%! %%BoundingBox: (atend) %%Pages: (atend) %%DocumentFonts: (atend) %%EndComments % % FrameMaker PostScript Prolog 2.0, for use with FrameMaker 2.0 % Copyright (c) 1986,87,89 by Frame Technology, Inc. All rights reserved. % % Known Problems: % Due to bugs in Transcript, the 'PS-Adobe-' is omitted from line 1 /FMversion (2.0) def % Set up Color vs. Black-and-White /FMPrintInColor systemdict /colorimage known def % Uncomment this line to force b&w on color printer % /FMPrintInColor false def /FrameDict 190 dict def systemdict /errordict known not {/errordict 10 dict def errordict /rangecheck {stop} put} if % The readline in 23.0 doesn't recognize cr's as nl's on AppleTalk FrameDict /tmprangecheck errordict /rangecheck get put errordict /rangecheck {FrameDict /bug true put} put FrameDict /bug false put mark % Some PS machines read past the CR, so keep the following 3 lines together! currentfile 5 string readline 00 0000000000 cleartomark errordict /rangecheck FrameDict /tmprangecheck get put FrameDict /bug get { /readline { /gstring exch def /gfile exch def /gindex 0 def { gfile read pop dup 10 eq {exit} if dup 13 eq {exit} if gstring exch gindex exch put /gindex gindex 1 add def } loop pop gstring 0 gindex getinterval true } def } if /FMVERSION { FMversion ne { /Times-Roman findfont 18 scalefont setfont 100 100 moveto (FrameMaker version does not match postscript_prolog!) dup = show showpage } if } def /FMLOCAL { FrameDict begin 0 def end } def /gstring FMLOCAL /gfile FMLOCAL /gindex FMLOCAL /orgxfer FMLOCAL /orgproc FMLOCAL /organgle FMLOCAL /orgfreq FMLOCAL /yscale FMLOCAL /xscale FMLOCAL /manualfeed FMLOCAL /paperheight FMLOCAL /paperwidth FMLOCAL /FMDOCUMENT { array /FMfonts exch def /#copies exch def FrameDict begin 0 ne dup {setmanualfeed} if /manualfeed exch def /paperheight exch def /paperwidth exch def setpapername manualfeed {true} {papersize} ifelse {manualpapersize} {false} ifelse {desperatepapersize} if /yscale exch def /xscale exch def currenttransfer cvlit /orgxfer exch def currentscreen cvlit /orgproc exch def /organgle exch def /orgfreq exch def end } def /pagesave FMLOCAL /orgmatrix FMLOCAL /landscape FMLOCAL /FMBEGINPAGE { FrameDict begin /pagesave save def 3.86 setmiterlimit /landscape exch 0 ne def landscape { 90 rotate 0 exch neg translate pop } {pop pop} ifelse xscale yscale scale /orgmatrix matrix def gsave } def /FMENDPAGE { grestore pagesave restore end showpage } def /FMDEFINEFONT { FrameDict begin findfont ReEncode 2 index exch definefont exch scalefont FMfonts 3 1 roll put end } bind def /FMNORMALIZEGRAPHICS { newpath 0.0 0.0 moveto 1 setlinewidth 0 setlinecap 0 0 0 sethsbcolor 0 setgray } bind def /fx FMLOCAL /fy FMLOCAL /fh FMLOCAL /fw FMLOCAL /llx FMLOCAL /lly FMLOCAL /urx FMLOCAL /ury FMLOCAL /FMBEGINEPSF { end /FMEPSF save def /showpage {} def FMNORMALIZEGRAPHICS [/fy /fx /fh /fw /ury /urx /lly /llx] {exch def} forall fx fy translate rotate fw urx llx sub div fh ury lly sub div scale llx neg lly neg translate } bind def /FMENDEPSF { FMEPSF restore FrameDict begin } bind def FrameDict begin /setmanualfeed { %%BeginFeature *ManualFeed True statusdict /manualfeed true put %%EndFeature } def /max {2 copy lt {exch} if pop} bind def /min {2 copy gt {exch} if pop} bind def /inch {72 mul} def /pagedimen { paperheight sub abs 16 lt exch paperwidth sub abs 16 lt and {/papername exch def} {pop} ifelse } def /papersizedict FMLOCAL /setpapername { /papersizedict 14 dict def papersizedict begin /papername /unknown def /Letter 8.5 inch 11.0 inch pagedimen /LetterSmall 7.68 inch 10.16 inch pagedimen /Tabloid 11.0 inch 17.0 inch pagedimen /Ledger 17.0 inch 11.0 inch pagedimen /Legal 8.5 inch 14.0 inch pagedimen /Statement 5.5 inch 8.5 inch pagedimen /Executive 7.5 inch 10.0 inch pagedimen /A3 11.69 inch 16.5 inch pagedimen /A4 8.26 inch 11.69 inch pagedimen /A4Small 7.47 inch 10.85 inch pagedimen /B4 10.125 inch 14.33 inch pagedimen /B5 7.16 inch 10.125 inch pagedimen end } def /papersize { papersizedict begin /Letter {lettertray} def /LetterSmall {lettertray lettersmall} def /Tabloid {11x17tray} def /Ledger {ledgertray} def /Legal {legaltray} def /Statement {statementtray} def /Executive {executivetray} def /A3 {a3tray} def /A4 {a4tray} def /A4Small {a4tray a4small} def /B4 {b4tray} def /B5 {b5tray} def /unknown {unknown} def papersizedict dup papername known {papername} {/unknown} ifelse get end /FMdicttop countdictstack 1 add def statusdict begin stopped end countdictstack -1 FMdicttop {pop end} for } def /manualpapersize { papersizedict begin /Letter {letter} def /LetterSmall {lettersmall} def /Tabloid {11x17} def /Ledger {ledger} def /Legal {legal} def /Statement {statement} def /Executive {executive} def /A3 {a3} def /A4 {a4} def /A4Small {a4small} def /B4 {b4} def /B5 {b5} def /unknown {unknown} def papersizedict dup papername known {papername} {/unknown} ifelse get end stopped } def /desperatepapersize { statusdict /setpageparams known { paperwidth paperheight 0 1 statusdict begin {setpageparams} stopped pop end } if } def /savematrix { orgmatrix currentmatrix pop } bind def /restorematrix { orgmatrix setmatrix } bind def /dmatrix matrix def /dpi 72 0 dmatrix defaultmatrix dtransform dup mul exch dup mul add sqrt def /freq dpi 18.75 div 8 div round dup 0 eq {pop 1} if 8 mul dpi exch div def /sangle 1 0 dmatrix defaultmatrix dtransform exch atan def /DiacriticEncoding [ /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /space /exclam /quotedbl /numbersign /dollar /percent /ampersand /quotesingle /parenleft /parenright /asterisk /plus /comma /hyphen /period /slash /zero /one /two /three /four /five /six /seven /eight /nine /colon /semicolon /less /equal /greater /question /at /A /B /C /D /E /F /G /H /I /J /K /L /M /N /O /P /Q /R /S /T /U /V /W /X /Y /Z /bracketleft /backslash /bracketright /asciicircum /underscore /grave /a /b /c /d /e /f /g /h /i /j /k /l /m /n /o /p /q /r /s /t /u /v /w /x /y /z /braceleft /bar /braceright /asciitilde /.notdef /Adieresis /Aring /Ccedilla /Eacute /Ntilde /Odieresis /Udieresis /aacute /agrave /acircumflex /adieresis /atilde /aring /ccedilla /eacute /egrave /ecircumflex /edieresis /iacute /igrave /icircumflex /idieresis /ntilde /oacute /ograve /ocircumflex /odieresis /otilde /uacute /ugrave /ucircumflex /udieresis /dagger /.notdef /cent /sterling /section /bullet /paragraph /germandbls /registered /copyright /trademark /acute /dieresis /.notdef /AE /Oslash /.notdef /.notdef /.notdef /.notdef /yen /.notdef /.notdef /.notdef /.notdef /.notdef /.notdef /ordfeminine /ordmasculine /.notdef /ae /oslash /questiondown /exclamdown /logicalnot /.notdef /florin /.notdef /.notdef /guillemotleft /guillemotright /ellipsis /.notdef /Agrave /Atilde /Otilde /OE /oe /endash /emdash /quotedblleft /quotedblright /quoteleft /quoteright /.notdef /.notdef /ydieresis /Ydieresis /fraction /currency /guilsinglleft /guilsinglright /fi /fl /daggerdbl /periodcentered /quotesinglbase /quotedblbase /perthousand /Acircumflex /Ecircumflex /Aacute /Edieresis /Egrave /Iacute /Icircumflex /Idieresis /Igrave /Oacute /Ocircumflex /.notdef /Ograve /Uacute /Ucircumflex /Ugrave /dotlessi /circumflex /tilde /macron /breve /dotaccent /ring /cedilla /hungarumlaut /ogonek /caron ] def /ReEncode { dup length dict begin { 1 index /FID ne {def} {pop pop} ifelse } forall Encoding StandardEncoding eq { /Encoding DiacriticEncoding def }if currentdict end } bind def /graymode true def /bwidth FMLOCAL /bpside FMLOCAL /bstring FMLOCAL /onbits FMLOCAL /offbits FMLOCAL /xindex FMLOCAL /yindex FMLOCAL /x FMLOCAL /y FMLOCAL /setpattern { /bwidth exch def /bpside exch def /bstring exch def /onbits 0 def /offbits 0 def freq sangle landscape {90 add} if {/y exch def /x exch def /xindex x 1 add 2 div bpside mul cvi def /yindex y 1 add 2 div bpside mul cvi def bstring yindex bwidth mul xindex 8 idiv add get 1 7 xindex 8 mod sub bitshift and 0 ne {/onbits onbits 1 add def 1} {/offbits offbits 1 add def 0} ifelse } setscreen {} settransfer offbits offbits onbits add div FMsetgray /graymode false def } bind def /grayness { FMsetgray graymode not { /graymode true def orgxfer cvx settransfer orgfreq organgle orgproc cvx setscreen } if } bind def /HUE FMLOCAL /SAT FMLOCAL /BRIGHT FMLOCAL /Colors FMLOCAL FMPrintInColor { /HUE 0 def /SAT 0 def /BRIGHT 0 def % array of arrays Hue and Sat values for the separations [HUE BRIGHT] /Colors [[0 0 ] % black [0 0 ] % white [0.00 1.0] % red [0.37 1.0] % green [0.60 1.0] % blue [0.50 1.0] % cyan [0.83 1.0] % magenta [0.16 1.0] % comment / yellow ] def /BEGINBITMAPCOLOR { BITMAPCOLOR} def /BEGINBITMAPCOLORc { BITMAPCOLORc} def /K { Colors exch get dup 0 get /HUE exch store 1 get /BRIGHT exch store HUE 0 eq BRIGHT 0 eq and {1.0 SAT sub setgray} {HUE SAT BRIGHT sethsbcolor} ifelse } def /FMsetgray { /SAT exch 1.0 exch sub store HUE 0 eq BRIGHT 0 eq and {1.0 SAT sub setgray} {HUE SAT BRIGHT sethsbcolor} ifelse } bind def } { /BEGINBITMAPCOLOR { BITMAPGRAY} def /BEGINBITMAPCOLORc { BITMAPGRAYc} def /FMsetgray {setgray} bind def /K { pop } def } ifelse /normalize { transform round exch round exch itransform } bind def /dnormalize { dtransform round exch round exch idtransform } bind def /lnormalize { 0 dtransform exch cvi 2 idiv 2 mul 1 add exch idtransform pop } bind def /H { lnormalize setlinewidth } bind def /Z { setlinecap } bind def /X { fillprocs exch get exec } bind def /V { gsave eofill grestore } bind def /N { stroke } bind def /M {newpath moveto} bind def /E {lineto} bind def /D {curveto} bind def /O {closepath} bind def /n FMLOCAL /L { /n exch def newpath normalize moveto 2 1 n {pop normalize lineto} for } bind def /Y { L closepath } bind def /x1 FMLOCAL /x2 FMLOCAL /y1 FMLOCAL /y2 FMLOCAL /rad FMLOCAL /R { /y2 exch def /x2 exch def /y1 exch def /x1 exch def x1 y1 x2 y1 x2 y2 x1 y2 4 Y } bind def /RR { /rad exch def normalize /y2 exch def /x2 exch def normalize /y1 exch def /x1 exch def newpath x1 y1 rad add moveto x1 y2 x2 y2 rad arcto x2 y2 x2 y1 rad arcto x2 y1 x1 y1 rad arcto x1 y1 x1 y2 rad arcto closepath 16 {pop} repeat } bind def /C { grestore gsave R clip } bind def /U { grestore gsave } bind def /F { FMfonts exch get setfont } bind def /T { moveto show } bind def /RF { rotate 0 ne {-1 1 scale} if } bind def /TF { gsave moveto RF show grestore } bind def /P { moveto 0 32 3 2 roll widthshow } bind def /PF { gsave moveto RF 0 32 3 2 roll widthshow grestore } bind def /S { moveto 0 exch ashow } bind def /SF { gsave moveto RF 0 exch ashow grestore } bind def /B { moveto 0 32 4 2 roll 0 exch awidthshow } bind def /BF { gsave moveto RF 0 32 4 2 roll 0 exch awidthshow grestore } bind def /x FMLOCAL /y FMLOCAL /dx FMLOCAL /dy FMLOCAL /dl FMLOCAL /t FMLOCAL /t2 FMLOCAL /Cos FMLOCAL /Sin FMLOCAL /r FMLOCAL /W { dnormalize /dy exch def /dx exch def normalize /y exch def /x exch def /dl dx dx mul dy dy mul add sqrt def dl 0.0 gt { /t currentlinewidth def savematrix /Cos dx dl div def /Sin dy dl div def /r [Cos Sin Sin neg Cos 0.0 0.0] def /t2 t 2.5 mul 3.5 max def newpath x y translate r concat 0.0 0.0 moveto dl t 2.7 mul sub 0.0 rlineto stroke restorematrix x dx add y dy add translate r concat t 0.67 mul setlinewidth t 1.61 mul neg 0.0 translate 0.0 0.0 moveto t2 1.7 mul neg t2 2.0 div moveto 0.0 0.0 lineto t2 1.7 mul neg t2 2.0 div neg lineto stroke t setlinewidth restorematrix } if } bind def /G { gsave newpath normalize translate 0.0 0.0 moveto dnormalize scale 0.0 0.0 1.0 5 3 roll arc closepath fill grestore } bind def /A { gsave savematrix newpath 2 index 2 div add exch 3 index 2 div sub exch normalize 2 index 2 div sub exch 3 index 2 div add exch translate scale 0.0 0.0 1.0 5 3 roll arc restorematrix stroke grestore } bind def /x FMLOCAL /y FMLOCAL /w FMLOCAL /h FMLOCAL /xx FMLOCAL /yy FMLOCAL /ww FMLOCAL /hh FMLOCAL /FMsaveobject FMLOCAL /FMoptop FMLOCAL /FMdicttop FMLOCAL /BEGINPRINTCODE { /FMdicttop countdictstack 1 add def /FMoptop count 4 sub def /FMsaveobject save def userdict begin /showpage {} def FMNORMALIZEGRAPHICS 3 index neg 3 index neg translate } bind def /ENDPRINTCODE { count -1 FMoptop {pop pop} for countdictstack -1 FMdicttop {pop end} for FMsaveobject restore } bind def /gn { 0 { 46 mul cf read pop 32 sub dup 46 lt {exit} if 46 sub add } loop add } bind def /str FMLOCAL /cfs { /str sl string def 0 1 sl 1 sub {str exch val put} for str def } bind def /ic [ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0223 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0223 0 {0 hx} {1 hx} {2 hx} {3 hx} {4 hx} {5 hx} {6 hx} {7 hx} {8 hx} {9 hx} {10 hx} {11 hx} {12 hx} {13 hx} {14 hx} {15 hx} {16 hx} {17 hx} {18 hx} {19 hx} {gn hx} {0} {1} {2} {3} {4} {5} {6} {7} {8} {9} {10} {11} {12} {13} {14} {15} {16} {17} {18} {19} {gn} {0 wh} {1 wh} {2 wh} {3 wh} {4 wh} {5 wh} {6 wh} {7 wh} {8 wh} {9 wh} {10 wh} {11 wh} {12 wh} {13 wh} {14 wh} {gn wh} {0 bl} {1 bl} {2 bl} {3 bl} {4 bl} {5 bl} {6 bl} {7 bl} {8 bl} {9 bl} {10 bl} {11 bl} {12 bl} {13 bl} {14 bl} {gn bl} {0 fl} {1 fl} {2 fl} {3 fl} {4 fl} {5 fl} {6 fl} {7 fl} {8 fl} {9 fl} {10 fl} {11 fl} {12 fl} {13 fl} {14 fl} {gn fl} ] def /sl FMLOCAL /val FMLOCAL /ws FMLOCAL /im FMLOCAL /bs FMLOCAL /cs FMLOCAL /len FMLOCAL /pos FMLOCAL /ms { /sl exch def /val 255 def /ws cfs /im cfs /val 0 def /bs cfs /cs cfs } bind def 400 ms /ip { is 0 cf cs readline pop { ic exch get exec add } forall pop } bind def /wh { /len exch def /pos exch def ws 0 len getinterval im pos len getinterval copy pop pos len } bind def /bl { /len exch def /pos exch def bs 0 len getinterval im pos len getinterval copy pop pos len } bind def /s1 1 string def /fl { /len exch def /pos exch def /val cf s1 readhexstring pop 0 get def pos 1 pos len add 1 sub {im exch val put} for pos len } bind def /hx { 3 copy getinterval cf exch readhexstring pop pop } bind def /h FMLOCAL /w FMLOCAL /d FMLOCAL /lb FMLOCAL /bitmapsave FMLOCAL /is FMLOCAL /cf FMLOCAL /wbytes { dup 8 eq {pop} {1 eq {7 add 8 idiv} {3 add 4 idiv} ifelse} ifelse } bind def /BEGINBITMAPBWc { 1 {} COMMONBITMAPc } bind def /BEGINBITMAPGRAYc { 8 {} COMMONBITMAPc } bind def /BEGINBITMAP2BITc { 2 {} COMMONBITMAPc } bind def /COMMONBITMAPc { /r exch def /d exch def gsave translate rotate scale /h exch def /w exch def /lb w d wbytes def sl lb lt {lb ms} if /bitmapsave save def r /is im 0 lb getinterval def ws 0 lb getinterval is copy pop /cf currentfile def w h d [w 0 0 h neg 0 h] {ip} image bitmapsave restore grestore } bind def /BEGINBITMAPBW { 1 {} COMMONBITMAP } bind def /BEGINBITMAPGRAY { 8 {} COMMONBITMAP } bind def /BEGINBITMAP2BIT { 2 {} COMMONBITMAP } bind def /COMMONBITMAP { /r exch def /d exch def gsave translate rotate scale /h exch def /w exch def /bitmapsave save def r /is w d wbytes string def /cf currentfile def w h d [w 0 0 h neg 0 h] {cf is readhexstring pop} image bitmapsave restore grestore } bind def /proc1 FMLOCAL /proc2 FMLOCAL /newproc FMLOCAL /Fmcc { /proc2 exch cvlit def /proc1 exch cvlit def /newproc proc1 length proc2 length add array def newproc 0 proc1 putinterval newproc proc1 length proc2 putinterval newproc cvx } bind def /ngrayt 256 array def /nredt 256 array def /nbluet 256 array def /ngreent 256 array def /gryt FMLOCAL /blut FMLOCAL /grnt FMLOCAL /redt FMLOCAL /indx FMLOCAL /cynu FMLOCAL /magu FMLOCAL /yelu FMLOCAL /k FMLOCAL /u FMLOCAL /colorsetup { currentcolortransfer /gryt exch def /blut exch def /grnt exch def /redt exch def 0 1 255 { /indx exch def /cynu 1 red indx get 255 div sub def /magu 1 green indx get 255 div sub def /yelu 1 blue indx get 255 div sub def /k cynu magu min yelu min def /u k currentundercolorremoval exec def nredt indx 1 0 cynu u sub max sub redt exec put ngreent indx 1 0 magu u sub max sub grnt exec put nbluet indx 1 0 yelu u sub max sub blut exec put ngrayt indx 1 k currentblackgeneration exec sub gryt exec put } for {255 mul cvi nredt exch get} {255 mul cvi ngreent exch get} {255 mul cvi nbluet exch get} {255 mul cvi ngrayt exch get} setcolortransfer {pop 0} setundercolorremoval {} setblackgeneration } bind def /tran FMLOCAL /fakecolorsetup { /tran 256 string def 0 1 255 {/indx exch def tran indx red indx get 77 mul green indx get 151 mul blue indx get 28 mul add add 256 idiv put} for currenttransfer {255 mul cvi tran exch get 255.0 div} exch Fmcc settransfer } bind def /BITMAPCOLOR { /d 8 def gsave translate rotate scale /h exch def /w exch def /bitmapsave save def colorsetup /is w d wbytes string def /cf currentfile def w h d [w 0 0 h neg 0 h] {cf is readhexstring pop} {is} {is} true 3 colorimage bitmapsave restore grestore } bind def /BITMAPCOLORc { /d 8 def gsave translate rotate scale /h exch def /w exch def /lb w d wbytes def sl lb lt {lb ms} if /bitmapsave save def colorsetup /is im 0 lb getinterval def ws 0 lb getinterval is copy pop /cf currentfile def w h d [w 0 0 h neg 0 h] {ip} {is} {is} true 3 colorimage bitmapsave restore grestore } bind def /BITMAPGRAY { 8 {fakecolorsetup} COMMONBITMAP } bind def /BITMAPGRAYc { 8 {fakecolorsetup} COMMONBITMAPc } bind def /ENDBITMAP { } bind def end %%EndProlog %%BeginSetup (2.0) FMVERSION 1 1 612 792 0 1 8 FMDOCUMENT /fillprocs 32 array def fillprocs 0 { 0.000000 grayness } put fillprocs 1 { 0.100000 grayness } put fillprocs 2 { 0.300000 grayness } put fillprocs 3 { 0.500000 grayness } put fillprocs 4 { 0.700000 grayness } put fillprocs 5 { 0.900000 grayness } put fillprocs 6 { 0.970000 grayness } put fillprocs 7 { 1.000000 grayness } put fillprocs 8 {<0f87c3e1f0783c1e> 8 1 setpattern } put fillprocs 9 {<0f1e3c78f0e1c387> 8 1 setpattern } put fillprocs 10 { 8 1 setpattern } put fillprocs 11 { 8 1 setpattern } put fillprocs 12 {<8142241818244281> 8 1 setpattern } put fillprocs 13 {<8040201008040201> 8 1 setpattern } put fillprocs 14 {<03060c183060c081> 8 1 setpattern } put fillprocs 15 {} put fillprocs 16 { 1.000000 grayness } put fillprocs 17 { 0.900000 grayness } put fillprocs 18 { 0.700000 grayness } put fillprocs 19 { 0.500000 grayness } put fillprocs 20 { 0.300000 grayness } put fillprocs 21 { 0.100000 grayness } put fillprocs 22 { 0.030000 grayness } put fillprocs 23 { 0.000000 grayness } put fillprocs 24 { 8 1 setpattern } put fillprocs 25 { 8 1 setpattern } put fillprocs 26 {<3333333333333333> 8 1 setpattern } put fillprocs 27 {<0000ffff0000ffff> 8 1 setpattern } put fillprocs 28 {<7ebddbe7e7dbbd7e> 8 1 setpattern } put fillprocs 29 {<7fbfdfeff7fbfdfe> 8 1 setpattern } put fillprocs 30 { 8 1 setpattern } put fillprocs 31 {} put %%EndSetup 0 12 /Times-Roman FMDEFINEFONT 1 24 /Times-Roman FMDEFINEFONT 2 16 /Times-Bold FMDEFINEFONT %%Page: "1" 1 %%BeginPaperSize: Letter %%EndPaperSize 612 792 0 FMBEGINPAGE 72 675 540 720 R 7 X 0 K V 0 F 0 X (Network Working Group) 72 712 T (J. Moy, Editor) 470.7 712 T (Request for Comments: 1246) 72 698 T (Proteon, Inc.) 478.38 698 T (July 1991) 493.02 684 T 72 72 540 83.95 R 7 X V 0 X ([Moy]) 72 75.95 T ([Page 1]) 499.7 75.95 T 72 117 540 603 R 7 X V 1 F 0 X (Experience with the OSPF protocol) 135.44 587 T 2 F (Status of this Memo) 72 514.33 T 0 F -0.23 (This memo provides information for the Internet community) 72 487 P -0.23 (. It does not specify any Internet stan-) 360.42 487 P (dard. Distribution of this memo is unlimited ) 72 473 T 2 F (Abstract) 72 442.33 T 0 F (This is the second of two reports on the OSPF protocol. These reports are required by the IAB/) 72 415 T (IESG in order for an Internet routing protocol to advance to Draft Standard Status. OSPF is a ) 72 401 T -0.05 (TCP/IP routing protocol, designed to be used internal to an Autonomous System \050in other words, ) 72 387 P (OSPF is an Interior Gateway Protocol\051.) 72 373 T -0.09 (V) 72 347 P -0.09 (ersion 1 of the OSPF protocol was published in RFC 1) 79.33 347 P -0.09 (131. Since then OSPF version 2 has been ) 339.85 347 P -0.22 (developed. V) 72 333 P -0.22 (ersion 2 has been documented in RFC 1247. The changes between version 1 and ver-) 134.4 333 P -0 (sion 2 of the OSPF protocol are explained in Appendix F of RFC 1247. It is OSPF V) 72 319 P -0 (ersion 2 that ) 477.72 319 P (is the subject of this report.) 72 305 T (This report documents experience with OSPF V2. This includes reports on interoperability test-) 72 279 T (ing, \336eld experience, simulations and the current state of OSPF implementations. It also presents ) 72 265 T -0.19 (a summary of the OSPF Management Information Base \050MIB\051, and a summary of OSPF authenti-) 72 251 P (cation mechanism.) 72 237 T (Please send comments to ospf@trantor) 72 211 T (.umd.edu.) 258.27 211 T FMENDPAGE %%EndPage: "1" 2 1 10 /Times-Roman FMDEFINEFONT %%Page: "2" 2 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 2]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (T) 72 673.33 T (able of Contents) 81.19 673.33 T 0 F (1.0) 72 650 T (Introduction) 108 650 T (..............................................................................................................) 167.91 650 T (3) 498 650 T 1 F (1.1) 108 635.33 T (Acknowledgments) 144 635.33 T (...............................................................................................................) 219.88 635.33 T (3) 499 635.33 T 0 F (2.0) 72 616 T (Documentation) 108 616 T (.........................................................................................................) 182.9 616 T (4) 498 616 T (3.0) 72 596 T (MIB) 108 596 T (..........................................................................................................................) 131.93 596 T (5) 498 596 T (4.0) 72 576 T (Security architecture) 108 576 T (................................................................................................) 209.88 576 T (7) 498 576 T (5.0) 72 556 T (Implementations) 108 556 T (.......................................................................................................) 188.9 556 T (8) 498 556 T (6.0) 72 536 T (Operational experience) 108 536 T (............................................................................................) 221.88 536 T (9) 498 536 T 1 F (6.1) 108 521.33 T (NSI) 144 521.33 T (......................................................................................................................................) 162.41 521.33 T (9) 499 521.33 T (6.1.1) 144 509.33 T (NSI\325s OSPF system) 180 509.33 T (............................................................................................) 262.36 509.33 T (10) 494.01 509.33 T (6.1.2) 144 497.33 T (NSI - Deployment analysis) 180 497.33 T (.................................................................................) 289.84 497.33 T (11) 494.01 497.33 T (6.2) 108 483.33 T (BARRNet) 144 483.33 T (..........................................................................................................................) 187.4 483.33 T (1) 494.38 483.33 T (1) 499 483.33 T (6.2.1) 144 471.33 T (BARRNet\325s OSPF system) 180 471.33 T (..................................................................................) 287.34 471.33 T (13) 494.01 471.33 T (6.2.2) 144 459.33 T (BARRNet - Deployment analysis) 180 459.33 T (.......................................................................) 314.83 459.33 T (14) 494.01 459.33 T (6.3) 108 445.33 T (OARnet) 144 445.33 T (..............................................................................................................................) 177.4 445.33 T (14) 494.01 445.33 T (6.3.1) 144 433.33 T (OARnet\325s OSPF system) 180 433.33 T (......................................................................................) 277.35 433.33 T (15) 494.01 433.33 T (6.3.2) 144 421.33 T (OARnet - Deployment analysis) 180 421.33 T (..........................................................................) 307.33 421.33 T (17) 494.01 421.33 T (6.4) 108 407.33 T (Features exercised during operational deployment) 144 407.33 T (...........................................................) 344.81 407.33 T (20) 494.01 407.33 T (6.5) 108 393.33 T (Limitations of operational deployments) 144 393.33 T (............................................................................) 302.33 393.33 T (21) 494.01 393.33 T (6.6) 108 379.33 T (Conclusions) 144 379.33 T (.......................................................................................................................) 194.89 379.33 T (21) 494.01 379.33 T 0 F (7.0) 72 360 T ( Interoperability T) 108 360 T (esting) 195.1 360 T (.........................................................................................) 224.88 360 T (23) 492.01 360 T 1 F (7.1) 108 345.33 T (T) 144 345.33 T (esting methodology) 149.41 345.33 T (.........................................................................................................) 229.87 345.33 T (23) 494.01 345.33 T (7.2) 108 331.33 T (First round \050Proteon, 9/25/90 - 9/29/90\051) 144 331.33 T (...........................................................................) 304.83 331.33 T (24) 494.01 331.33 T (7.2.1) 144 319.33 T (Problems found in the First round testing) 180 319.33 T (...........................................................) 344.81 319.33 T (25) 494.01 319.33 T (7.3) 108 305.33 T (Second round \050SURAnet, 12/17/90 - 12/21/90\051) 144 305.33 T (................................................................) 332.32 305.33 T (25) 494.01 305.33 T (7.3.1) 144 293.33 T (Official report of the Second round testing) 180 293.33 T (........................................................) 352.31 293.33 T (26) 494.01 293.33 T (7.3.2) 144 281.33 T (Problems found in the Second round testing) 180 281.33 T (......................................................) 357.3 281.33 T (28) 494.01 281.33 T (7.4) 108 267.33 T (Third round \0503com, 2/4/91 - 2/8/91\051) 144 267.33 T (.................................................................................) 289.84 267.33 T (28) 494.01 267.33 T (7.4.1) 144 255.33 T (Official report of the Third round testing) 180 255.33 T (...........................................................) 344.81 255.33 T (29) 494.01 255.33 T (7.4.2) 144 243.33 T (Problems found in the Third round testing) 180 243.33 T (.........................................................) 349.81 243.33 T (31) 494.01 243.33 T (7.5) 108 229.33 T (Overall: Features tested) 144 229.33 T (.....................................................................................................) 239.87 229.33 T (31) 494.01 229.33 T (7.6) 108 215.33 T (T) 144 215.33 T (esting conclusions) 149.41 215.33 T (...........................................................................................................) 224.88 215.33 T (33) 494.01 215.33 T (7.7) 108 201.33 T (Future work) 144 201.33 T (.......................................................................................................................) 194.89 201.33 T (33) 494.01 201.33 T 0 F (8.0) 72 182 T (Simulation) 108 182 T (..............................................................................................................) 161.91 182 T (34) 492.01 182 T 1 F (8.1) 108 167.33 T (Simulator setup) 144 167.33 T (..................................................................................................................) 207.39 167.33 T (34) 494.01 167.33 T (8.2) 108 153.33 T (Simulation results) 144 153.33 T (..............................................................................................................) 217.38 153.33 T (34) 494.01 153.33 T 0 F (9.0) 72 134 T (Reference Documents) 108 134 T (............................................................................................) 215.88 134 T (36) 492.01 134 T (10.0) 72 114 T (People) 108 114 T (.....................................................................................................................) 140.92 114 T (37) 492.01 114 T FMENDPAGE %%EndPage: "2" 3 3 14 /Times-Bold FMDEFINEFONT %%Page: "3" 3 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 3]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (1.0 Intr) 72 673.33 T (oduction) 127.23 673.33 T 0 F -0.02 (This document addresses, for OSPF V2, the requirements set forth by the IAB/IESG for an Inter-) 72 646 P -0.19 (net routing protocol to advance to Draft Standard state. This requirements are brie\337y summarized ) 72 632 P (below) 72 618 T (. The remaining sections of this report document how OSPF V2 satis\336es these require-) 100.53 618 T (ments:) 72 604 T (\245) 72 584 T (The speci\336cation for the routing protocol must be well written such that independent, interop-) 85.54 584 T (erable implementations can be developed solely based on the speci\336cation. For example, it ) 85.54 570 T (should be possible to develop an interoperable implementation without consulting the original ) 85.54 556 T (developers of the routing protocol.) 85.54 542 T (\245) 72 522 T -0.1 (A Management Information Base \050MIB\051 must be written for the protocol. The MIB must be in ) 85.54 522 P (the standardization process, but does not need to be at the same level of standardization as the ) 85.54 508 T (routing protocol.) 85.54 494 T (\245) 72 474 T (The security architecture of the protocol must be set forth explicitly) 85.54 474 T (. The security architecture ) 409.19 474 T -0.02 (must include mechanisms for authenticating routing messages and may include other forms of ) 85.54 460 P (protection.) 85.54 446 T (\245) 72 426 T (T) 85.54 426 T (wo or more interoperable implementations must exist. At least two must be written indepen-) 92.02 426 T (dently) 85.54 412 T (.) 114.74 412 T (\245) 72 392 T (There must be evidence that all features of the protocol have been tested, running between at ) 85.54 392 T -0.21 (least two implementations. This must include that all of the security features have been demon-) 85.54 378 P (strated to operate, and that the mechanisms de\336ned in the protocol actually provide the ) 85.54 364 T (intended protection.) 85.54 350 T (\245) 72 330 T (There must be signi\336cant operational experience. This must include running in a moderate ) 85.54 330 T (number routers con\336gured in a moderately complex topology) 85.54 316 T (, and must be part of the opera-) 379.55 316 T -0.07 (tional Internet. All signi\336cant features of the protocol must be exercised. In the case of an Inte-) 85.54 302 P (rior Gateway Protocol \050IGP\051, both interior and exterior routes must be carried \050unless another ) 85.54 288 T (mechanism is provided for the exterior routes\051. In the case of a Exterior Gateway Protocol ) 85.54 274 T (\050EGP\051, it must carry the full complement of exterior routes. ) 85.54 260 T (This report is a compilation of information obtained from many people. The reader is referred to ) 72 234 T (speci\336c people when more information on a subject is available. People references are gathered ) 72 220 T (into Section 10.0, in a format similar to that used in [4].) 72 206 T 3 F (1.1 Acknowledgments) 72 172.67 T 0 F -0.03 (The OSPF protocol has been developed by the OSPF W) 72 146 P -0.03 (orking Group of the Internet Engineering ) 339.64 146 P (T) 72 132 T (ask Force. Many people have contributed to this report. They are listed in Section 10.0 of this ) 78.49 132 T (report.) 72 118 T FMENDPAGE %%EndPage: "3" 4 4 12 /Times-Italic FMDEFINEFONT %%Page: "4" 4 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 4]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (2.0 Documentation) 72 673.33 T 0 F -0.01 (V) 72 646 P -0.01 (ersion 1 of the OSPF protocol is documented in RFC 1) 79.33 646 P -0.01 (131 [1]. OSPF V) 341.93 646 P -0.01 (ersion 2, which super-) 421.84 646 P -0.3 (sedes V) 72 632 P -0.3 (ersion 1, has been documented in RFC 1247 [2]. The dif) 108 632 P -0.3 (ferences between OSPF V) 375.29 632 P -0.3 (ersion 1 ) 499.29 632 P -0.22 (and V) 72 618 P -0.22 (ersion 2 are relatively minor) 99.43 618 P -0.22 (, and are listed in Appendix F of RFC 1247 [2]. All information ) 233.98 618 P (presented in this report concerns OSPF V2 unless explicitly mentioned otherwise.) 72 604 T -0.19 (The OSPF protocol was developed by the OSPF W) 72 578 P -0.19 (orking Group of the Internet Engineering T) 315.35 578 P -0.19 (ask ) 521.21 578 P -0.12 (Force. This W) 72 564 P -0.12 (orking Group has a mailing list, ) 139.75 564 P 4 F -0.12 (ospf@trantor) 294.6 564 P -0.12 (.umd.edu) 358.28 564 P 0 F -0.12 (, where discussions of proto-) 402.25 564 P (col features and operation are held. The OSPF W) 72 550 T (orking Group also meets during the quarterly ) 307.19 550 T (Internet Engineering T) 72 536 T (ask Force conferences. Reports of these meeting are published in the ) 180.41 536 T (IETF\325) 72 522 T (s Proceedings. In addition, two reports on the OSPF protocol have been presented to the ) 100.65 522 T (IETF plenary \050see \322Everything Y) 72 508 T (ou Ever W) 230.67 508 T (anted to Know about OSPFIGP\323 in [5] and \322OSPF ) 281.67 508 T (Update\323 in [6]\051.) 72 494 T (The OSPF protocol began under) 72 468 T (going \336eld trials in Spring of 1990. A mailing list,) 227.01 468 T 4 F ( ospf-) 467.9 468 T -0.47 (tests@seka.cso.uiuc.edu) 72 454 P 0 F -0.47 (, was formed to discuss how the \336eld trials were proceeding. This mailing ) 188.62 454 P (list is maintained by Ross V) 72 440 T (each of the University of Illinois [rrv]. Archives of this list are also ) 205.61 440 T (available. There has been quite a bit of discussion on the list concerning OSPF/RIP/EGP interac-) 72 426 T (tion.) 72 412 T (A OSPF V2 Management Information Base has also been developed and published in [3]. For ) 72 386 T (more information, see Section 3.0 of this report.) 72 372 T (There is a free implementation of OSPF available from the University of Maryland. This imple-) 72 346 T (mentation was written by Rob Coltun [rcoltun]. Contact Rob for details.) 72 332 T FMENDPAGE %%EndPage: "4" 5 5 10 /Times-Bold FMDEFINEFONT %%Page: "5" 5 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 5]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (3.0 MIB) 72 673.33 T 0 F (An OSPF Management Information Base has been published in RFC 1248 [3]. The MIB was ) 72 646 T (written by Rob Coltun [rcoltun] and Fred Baker [fbaker]. The OSPF MIB appears on the mgmt ) 72 632 T (subtree as SMI standard-mib 13.) 72 618 T (The OSPF MIB was originally developed by Rob Coltun of the University of Maryland, under ) 72 592 T (contract to Advanced Computer Communications. A subset of his proposal was implemented to ) 72 578 T (facilitate their development, and represents operational experience of a sort.) 72 564 T (The MIB consists of a general variables group and ten tables:) 72 538 T 72 502.01 540 510 C 72 507.98 540 507.98 2 L 0.5 H 0 Z 0 X 0 K N 0 0 612 792 C 5 F 0 X 0 K (T) 72 513.33 T (ABLE 1. OSPF MIB Organization) 77.93 513.33 T (Gr) 144 491.34 T (oup Name) 156.03 491.34 T (Description) 288 491.34 T 1 F (ospfGeneralGroup) 144 474.34 T (General Global V) 288 474.34 T (ariables) 357.94 474.34 T ( ) 72 458.34 T (ospfAreaT) 144 458.34 T (able) 186.04 458.34 T (Area Descriptions) 288 458.34 T ( ) 72 442.34 T (ospfStubAreaT) 144 442.34 T (able) 204.37 442.34 T (Default Metrics, by T) 288 442.34 T (ype of Service ) 373.9 442.34 T (ospfLsdbT) 144 426.34 T (able) 186.61 426.34 T (Link State Database) 288 426.34 T (ospfAreaRangeT) 144 410.34 T (able) 211.57 410.34 T (Address Range Speci\336cations) 288 410.34 T (ospfHostT) 144 394.34 T (able) 185.5 394.34 T (Directly connected Hosts) 288 394.34 T ( ) 72 378.34 T (ospfIfT) 144 378.34 T (able) 173.27 378.34 T (OSPF Interface V) 288 378.34 T (ariables) 358.5 378.34 T ( ) 72 362.34 T (ospfIfMetricT) 144 362.34 T (able) 199.92 362.34 T (Interface Metrics, by T) 288 362.34 T (ype of Service) 379.44 362.34 T ( ) 72 346.34 T (ospfV) 144 346.34 T (irtIfT) 167.83 346.34 T (able) 188.77 346.34 T (V) 288 346.34 T (irtual Links) 294.62 346.34 T (ospfNbrT) 144 330.34 T (able) 182.16 330.34 T (\050Non-virtual\051 OSPF Neighbors) 288 330.34 T ( ) 72 314.34 T (ospfV) 144 314.34 T (irtNbrT) 167.83 314.34 T (able) 197.66 314.34 T (V) 288 314.34 T (irtual OSPF Neighbors) 294.62 314.34 T 0 F -0.05 (As MIBs go, the OSPF MIB is quite lar) 72 289.01 P -0.05 (ge; 105 objects. The following are some statistics describ-) 261.61 289.01 P (ing the distribution of the MIB\325) 72 275.01 T (s variables:) 222.93 275.01 T (\245) 72 255.01 T (1) 85.54 255.01 T (1 de\336ne the above Group and T) 91.09 255.01 T (ables) 242.13 255.01 T (\245) 72 235.01 T (10 de\336ne the Entry in a T) 85.54 235.01 T (able) 207.28 235.01 T (\245) 72 215.01 T (7 are Counters) 85.54 215.01 T (\245) 72 195.01 T (6 are Gauges ) 85.54 195.01 T (\245) 72 175.01 T (68 objects mandated by the OSPF V) 85.54 175.01 T (ersion 2 Speci\336cation) 258.77 175.01 T (Section D.2 of the OSPF V2 speci\336cation [2] lists a set of required statistics that an implementa-) 72 149.01 T -0.02 (tion must maintain. These statistics have been incorporated into the OSPF MIB. The MIB\325) 72 135.01 P -0.02 (s thir-) 505.85 135.01 P (teen Counters and Gauges enable evaluation of the OSPF protocol\325) 72 121.01 T (s performance in an ) 393.45 121.01 T FMENDPAGE %%EndPage: "5" 6 %%Page: "6" 6 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 6]) 499.7 73 T 72 108 540 684 R 7 X V 0 X (operational environment. Most of the remainder of the MIB\325) 72 676 T (s variables parameterize the many ) 362.13 676 T (features that OSPF provides the network administrator) 72 662 T (. ) 333.17 662 T (For more information on the MIB contact Fred Baker [fbaker].) 72 636 T FMENDPAGE %%EndPage: "6" 7 %%Page: "7" 7 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 7]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (4.0 Security ar) 72 673.33 T (chitectur) 175.64 673.33 T (e) 236.63 673.33 T 0 F (In OSPF) 72 646 T (, all protocol packet exchanges are authenticated. The OSPF packet header \050which is ) 112.69 646 T -0.19 (common to all OSPF packets\051 contains a 16-bit Authentication type \336eld, and 64-bits of Authenti-) 72 632 P (cation data. Each particular OSPF area must run a single authentication scheme, as indicated by ) 72 618 T (the Authentication type \336eld. However) 72 604 T (, authentication keys can be con\336gured by the system ) 258.39 604 T (administrator on a per) 72 590 T (-network basis.) 177.35 590 T (When an OSPF packet is received from a network, the OSPF router \336rst veri\336es that it indicates ) 72 564 T (the correct Authentication type. The router then authenticates the packet, running a veri\336cation ) 72 550 T (algorithm using the con\336gured authentication key) 72 536 T (, the 64-bits of Authentication data and the rest ) 310.07 536 T (of the OSPF packet data as input. The precise algorithm used is dictated by the Authentication ) 72 522 T (type. Packets failing the authentication algorithm are dropped, and the authentication failure is ) 72 508 T (noted in a MIB-accessible variable \050see [3]\051.) 72 494 T (There are currently few Authentication types in use. The current assignments are:) 72 468 T 72 432.01 540 440 C 72 437.98 540 437.98 2 L 0.5 H 0 Z 0 X 0 K N 0 0 612 792 C 5 F 0 X 0 K (T) 72 443.33 T (ABLE 2. Curr) 77.93 443.33 T (ent OSPF Authentication types.) 139.1 443.33 T (T) 144 421.34 T (ype code) 149.93 421.34 T (Algorithm) 225 421.34 T 1 F (0) 144 404.34 T (No authentication performed.) 225 404.34 T (1) 144 388.34 T (Simple \050clear\051 password.) 225 388.34 T (2-255) 144 372.34 T (Reserved for assignment by the IANA \050iana@isi.edu\051) 225 372.34 T (> 255) 144 356.34 T (A) 225 356.34 T (vailable for local \050per) 231.48 356.34 T (-AS\051 de\336nition.) 317.59 356.34 T 0 F (For more information on OSPF\325) 72 331.01 T (s authentication procedures, see Sections 8.1, 8.2, and Appendix ) 225.92 331.01 T (E of [2].) 72 317.01 T FMENDPAGE %%EndPage: "7" 8 6 12 /Times-Bold FMDEFINEFONT %%Page: "8" 8 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 8]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (5.0 Implementations) 72 673.33 T 0 F -0.36 (The are multiple, interoperable implementations of OSPF currently available. This section gives a ) 72 646 P (brief overview of the \336ve implementations that have participated in at least one round of interop-) 72 632 T (erability testing) 72 618 T 1 F (1) 146.96 622.8 T 0 F (. Other implementations do exist, but because of commercial realities \050e.g., the ) 151.96 618 T (product is not yet announced\051 they unfortunately cannot be listed here.) 72 604 T -0.03 (The \336ve implementations that have participated in the OSPF interoperability testing are \050listed in ) 72 578 P (alphabetical order\051:) 72 564 T (\245) 72 544 T 6 F (3com) 85.54 544 T 0 F (. This implementation was wholly developed by 3com. It has participated in all three ) 112.85 544 T (rounds of interoperability testing. It is also the only implementation of OSPF\325) 85.54 530 T (s T) 458 530 T (OS routing.. ) 472.78 530 T (The 3com implementation consists of approximately 9000 lines of C code, including com-) 85.54 516 T (ments but excluding user interface and MIB code. Consult Dino Farinacci [dino] for more ) 85.54 502 T (details.) 85.54 488 T (\245) 72 468 T 6 F (ACC.) 85.54 468 T 0 F ( This implementation is based on the University of Maryland code. It participated in the ) 114.51 468 T (last two rounds of interoperability testing. It also contains the only implementation of \050a pre-) 85.54 454 T -0.04 (cursor to\051 the OSPF MIB \050see Section 3.0 for details\051, which it uses for monitoring and con\336g-) 85.54 440 P -0.15 (uration. The ACC implementation consists of approximately 24,000 lines of C code, including ) 85.54 426 P (its OSPF MIB code. Consult Fred Baker [fbaker] for more details.) 85.54 412 T (\245) 72 392 T 6 F -0.45 (Pr) 85.54 392 P -0.45 (oteon) 97.97 392 P 0 F -0.45 (. This implementation was wholly developed by Proteon. It has participated in all three ) 125.95 392 P (rounds of interoperability testing. It is also the only implementation that has a signi\336cant ) 85.54 378 T (amount of \336eld experience \050see Section 6.0 for details\051. The Proteon implementation consists ) 85.54 364 T -0.14 (of approximately 9500 lines of C code, including comments but excluding user interface code. ) 85.54 350 P (Consult John Moy [jmoy] for more details.) 85.54 336 T (\245) 72 316 T 6 F (W) 85.54 316 T (ell\337eet) 96.87 316 T 0 F (. This implementation has participated in all three rounds of interoperability testing. ) 130.18 316 T (Consult Jonathan Hsu [jhsu] for more details.) 85.54 302 T (\245) 72 282 T 6 F (University of Maryland. ) 85.54 282 T 0 F (This implementation was developed wholly by Rob Coltun at the ) 212.13 282 T (University of Maryland. It has formed the basis for a number of commercial OSPF implemen-) 85.54 268 T (tations, and also participated in the latest round of interoperability testing. The University of ) 85.54 254 T (Maryland implementation consists of approximately 10,000 lines of C code. Consult Rob Col-) 85.54 240 T (tun [rcoltun] for more details.) 85.54 226 T -0.18 (Note that, as required by the IAB/IESG for Draft Standard status, there are multiple interoperable ) 72 200 P (independent implementations, namely those from 3com, Proteon and the University of Maryland.) 72 186 T 72 122 540 142.09 C 72 129.99 203.98 129.99 2 L 0.25 H 2 Z 0 X 0 K N 0 0 612 792 C 1 F 0 X 0 K (1. For a detailed discussion of OSPF interoperability testing, see Section 7.0 of this report.) 72 115.33 T FMENDPAGE %%EndPage: "8" 9 %%Page: "9" 9 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 9]) 499.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (6.0 Operational experience) 72 673.33 T 0 F -0.24 (This section discusses operational experience with the OSPF protocol. V) 72 646 P -0.24 (ersion 1 of the OSPF pro-) 417.61 646 P (tocol began to be deployed in the Internet in Spring of 1990. The results of this original deploy-) 72 632 T (ment were reported to the mailing list ) 72 618 T 4 F (ospf-tests@seka.cso.uiuc.edu) 255.55 618 T 1 F (2) 396.16 622.8 T 0 F (. No protocol bugs were ) 401.15 618 T (found in this \336rst deployment, although several additional features were found to be desirable. ) 72 604 T (These new features were added to the protocol in OSPF V) 72 590 T (ersion 2.) 349.79 590 T -0.3 (The OSPF protocol is now deployed in a number of places in the Internet. In this section we focus ) 72 564 P (on three highly visible systems, namely the NASA Sciences Internet, BARRNet and OARnet. ) 72 550 T (The dimensions of these three OSPF systems is summarized in the following table:) 72 536 T 72 500.01 540 508 C 72 505.98 540 505.98 2 L 0.5 H 0 Z 0 X 0 K N 0 0 612 792 C 5 F 0 X 0 K (T) 72 511.33 T (ABLE 3. Thr) 77.93 511.33 T (ee operational OSPF deployments) 134.12 511.33 T (Name) 72 489.34 T (V) 144 489.34 T (ersion 1 date) 150.3 489.34 T (V) 252 489.34 T (ersion 2 date) 258.3 489.34 T (# r) 360 489.34 T (outers) 371.75 489.34 T (#externals) 432 489.34 T 1 F (NSI) 72 472.34 T (4/13/90) 144 472.34 T (1/1/91) 252 472.34 T (15) 360 472.34 T (496) 432 472.34 T (BARRNet) 72 456.34 T (4/90) 144 456.34 T (1) 252 456.34 T (1/90) 256.63 456.34 T (14) 360 456.34 T (1816) 432 456.34 T (OARnet) 72 440.34 T (10/15/90) 144 440.34 T (not yet) 252 440.34 T (13) 360 440.34 T (135-140) 432 440.34 T 0 F (All the above deployments are using the Proteon OSPF implementation. There is one other ) 72 415.01 T -0.38 (deployment worth mentioning in this context. 3com has started to deploy OSPF on their corporate ) 72 401.01 P -0.36 (network. They have 8 of their routers running OSPF \050the 3com implementation\051, and are planning ) 72 387.01 P (on cutting over the remaining routers \05020 in all\051. Currently they have two operational routers run-) 72 373.01 T (ning OSPF and RIP simultaneously) 72 359.01 T (. One converts OSPF data to RIP data, and the other RIP data ) 241.82 359.01 T (to OSPF data. For more details, contact Dino Farinacci [dino].) 72 345.01 T 3 F (6.1 NSI) 72 311.67 T 0 F (The NASA Science Internet \050NSI\051 is a multiprotocol network, currently supporting both DECnet ) 72 285.01 T (and TCP/IP protocols. NSI\325) 72 271.01 T (s mission is to provide reliable high-speed communications to the ) 204.6 271.01 T (NASA science community) 72 257.01 T (. The NASA Science Internet connects with other national networks ) 199.8 257.01 T -0.3 (including the National Science Foundation\325) 72 243.01 P -0.3 (s NSFNET) 279.33 243.01 P -0.3 (, the Department of Ener) 331.12 243.01 P -0.3 (gy\325) 448.6 243.01 P -0.3 (s ESnet and the ) 463.92 243.01 P (Department of Defense\325) 72 229.01 T (s MILNET) 187.23 229.01 T (. NSI also has international connections to Japan, Australia, ) 239.31 229.01 T (New Zealand, Chile and several European countries.) 72 215.01 T (For more information on NSI, contact Jef) 72 189.01 T (frey Bur) 270.65 189.01 T (gan [jef) 310.73 189.01 T (f] or Milo Medin [medin].) 347.48 189.01 T 72 122 540 142.09 C 72 129.99 203.98 129.99 2 L 0.25 H 2 Z 0 X 0 K N 0 0 612 792 C 1 F 0 X 0 K (2. Archives of this mailing list are available from Ross V) 72 115.33 T (each [rrv].) 301.85 115.33 T FMENDPAGE %%EndPage: "9" 10 %%Page: "10" 10 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 10]) 493.7 73 T 72 108 540 684 R 7 X V 6 F 0 X (6.1.1 NSI\325) 72 264.02 T (s OSPF system) 125.53 264.02 T 0 F (NSI was one of the initial deployment sites for OSPF V) 72 238.02 T (ersion 1, having deployed the protocol in ) 337.85 238.02 T (April 1990. NSI has been running OSPF V2 since 1/1/91. They currently have 15 routers in their ) 72 224.02 T (OSPF system. This system is pictured in Figure 1. It consists of a nationwide collection of serial ) 72 210.02 T (lines, with ethernets at hub sites. The numbers associated to interfaces/links in Figure1 are the ) 72 196.02 T -0.12 (associated OSPF costs. Note that certain links have been weighted so that they are less preferable ) 72 182.02 P (than others.) 72 168.02 T ( Many of NSI\325) 72 142.02 T (s OSPF routers are speaking either RIP and/or EGP as well as OSPF) 141.62 142.02 T (. Routes from ) 469.13 142.02 T (these other routing protocols are selectively imported into their OSPF system as externals. The ) 72 128.02 T (current number of imported externals is 496.) 72 114.02 T 72 108 540 684 C 72 272.02 540 684 C 70.99 272.02 541.01 684 R 7 X 0 K V 449.99 629 479.99 653 R 4 X V 1 H 2 Z 0 X N 119.99 565 155 589 R N 427 660 427 622 2 L 4 H N 430 594 430 498 2 L N 173 610 174 458 2 L N 377.99 632 410 656 R 1 H N 119.99 527 156 551 R N 191.99 573 226 597 R N 119.99 453 155.99 480 R 3 X V 0 X N 119.99 492 157 516 R N 380.99 573 413 597 R N 452.99 546 485 570 R N 452.99 510 485 534 R N 380.99 525 413 549 R N 384.98 426 416.99 450 R N 434.99 426 467 450 R N 308.99 408 341 432 R N 218.99 435 251 459 R N 330.98 555 362.99 579 R N 428 642 450 642 2 L N 381 587 226 587 2 L N 409 644 425 644 2 L N 414 584 430 584 2 L N 453 559 430 559 2 L N 453 523 430 523 2 L N 413 538 429 538 2 L N 363 575 381 575 2 L N 396 632 397 597 2 L N 398 525 398 450 2 L N 418 438 435 438 2 L N 341 420 385 436 2 L N 252 449 381 537 2 L N 252 441 309 416 2 L N 208 572 236 459 2 L N 174 586 192 585 2 L N 156 576 172 576 2 L N 156 540 172 540 2 L N 157 503 174 503 2 L N 156 466 174 466 2 L N 374 354 408 378 R 4 X V 0 X N 310 355 344 379 R N 326 408 326 379 2 L N 306 336 418 336 2 L 4 H N 325 354 325 336 2 L 1 H N 392 354 393 337 2 L N 1 F (SURA) 382 641 T (NSS9) 453 638 T (GSFC2) 455 556 T (GSFC4) 455 520 T (GSFC1) 383 582 T (GSFC3) 383 534 T (MSFC) 387 435 T (SSC) 441 436 T (STScI) 335 565 T (JSC) 317 419 T (RICE) 315 364 T (NSS1) 377 362 T (1) 399.96 362 T (NSS13) 124 463 T (ARC1) 197 582 T (FIX-East) 409 671 T (FIX-W) 156 626 T (est) 184.07 626 T (Sesquinet) 256 335 T (JPL) 226 444 T (ARC2) 126 574 T (ARC) 128 541 T (Ext1) 129 530 T (ARC) 128 507 T (Ext2) 129 495 T (1) 400 612 T (1) 439 528 T (1) 439 564 T (1) 181 592 T (1) 161 581 T (1) 160 545 T (1) 162 509 T (1) 330 393 T (1) 416 588 T (1) 418 544 T (2) 295 595 T (2) 359 434 T (2) 403 490 T (2) 323 486 T (2) 224 524 T (3) 281 433 T 6 F (Figur) 165 290 T (e 1: The NASA Science Internet OSPF System) 193.44 290 T 72 108 540 684 C 0 0 612 792 C FMENDPAGE %%EndPage: "10" 11 %%Page: "11" 11 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 11]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (All NSI externals are imported as OSPF type 2 metrics. In addition, NSI uses the OSPF external ) 72 676 T (route tag to manage the readvertisement of external routes. For example, a route learned at one ) 72 662 T (edge of the NSI system via EGP can be tagged with the number of the AS from which it was ) 72 648 T (learned. Then, as the OSPF external LSA describing this route is \337ooded through the OSPF sys-) 72 634 T (tem, this tagging information is distributed to all the other AS boundary routers. A router on the ) 72 620 T (other edge of the NSI can then say that it wants to readvertise \050via EGP\051 routes learned from one ) 72 606 T (particular AS but not routes learned from another AS. This allows NSI to implement transit poli-) 72 592 T (cies at the granularity of Autonomous Systems, instead of network numbers, which greatly ) 72 578 T (reduces the network\325) 72 564 T (s con\336guration burden.) 171.92 564 T (NSI has also experimented with OSPF stub areas, in order to support routers having a small ) 72 538 T (amount of memory) 72 524 T (. ) 163.16 524 T 6 F (6.1.2 NSI - Deployment analysis) 72 492 T 0 F -0.26 (NSI ran a couple of experiments after OSPF\325) 72 466 P -0.26 (s deployment to test OSPF\325) 285.7 466 P -0.26 (s conver) 415.94 466 P -0.26 (gence time in the ) 455.76 466 P -0.21 (face of network failures, and to compare the level of routing traf) 72 452 P -0.21 (\336c in OSPF with the level of rout-) 377.89 452 P (ing traf) 72 438 T (\336c in RIP) 106.76 438 T (. These experiments were included in NSI status reports to the OSPF plenary) 151.41 438 T (.) 519.38 438 T -0.46 (The \336rst experiment consisted of running a continuous ICMP ping, and then bringing down one of ) 72 412 P (the links in the ping packet\325) 72 398 T (s path. They then timed how long it took OSPF to \336nd an alternate ) 204.92 398 T (path, by noticing when the pings resumed. The result of this experiment is contained in Milo ) 72 384 T (Medin\325) 72 370 T (s \322NASA Sciences Internet Report\323 in [8]. It shows that the interrupted ping resumed in ) 106.65 370 T (three seconds.) 72 356 T (The second experiment consisted in analyzing the amount of routing protocol traf) 72 330 T (\336c that \337ow ) 463.18 330 T (over an NSI link. One of the NSI links was installed, but did not have any active users yet. For ) 72 316 T (this reason, all traf) 72 302 T (\336c that \337owed over the link was routing protocol traf) 161.05 302 T (\336c. The link was instru-) 415 302 T (mented to continuously measure the amount of bandwidth consumed, \336rst in the case where RIP ) 72 288 T -0.13 (was running, and then in the case of where OSPF was running. The result is shown graphically in ) 72 274 P (Jef) 72 260 T (frey Bur) 85.77 260 T (gan\325) 125.85 260 T (s \322NASA Sciences Internet\323 report in [9]. It shows that OSPF consumes many ) 146.5 260 T (times less network bandwidth than RIP) 72 246 T (.) 258.89 246 T 3 F (6.2 BARRNet) 72 212.67 T 0 F (BARRNet is the NSFNet regional network in Northern California. At the present time, it serves ) 72 186 T (approximately 80 member sites in an area stretching from Sacramento in the north-east to ) 72 172 T (Monterey in the in the south-west. Sites are connected to the network at speeds from 9.6Kbps to ) 72 158 T (full T1 using Proteon and cisco routers as well as a Xylogics terminal server) 72 144 T (. The membership is ) 437.42 144 T (composed of a mix of university) 72 130 T (, government, and commercial or) 226.79 130 T (ganizations. BARRNet has ) 385.78 130 T -0.1 (interconnections to the NSFNet \050peering with both T1 and T3 backbones at Stanford University\051, ) 72 116 P FMENDPAGE %%EndPage: "11" 12 %%Page: "12" 12 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 12]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (ESNet \050peering at LLNL, with additional multi-homed sites at LBL, SLAC, and NASA Ames\051, ) 72 676 T (and DDN national networks \050peering at the FIX network at NASA Ames\051, and to the statewide ) 72 662 T (networks of the University of California \050peering at U.C. Berkeley\051 and the California State Uni-) 72 648 T (versity system \050peering at San Francisco State and Sacramento State\051.) 72 634 T (T) 72 608 T (opologically) 78.49 608 T (, the network consists of fourteen OSPF-speaking Proteon routers, which as a ) 137.68 608 T -0.27 (\322core\323, with six of these redundantly connected into a ring. All \322core\323 sites are interconnected via ) 72 594 P (full T1 circuits. Other member sites attach as \322stub\323 connections to the \322core\323 sites. The bulk of ) 72 580 T (these are connected in a \322star\323 con\336guration at Stanford University) 72 566 T (, with lesser numbers at other ) 393.3 566 T (\322core\323 sites. ) 72 552 T (Contact V) 72 526 T (ince Fuller [vaf] for more information on BARRNet.) 120.25 526 T FMENDPAGE %%EndPage: "12" 13 %%Page: "13" 13 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 13]) 493.7 73 T 72 108 540 684 R 7 X V 6 F 0 X (6.2.1 BARRNet\325) 72 217 T (s OSPF system) 157.49 217 T 0 F (BARRNet was also one of the initial deployment sites for OSPF V) 72 191 T (ersion 1, having deployed the ) 390.82 191 T -0.02 (protocol in April 1990. BARRNet has been running OSPF V2 since November 1990. They cur-) 72 177 P -0.16 (rently have 14 routers in their OSPF system. The BARRNet OSPF system is pictured in Figure 2. ) 72 163 P (It consists of a collection of T1 serial lines, with ethernets at hub sites. ) 72 149 T 72 108 540 684 C 72 225 540 684 C 70.99 225 541.01 684 R 7 X 0 K V 114.99 594.18 150 618.18 R 1 H 2 Z 0 X N 441.98 592 474.99 617 R N 114.99 556.18 151 580.18 R N 186.99 439 221 463 R N 114.98 637 150.98 664 R 3 X V 0 X N 114.99 521.18 152 545.18 R N 438.99 655 471 679 R N 249.99 505 282 529 R N 370.98 559 402.99 583 R N 375.99 505 408 532 R N 433.98 415 465.99 439 R N 186.99 367 219 391 R N 289.98 559 321.99 583 R N 151 605.18 167 605.18 2 L N 151 569.18 167 569.18 2 L N 152 532.18 169 532.18 2 L N 150.99 650 168.99 650 2 L N 272.99 391 307 416 R 7 X V 0 X N 287.98 304 321.99 328 R N 1 F (NSS13) 118.99 647 T 483.99 475 483.99 376 2 L 4 H N 168.99 664 168.99 358 2 L N 240.99 541 429.99 541 2 L N 249.99 286 465.99 286 2 L N 186.99 448 168.99 448 2 L 1 H N 187 379 168 379 2 L N 222 455 265 505 2 L N 266 529 266 541 2 L N 306 559 306 543 2 L N 388 559 388 543 2 L N 403 570 458 592 2 L N 392 531 393 542 2 L N 460 617 456 655 2 L N 394 505 434 427 2 L N 323 315 450 415 2 L N 202 367 288 315 2 L N 305 304 305 287 2 L 10 X N 475 605 504 605 2 L 0.5 H N 471 668 502 669 2 L N 389 583 379 612 2 L 11 X N 305 585 288 621 2 L N 310 584 327 623 2 L N 282 517 314 517 2 L 10 X N 221 448 295 446 2 L N 116 611 88 627 2 L N 115 605 82 615 2 L N 113 601 82 603 2 L N 115 598 82 589 2 L N 114 573 84 580 2 L N 115 569 80 567 2 L N 114 564 78 557 2 L N 114 559 78 547 2 L N 466 428 482 428 2 L 1 H 0 X N 306 304 306 287 2 L N 322 310 360 303 2 L 0.5 H 10 X N 219 379 273 403 2 L 1 H 0 X N 116 534 86 535 2 L 0.5 H 10 X N 116 529 86 519 2 L N 332 437 332 371 2 L 4 H 0 X N 307 403 331 403 2 L 1 H N (BR10) 121 602 T (BR5) 193 448 T (BR6) 193 376 T (BR16) 120 565 T (BR28) 122 530 T (BR4) 256 515 T (BR29) 295 568 T (BR3) 378 569 T (BR21) 444 665 T (BR1) 449 602 T (BR22) 381 516 T (BR9) 439 424 T (BR17) 279 400 T (BR8) 296 313 T 6 F (Figur) 208 248 T (e 2; The BARRNet OSPF system) 236.44 248 T 72 108 540 684 C 0 0 612 792 C FMENDPAGE %%EndPage: "13" 14 %%Page: "14" 14 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 14]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (Most of BARRNet\325) 72 676 T (s OSPF routers are speaking either RIP and/or EGP as well as OSPF) 165.96 676 T (. Routes ) 493.46 676 T -0.38 (from these other routing protocols are selectively imported into their OSPF system as externals. A ) 72 662 P -0.4 (lar) 72 648 P -0.4 (ge number of external routes are imported; the current number is1816. The bulk of these are the ) 84.44 648 P (T1 NSFNet routes, followed by several hundred NSN routes, around 60-80 BARRNet routes ) 72 634 T (from the non-OSPF system, and several dozen from ESNet.) 72 620 T -0.23 (All external routes are imported into the BARRNet system as external type 1 metrics. In addition, ) 72 594 P -0.02 (BARRnet, like NSI, uses the OSPF\325) 72 580 P -0.02 (s external route tagging feature to help manage the readver-) 244.84 580 P (tisement of external routes via EGP) 72 566 T (.) 241.55 566 T (BARRnet is also using four stub OSPF areas in order to collapse subnet information. These stub ) 72 540 T (areas all consist of a single LAN. They do not contain any OSPF routers in their interiors.) 72 526 T 6 F (6.2.2 BARRNet - Deployment analysis) 72 494 T 0 F (Initial deployment of OSPF V) 72 468 T (ersion 1 in BARRNet pointed to the need for two new protocol fea-) 215.26 468 T (tures that were added to OSPF V2, namely:) 72 454 T (\245) 72 434 T (Addition of the ) 85.54 434 T 6 F (forwarding addr) 161.83 434 T (ess) 247.23 434 T 0 F ( to OSPF external LSAs. This eliminated the extra hops ) 261.89 434 T (that were being taken in BARRNet when only routers BR5 and BR6 were exchanging EGP ) 85.54 420 T (information with the NSS \050see Figure 2\051. W) 85.54 406 T (ithout the forwarding address feature, that meant ) 295.59 406 T -0.45 (that NSFNet traf) 85.54 392 P -0.45 (\336c handled by routers BR10, BR16 and BR28 was taking an extra hop to get to ) 164.38 392 P (the NSS.) 85.54 378 T (\245) 72 358 T (Addition of ) 85.54 358 T 6 F (stub ar) 144.17 358 T (eas) 180.27 358 T 0 F (. This was an attempt to get OSPF running on some of the BARRNet ) 196.26 358 T (routers that had insuf) 85.54 344 T (\336cient memory to deal with all of BARRNet\325) 186.92 344 T (s external routes.) 403.79 344 T 3 F (6.3 OARnet) 72 310.67 T 0 F -0.04 (OARnet, the Ohio Academic Resources Network, is the regional network for the state of Ohio. It ) 72 284 P -0.27 (serves the entire higher education community) 72 270 P -0.27 (, providing Ohio schools access to colleagues world-) 288.72 270 P (wide. The Ohio Supercomputer Center and the NSF Supercomputer Centers are reached through ) 72 256 T (OARnet. Libraries, databases, national and international laboratories and research centers are ) 72 242 T (accessible to faculty) 72 228 T (, helping make Ohio schools competitive.) 168.48 228 T -0.28 (OARnet was established in 1987 to provide state-wide access to the CRA) 72 202 P -0.28 (Y at the Ohio Supercom-) 420.55 202 P -0.43 (puter Center in Columbus, Ohio. Since then it has evolved into a network supporting all aspects of ) 72 188 P -0.42 (higher education within Ohio. A primary goal of OARnet is to facilitate collaborative projects and ) 72 174 P (sharing of resources between institutions, including those outside the state. OARnet connections ) 72 160 T (are available to Ohio academic institutions and corporations engaged in research, product devel-) 72 146 T -0.36 (opment, or instruction. Colleges, universities, and industries currently use OARnet connections to ) 72 132 P (communicate within the state and with colleagues around the country) 72 118 T (.) 403.99 118 T FMENDPAGE %%EndPage: "14" 15 %%Page: "15" 15 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 15]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (OARnet uses the Internet \050TCP/IP\051 and DECNET protocols. OARnet participants using TP/IP ) 72 676 T -0.07 (protocols are connected to the worldwide Internet, which includes all the major networks open to ) 72 662 P -0.41 (non-classi\336ed research. OARnet is also connected to NSFNet, the national research and education ) 72 648 P (network sponsored by the National Science Foundation. It has gateways to BITNET) 72 634 T (, CSNET) 474.83 634 T (, ) 517.93 634 T (CICNet \050a network connecting the Big T) 72 620 T (en universities\051, and the NASA Science Internet. ) 267.02 620 T (For more information on OARnet, contact Kannan V) 72 594 T (aradhan [kannan].) 325.16 594 T 6 F (6.3.1 OARnet\325) 72 562 T (s OSPF system) 148.17 562 T 0 F (OARnet has been running OSPF V) 72 536 T (ersion 1 since October 15, 1990. They currently have 14 rout-) 238.9 536 T (ers in their OSPF system. The OARnet OSPF system is pictured in Figure 3.) 72 522 T FMENDPAGE %%EndPage: "15" 16 %%Page: "16" 16 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 16]) 493.7 73 T 72 108 540 684 R 7 X V 0 X -0.08 (There are 29 sites connected directly to the OARnet backbone. All 13 of OARnet\325) 72 217 P -0.08 (s OSPF routers ) 464.62 217 P (act as ASBRs. There are 40 OSPF internal routes on OARnet\325) 72 203 T (s network, and they import about ) 369.47 203 T (120 routes from RIP) 72 189 T (. OARnet runs EGP on the DMZnet at Columbus, which connects them to ) 168.94 189 T (CICNet. The router connecting OARnet to DMZnet \050OAR1 in Figure 3\051 runs EGP on the ) 72 175 T -0.12 (DMZnet side, and OSPF and RIP on the OARnet backbone. No EGP routes are imported into the ) 72 161 P (OSPF system. The OAR1 router is con\336gured to generate a default when EGP routes are avail-) 72 147 T -0.2 (able. The OAR1 router is the keystone for routing on OARnet\325) 72 133 P -0.2 (s network, in that it acts as an inter-) 370.75 133 P (mediary for all of OARnet\325) 72 119 T (s RIP centric routers.) 202.57 119 T 72 108 540 684 C 72 225 540 684 C 71.5 225 540.5 684 R 7 X 0 K V 251.5 495 286.51 519 R 1 H 2 Z 0 X N 377.5 477 410.51 502 R N 350.5 423 386.51 447 R N 209.99 621 242.5 648 R N 413.5 525 450.51 549 R N 296.5 621 328.51 645 R N 462.49 576 494.5 600 R N 255.49 549 287.5 576 R N 413.5 360 445.51 384 R N 300.49 297 332.5 321 R N 255.49 423 287.5 447 R N 179.5 360 213.51 385 R 7 X V 0 X N 298.49 363 332.5 387 R N 263.99 281.85 479.99 281.85 2 L 4 H N 6 F (Figur) 222 243.85 T (e 3; The OARnet OSPF system) 250.44 243.85 T 116.5 342 476.5 342 2 L N 224.5 666 332.5 666 2 L N 197.5 387 251.5 567 2 L 1 H N 206.5 387 251.5 504 2 L N 197.5 360 197.5 342 2 L N 314.5 360 314.5 342 2 L N 431.5 360 431.5 342 2 L N 314.5 387 269.5 423 2 L N 269.5 450 269.5 495 2 L N 287.5 504 350.5 450 2 L N 269.5 576 314.5 621 2 L N 314 645 314 666 2 L N 228 648 228 666 2 L N 329 634 462 589 2 L N 414 384 387 424 2 L N 433 384 399 477 2 L N 438 384 433 525 2 L N 442 384 478 576 2 L N 288 563 414 536 2 L N 289 556 378 489 2 L N 316 321 316 342 2 L N 317 297 317 284 2 L N 1 F (OAR1) 304 305 T (OAR2) 417 369 T (OAR4) 182 368 T (OEB1) 303 371 T (kent) 384 486 T (mco) 360 433 T (T) 255 505 T (oledo) 260.41 505 T (akron) 421 534 T (cwru) 467 586 T (ASP_1) 258 558 T (CSP_1) 300 631 T (CSP_2) 213 632 T (bgsu) 261 432 T (DMZnet) 385 292 T 446 378 478 411 2 L 0.5 H 10 X N 447 373 483 387 2 L N 447 368 486 374 2 L N 209 637 161 659 2 L N 210 631 156 640 2 L N 210 627 148 623 2 L N 329 640 373 653 2 L N 255 571 208 581 2 L N 264 576 258 594 2 L 11 X N 289 440 312 448 2 L 10 X N 288 434 310 434 2 L N 288 510 304 514 2 L N 185 386 172 414 2 L 11 X N 180 380 155 402 2 L 10 X N 181 374 150 380 2 L N 414 375 386 388 2 L N 334 374 346 376 2 L N 351 434 338 434 2 L N 389 503 376 520 2 L 11 X N 398 504 405 518 2 L N 420 548 404 571 2 L N 428 550 426 575 2 L N 436 550 441 573 2 L N 442 549 452 572 2 L N 470 600 452 627 2 L N 476 600 476 632 2 L N 482 601 491 630 2 L N 72 108 540 684 C 0 0 612 792 C FMENDPAGE %%EndPage: "16" 17 %%Page: "17" 17 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 17]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (OARnet uses the Event Logging System on its Proteon routers to generate traps for \322interesting\323 ) 72 676 T (events related to routing. They have these traps sent to an SNMP management station, where the ) 72 662 T (logs are collected for later perusal. ) 72 648 T 6 F (6.3.2 OARnet - Deployment analysis) 72 616 T 4 F (OARnet is monitoring their OSPF system via collection of traps on their SNMP management sta-) 72 590 T (tion. The following is a r) 72 576 T (eport on their observations. It has been edited slightly to conform better ) 190.18 576 T (with the other text and maps pr) 72 562 T (esented in this r) 221.47 562 T (eport. For mor) 297.32 562 T (e information, contact Kannan ) 368.5 562 T (V) 72 548 T (aradhan [kannan]:) 78 548 T 0 F -0.01 (3 of our 10 DS1 circuits are on digital microwave, and these tend to \337ap occasionally) 72 522 P -0.01 (. Our obser-) 480.05 522 P (vations indicate that the routers bring up links, and adjacencies, on average, in about 2 seconds. ) 72 508 T -0.37 (Routes fallback to alternate backup paths instantly) 72 494 P -0.37 (. Whole blocks of routes cut over the instant the ) 310.85 494 P (adjacencies are formed.) 72 480 T (In contrast to this, our RIP routes would take about 3-6 minutes to cutover) 72 454 T (, and, on occasion, ) 427.96 454 T (would not cut back to the preferred paths. This was our prime motivation in switching to OSPF) 72 440 T (.) 527.76 440 T (W) 72 414 T (e attempted to duplicate Milo Medin\325) 82.36 414 T (s ping test to dramatically illustrate the performance of ) 261.26 414 T -0.06 (RIP over OSPF) 72 400 P -0.06 (. T) 145.55 400 P -0.06 (o do this, we selected a host on the farthest point from our workstation, and ran ) 157.97 400 P (a continuous ping to it. W) 72 386 T (e would then bring down a primary DS1 circuit, and watch the time it ) 195.64 386 T -0.19 (took to switch to the fallback route. Following this, we would bring the circuit back up, and study ) 72 372 P -0.01 (the time it took to re-sync to the new path. ) 72 358 P -0.01 (W) 278.08 358 P -0.01 (ith RIP) 288.92 358 P -0.01 (, we were unable to fully complete the exper-) 321.9 358 P (iment, because the farthest point was exactly equal to the new \050and preferred\051 primary path, and ) 72 344 T -0.2 (therefore, RIP would never choose it on it\325) 72 330 P -0.2 (s own, until the path it was currently using failed. W) 275.14 330 P -0.2 (ith ) 524.53 330 P -0.1 (OSPF) 72 316 P -0.1 (, it took about 2 seconds to synchronize over a new) 99.71 316 P -0.1 (, much slower 56kb path, and less than a ) 343.39 316 P (second when the DS1 circuit came back up.) 72 302 T (Here are some more observations of the OARnet OSPF system\325) 72 276 T (s behavior:) 377.46 276 T (\245) 72 256 T -0.2 (131.187.36.0 is the 56kb line to Kent State University) 85.54 256 P -0.2 (. Kent also has a DS1 circuit leading into ) 341.97 256 P -0.09 (ASP) 85.54 242 P -0.09 (, the Akron Pop. Likewise, UAkron.edu has a similar con\336guration. A roundabout backup ) 106.2 242 P (path exists when traf) 85.54 228 T (\336c heads up to Cleveland over a couple of DS1 circuits, and then down a ) 184.91 228 T (56kb backup path used by another school in the Cleveland area.) 85.54 214 T (Some statistical information: ) 85.54 196 T 1 F (1. 09:55:17: SPF) 144 179.33 T (.37: new route to Net 131.187.36.5, type SPF cost 32) 210.68 179.33 T (2. 09:55:18: SPF) 144 163.33 T (.37: new route to Net 131.187.36.6, type SPF cost 22) 210.68 163.33 T (3. 09:55:20: SPF) 144 147.33 T (.21: State Change, nbr 131.187.27.6, new state , event 9) 210.68 147.33 T ( ) 72 131.33 T (4. 09:55:21: SPF) 144 131.33 T (.37: new route to Net 131.187.36.5, type SPF cost 31) 210.68 131.33 T (5. 09:55:22: SPF) 144 115.33 T (.37: new route to Net 131.187.36.6, type SPF cost 21) 210.68 115.33 T FMENDPAGE %%EndPage: "17" 18 %%Page: "18" 18 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 18]) 493.7 73 T 72 108 540 684 R 7 X V 1 F 0 X ( ) 72 677.33 T (6. 09:55:28: SPF) 144 677.33 T (.21: State Change, nbr 131.187.21.5, new state , event 9 ) 210.68 677.33 T (7. 09:55:29: SPF) 144 661.33 T (.21: State Change, nbr 131.187.51.6, new state , event 9) 210.68 661.33 T ( ) 72 645.33 T (8. 09:55:31: SPF) 144 645.33 T (.37: new route to Net 131.187.36.5, type SPF cost 22) 210.68 645.33 T ( ) 72 629.33 T (9. 09:55:33: SPF) 144 629.33 T (.37: new route to Net 131.187.36.5, type SPF cost 1) 210.68 629.33 T (1) 417.42 629.33 T 0 F (The Akron router restarts, and has to re-sync with all the lines. This restart is con\336rmed when ) 85.54 612 T (one looks at the traps from gwCSP1. T) 85.54 598 T (raps from gwASP1 still do not get through to us, ) 271.67 598 T (because traps are sent via UDP) 85.54 584 T (, and gwASP1\325) 233.08 584 T (s routing tables are not fully con\336gured yet.) 305.38 584 T -0.44 (Events 1 and 2 are route changes routing traf) 85.54 566 P -0.44 (\336c via Cleveland, across 2 DS1 circuits and a 56kb ) 296.94 566 P (line.) 85.54 552 T -0.03 (When the DS1 circuit to UAkron came up, routes instantly cut over to use this as a better least ) 85.54 534 P (cost path. This is shown in events 3, 4 and 5.) 85.54 520 T -0.41 (In a few seconds, the line to Columbus is the next one up. This is event 6. Event 8 relates to this ) 85.54 502 P (cutover) 85.54 488 T (, and is the best path yet. When the DS1 circuit to Kent is up, the link is used instantly) 121.02 488 T (.) 533.33 488 T (W) 85.54 470 T (e are able to make such a de\336nitive conclusion of these traps on the basis of the topological ) 95.9 470 T (information that we have about the network and the means used to monitor them.) 85.54 456 T (\245) 72 436 T (T) 85.54 436 T (o illustrate the time required to fully synchronize a database, we piece together a few adja-) 92.02 436 T (cency forming traces...) 85.54 422 T -0.19 (Please bear in mind that these time stamps are the time stamps on the management station, and ) 85.54 404 P -0.11 (are not to be taken as the absolute truth. Things we haven\325) 85.54 390 P -0.11 (t taken into account are ) 363.9 390 P -0.11 (transit times ) 478.25 390 P -0.16 (of messages, ) 85.54 376 P -0.16 (ordering of events \050SNMP traps are sent using UDP\051, ) 149.5 376 P -0.16 (loss of event reports \050recall ) 407.23 376 P (that an entire synchronization ) 85.54 362 T (sequence of gwASP1 on the ASP-CSP link is missing\051, ) 230.76 362 T (etc.) 499.3 362 T (The trace below corresponds to the Akron router) 85.54 344 T (, gwASP1 bring up the link in the previous ) 318.53 344 T (section. This is as observed on the other end of the line, gwCSP1.) 85.54 330 T 1 F (REPOR) 144 313.33 T (T DA) 175.61 313.33 T (TE: 02/26/91) 197.54 313.33 T (ROUTER: gwcsp1) 250.57 313.33 T (09:55:06: SPF) 144 297.33 T (.15: State Change, ifc 131.187.22.6, new state , event 1) 422.44 297.33 T -0.13 (09:55:06: GW) 144 281.33 P -0.13 (.xxx: Link Up T) 200.41 281.33 P -0.13 (rap: 09:55:07: SPF) 264.63 281.33 P -0.13 (.37: new route to Net 131.187.22.5, type SPF cost ) 339.09 281.33 P (1) 225 269.33 T (09:55:07: SPF) 144 253.33 T (.21: State Change, nbr 131.187.22.5, new state , event 1) 200.69 253.33 T (09:55:09: SPF) 144 237.33 T (.37: new route to Net 131.187.27.5, type SPF cost 22) 200.69 237.33 T (09:55:1) 144 221.33 T (1: SPF) 174.17 221.33 T (.21: State Change, nbr 131.187.22.5, new state , event 14 ) 200.32 221.33 T (09:55:1) 144 205.33 T (1: SPF) 174.17 205.33 T (.21: State Change, nbr 131.187.22.5, new state <2-W) 200.32 205.33 T (ay>, event 3) 411.96 205.33 T (09:55:12: SPF) 144 189.33 T (.21: State Change, nbr 131.187.22.5, new state , event 5) 200.69 189.33 T (09:55:12: SPF) 144 173.33 T (.21: State Change, nbr 131.187.22.5, new state , event 9) 200.69 173.33 T (09:55:12: SPF) 144 157.33 T (.21: State Change, nbr 131.187.22.5, new state , event 6) 200.69 157.33 T 0 F (Below) 85.54 140 T (, is another trace of the same router restart sequence, where the router is proceeding to ) 116.07 140 T (bring up other DS1 circuits. Bringing up the \336rst adjacency took about 5 seconds. Subsequent ) 85.54 126 T (adjacencies take the router less than a second as seen below) 85.54 112 T (.) 371.18 112 T FMENDPAGE %%EndPage: "18" 19 %%Page: "19" 19 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 19]) 493.7 73 T 72 108 540 684 R 7 X V 1 F 0 X (REPOR) 144 677.33 T (T DA) 175.61 677.33 T (TE: 02/26/91) 197.54 677.33 T (ROUTER: gwasp1 ) 250.57 677.33 T (09:55:20: SPF) 144 661.33 T (.15: State Change, ifc 131.187.27.5, new state , event 1 ) 422.44 661.33 T (09:55:20: GW) 144 645.33 T (.xxx: Link Up T) 200.55 645.33 T (rap: 09:55:20: SPF) 265.16 645.33 T (.21: State Change, nbr 131.187.27.6, new state ) 339.89 645.33 T (, event 1 ) 225 633.33 T (09:55:20: SPF) 144 617.33 T (.21: State Change, nbr 131.187.27.6, new state , event 14 ) 200.69 617.33 T (09:55:20: SPF) 144 601.33 T (.21: State Change, nbr 131.187.27.6, new state , event 5 ) 200.69 601.33 T (09:55:20: SPF) 144 585.33 T (.21: State Change, nbr 131.187.27.6, new state , event 9 ) 200.69 585.33 T (09:55:21: SPF) 144 569.33 T (.21: State Change, nbr 131.187.27.6, new state , event 6) 200.69 569.33 T (09:55:24: SPF) 144 553.33 T (.21: State Change, nbr 131.187.51.6, new state , event 1 ) 200.69 553.33 T (09:55:24: SPF) 144 537.33 T (.21: State Change, nbr 131.187.21.5, new state , event 1 ) 200.69 537.33 T (09:55:25: SPF) 144 521.33 T (.37: new route to Net 131.187.21.6, type SPF cost 13 ) 200.69 521.33 T (09:55:25: SPF) 144 505.33 T (.37: new route to Net 131.187.51.5, type SPF cost 22 ) 200.69 505.33 T (09:55:28: SPF) 144 489.33 T (.21: State Change, nbr 131.187.21.5, new state , event 14 ) 200.69 489.33 T (09:55:28: SPF) 144 473.33 T (.21: State Change, nbr 131.187.21.5, new state <2-W) 200.69 473.33 T (ay>, event 3 ) 412.33 473.33 T (09:55:28: SPF) 144 457.33 T (.21: State Change, nbr 131.187.21.5, new state , event 5 ) 200.69 457.33 T (09:55:28: SPF) 144 441.33 T (.21: State Change, nbr 131.187.21.5, new state , event 9 ) 200.69 441.33 T (09:55:28: SPF) 144 425.33 T (.21: State Change, nbr 131.187.21.5, new state , event 6 ) 200.69 425.33 T (09:55:29: SPF) 144 409.33 T (.37: new route to Net 131.187.51.6, type SPF cost 1 ) 200.69 409.33 T (09:55:29: SPF) 144 393.33 T (.37: new route to Net 131.187.21.5, type SPF cost 1 ) 200.69 393.33 T (09:55:29: SPF) 144 377.33 T (.21: State Change, nbr 131.187.51.6, new state , event 5 ) 200.69 377.33 T (09:55:29: SPF) 144 361.33 T (.21: State Change, nbr 131.187.51.6, new state , event 14 ) 200.69 361.33 T (09:55:29: SPF) 144 345.33 T (.21: State Change, nbr 131.187.51.6, new state <2-W) 200.69 345.33 T (ay>, event 3 ) 412.33 345.33 T (09:55:29: SPF) 144 329.33 T (.21: State Change, nbr 131.187.51.6, new state , event 9 ) 200.69 329.33 T (09:55:29: SPF) 144 313.33 T (.21: State Change, nbr 131.187.51.6, new state , event 6) 200.69 313.33 T 0 F -0.25 (A transient fault on a DS1 circuit, causes the line to \337ap. All routers quickly reroute around the ) 85.54 296 P (\337ap, and the router itself takes about 2 seconds to bring up the adjacency once more.) 85.54 282 T 1 F (REPOR) 144 265.33 T (T DA) 175.61 265.33 T (TE: 02/26/91) 197.54 265.33 T (ROUTER: gwasp1 ) 250.57 265.33 T (14:33:43: GW) 144 249.33 T (.xxx: Link Up T) 200.55 249.33 T (rap: ) 265.16 249.33 T (14:34:19: SPF) 144 233.33 T (.15: State Change, ifc 131.187.22.5, new state , event 7 ) 200.69 233.33 T (14:34:19: GW) 144 217.33 T (.xxx: Link Failure T) 200.55 217.33 T (rap:) 281.27 217.33 T (14:34:19: SPF) 144 201.33 T (.47: Net 131.187.22.6 now unreachable ) 200.69 201.33 T (14:34:36: SPF) 144 185.33 T (.15: State Change, ifc 131.187.22.5, new state , event 1 ) 422.44 185.33 T (14:34:36: GW) 144 169.33 T (.xxx: Link Up T) 200.55 169.33 T (rap: ) 265.16 169.33 T (14:34:37: SPF) 144 153.33 T (.37: new route to Net 131.187.22.6, type SPF cost 1 ) 200.69 153.33 T (14:34:45: SPF) 144 137.33 T (.21: State Change, nbr 131.187.22.6, new state <2-W) 200.69 137.33 T (ay>, event 3 ) 412.33 137.33 T (14:34:45: SPF) 144 121.33 T (.21: State Change, nbr 131.187.22.6, new state , event 1 ) 200.69 121.33 T FMENDPAGE %%EndPage: "19" 20 %%Page: "20" 20 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 20]) 493.7 73 T 72 108 540 684 R 7 X V 1 F 0 X (14:34:46: SPF) 144 677.33 T (.21: State Change, nbr 131.187.22.6, new state , event 14 ) 200.69 677.33 T (14:34:46: SPF) 144 661.33 T (.21: State Change, nbr 131.187.22.6, new state , event 5 ) 200.69 661.33 T (14:34:47: SPF) 144 645.33 T (.21: State Change, nbr 131.187.22.6, new state , event 9 ) 200.69 645.33 T (14:34:47: SPF) 144 629.33 T (.21: State Change, nbr 131.187.22.6, new state , event 6) 200.69 629.33 T 0 F (\245) 72 610 T (On the amount of time it takes for a router to restart, and become fully synchronized. T) 85.54 610 T (aking ) 502.39 610 T (the logs in the previous instance, we notice that the CSP-ASP link comes up at 9:55:06. The ) 85.54 596 T (last link is observed to be up at 9:55:29, which is less than a minute.) 85.54 582 T (\245) 72 562 T (On the RIP equivalent of the tests, it took us 3 minutes to cutover to the slower speed fallback ) 85.54 562 T (route, and we lost countless many packets. The routes never cutover to the higher speed paths ) 85.54 548 T (when available, and we waited well over 30 minutes watching this, wondering why) 85.54 534 T (. Unfortu-) 484.47 534 T (nately) 85.54 520 T (, at this point, we seem to have lost the RIP statistics.) 114.07 520 T (On the OSPF version, we have...) 85.54 502 T 1 F ({nisca danw 51}) 144 485.33 T (ping 131.187.25.6 PING 131.187.25.6 \050131.187.25.6\051: ) 144 469.33 T (56 data bytes 64 bytes from 131.187.25.6: icmp seq=0 ttl\050255-ttl\051=54\050201\051. time=20 ms) 144 453.33 T ([...] ) 225 437.33 T (64 bytes from 131.187.25.6: icmp seq=10 ttl\050255-ttl\051=54\050201\051. time=20 ms) 144 421.33 T ( ) 72 405.33 T ( ||) 225 405.33 T (T1 down ) 231.5 405.33 T (64 bytes from 131.187.25.6: icmp seq=14 ttl\050255-ttl\051=54\050201\051. time=180 ms ) 144 389.33 T (64 bytes from 131.187.25.6: icmp seq=15 ttl\050255-ttl\051=54\050201\051. time=60 ms) 144 373.33 T ( ) 72 357.33 T ([...]) 225 357.33 T (64 bytes from 131.187.25.6: icmp seq=38 ttl\050255-ttl\051=8\050247\051. time=1300 ms ) 144 341.33 T (64 bytes from 131.187.25.6: icmp seq=39 ttl\050255-ttl\051=54\050201\051. time=820 ms) 144 325.33 T ( ) 72 309.33 T ( ||) 225 309.33 T (Tl Up) 231.5 309.33 T (64 bytes from 131.187.25.6: icmp seq=40 ttl\050255-ttl\051=54\050201\051. time=20 ms ) 144 293.33 T (64 bytes from 131.187.25.6: icmp seq=41 ttl\050255-ttl\051=54\050201\051. time=20 ms) 144 277.33 T (131.187.25.6 PING Statistics ) 144 261.33 T (51 packets transmitted, 48 packets received, 5% packet loss) 144 245.33 T ( ) 72 229.33 T (round-trip \050ms\051 min/avg/max = 20/277/1300 ) 144 229.33 T 3 F (6.4 Featur) 72 196.67 T (es exer) 136.64 196.67 T (cised during operational deployment) 177.18 196.67 T 0 F (In operational environments, all basic mechanisms of the OSPF protocol have been exercised. ) 72 170 T (These mechanisms include:) 72 156 T (\245) 72 136 T 6 F (Designated Router election) 85.54 136 T 0 F (. There have been operational deployments have as many as 8 ) 223.44 136 T (OSPF routers attached to a single broadcast network.) 85.54 122 T FMENDPAGE %%EndPage: "20" 21 %%Page: "21" 21 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 21]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T 6 F (Database synchr) 85.54 676 T (onization.) 170.27 676 T 0 F ( This includes OSPF\325) 220.58 676 T (s ) 322.87 676 T 6 F (adjacency bringup) 330.54 676 T 0 F ( and ) 426.16 676 T 6 F (r) 449.48 676 T (eliable \337ooding) 454.58 676 T 0 F ( ) 532.23 676 T (procedures. Lar) 85.54 662 T (ge operational OSPF link state databases \050e.g., BARRNet\051 have provided a ) 160.58 662 T (thorough test of these mechanisms.) 85.54 648 T (\245) 72 628 T 6 F (Flushing advertisements.) 85.54 628 T 0 F ( The procedure for \337ushing old or unreachable advertisements \050the ) 213.47 628 T 6 F (MaxAge ) 85.54 614 T 0 F (procedure\051 has been tested operationally) 131.83 614 T (. It is interesting to note that \337ushing of ) 324.89 614 T -0.14 (advertisements does occur more during interoperability testing \050because of the constant restart-) 85.54 600 P (ing of routers\051 that it does operationally) 85.54 586 T (. For example, in a week of BARRNet statistics, 9650 ) 275.3 586 T (advertisements were \337ushed, while 688,278 new advertisements were \337ooded.) 85.54 572 T (\245) 72 552 T 6 F (Import of external r) 85.54 552 T (outes) 188.24 552 T 0 F (. All options of external LSAs have been tested operationally: ) 214.88 552 T 6 F (type ) 514.35 552 T (1 metrics) 85.54 538 T 0 F (, ) 132.49 538 T 6 F (type 2 metrics) 138.49 538 T 0 F (, ) 210.42 538 T 6 F (forwarding addr) 216.42 538 T (esses) 301.83 538 T 0 F ( and the ) 326.47 538 T 6 F (external r) 367.44 538 T (oute tag) 417.51 538 T 0 F (.) 458.48 538 T (\245) 72 518 T 6 F (Authentication.) 85.54 518 T 0 F ( The OSPF authentication procedure has been tested operationally) 165.16 518 T (.) 481.49 518 T (\245) 72 498 T 6 F (Equal-cost multipath) 85.54 498 T 0 F (. Operational deployments have included topologies with equal-cost, ) 193.82 498 T (redundant paths.) 85.54 484 T (\245) 72 464 T 6 F (Stub ar) 85.54 464 T (eas.) 123.64 464 T 0 F ( These have been deployed both in BARRNet and NSI.) 142.62 464 T 3 F (6.5 Limitations of operational deployments) 72 430.67 T 0 F (The following things have not been tested in an operational environment:) 72 404 T (\245) 72 384 T 6 F (Multi-vendor deployments) 85.54 384 T 0 F (. So far all deployments have used a single implementation. How-) 221.79 384 T (ever) 85.54 370 T (, extensive interoperability testing of OSPF has been done \050see Section 7.0 of this report\051.) 105.7 370 T (\245) 72 350 T 6 F -0.02 (Regular OSPF ar) 85.54 350 P -0.02 (eas.) 174.57 350 P 0 F -0.02 ( These have however been tested in all three rounds of the OSPF inter-) 193.55 350 P (operability testing.) 85.54 336 T (\245) 72 316 T 6 F (V) 85.54 316 T (irtual links.) 93.75 316 T 0 F ( These have however been tested in OSPF\325) 153.07 316 T (s interoperability testing.) 358.59 316 T (\245) 72 296 T 6 F (Non-br) 85.54 296 T (oadcast networks) 122.63 296 T 0 F (. However) 211.57 296 T (, OSPF interoperability testing has been performed over ) 261.05 296 T (X.25 networks.) 85.54 282 T (\245) 72 262 T 6 F (T) 85.54 262 T (OS r) 93.32 262 T (outing.) 117.43 262 T 0 F ( However) 153.08 262 T (, this has been tested in OSPF\325) 199.56 262 T (s interoperability testing.) 345.82 262 T 3 F (6.6 Conclusions) 72 228.67 T 0 F -0.4 (All basic features of the OSPF protocol have been exercised. V) 72 202 P -0.4 (ery lar) 370.08 202 P -0.4 (ge OSPF link state databases ) 400.43 202 P (\050e.g., BARRNet\325) 72 188 T (s OSPF system\051 have been deployed, providing a thorough test of OSPF\325) 152.62 188 T (s data-) 502.42 188 T (base synchronization mechanisms. No OSPF protocol problems have been found in operational ) 72 174 T (deployments.) 72 160 T FMENDPAGE %%EndPage: "21" 22 %%Page: "22" 22 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 22]) 493.7 73 T 72 108 540 684 R 7 X V 0 X -0 (Most of the hassles in operation deployments has to do with the OSPF/RIP interchange. Many of ) 72 676 P (these issues have been ironed out on the ospf-tests mailing list \050see Section 2.0\051. However) 72 662 T (, the ) 503.22 662 T (interaction between OSPF) 72 648 T (, RIP) 196.96 648 T (, and EGP continues to be an active area of research.) 220.29 648 T FMENDPAGE %%EndPage: "22" 23 %%Page: "23" 23 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 23]) 493.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (7.0 Inter) 72 673.33 T (operability T) 138.33 673.33 T (esting) 226.15 673.33 T 0 F (There have been three separate OSPF V2 interoperability testing sessions. Five separate imple-) 72 646 T (mentations have participated in at least one session: implementations from the companies 3com, ) 72 632 T (ACC, Proteon and W) 72 618 T (ell\337eet, and the publicly available implementation from the University of ) 173.64 618 T (Maryland. ) 72 604 T -0.35 (Each of the testing sessions is described in a succeeding section. For each session, the participants ) 72 578 P (are identi\336ed, and the testing topologies are described along with the particular protocol features ) 72 564 T (that were exercised. Any protocol problems that were encountered during the testing are also ) 72 550 T (described. In addition, for the second and third rounds testing reports were sent to the ospf mail-) 72 536 T (ing lists. These reports are reproduced in this document.) 72 522 T (There is quite a bit of commonality in the features that have been tested from session to session. ) 72 496 T (There are several reasons for this commonality) 72 482 T (. First, in each testing session an attempt has been ) 296.37 482 T (made to increase the size of the OSPF system under test. For example, the number of external ) 72 468 T (routes imported has doubled each session. Secondly) 72 454 T (, the interoperability sessions have been ) 320.72 454 T -0.29 (debugging sessions as well as protocol sessions. Many things tested in the third round were to ver-) 72 440 P -0.02 (ify that implementations had successfully \336xed problems found in earlier sessions. A brief over-) 72 426 P (view of the testing session is presented in the following table:) 72 412 T 72 376.01 540 384 C 72 381.98 540 381.98 2 L 0.5 H 0 Z 0 X 0 K N 0 0 612 792 C 5 F 0 X 0 K (T) 72 387.33 T (ABLE 4. OSPF inter) 77.93 387.33 T (operability testing at a glance.) 166.04 387.33 T (Site) 72 365.34 T (Date) 144 365.34 T (# Routers) 225 365.34 T (# Externals) 288 365.34 T (Implementations) 360 365.34 T 1 F (Proteon) 72 348.34 T (9/25/90-9/29/90) 144 348.34 T (6) 225 348.34 T (20-30) 288 348.34 T (3com, Proteon, W) 360 348.34 T (ell\337eet) 431.93 348.34 T (SURAnet) 72 332.34 T (12/17/90-12/21/90) 144 332.34 T (10) 225 332.34 T (96) 288 332.34 T (3com, ACC, Proteon, W) 360 332.34 T (ell\337eet) 457.48 332.34 T (3com) 72 316.34 T (2/4/91-2/8/91) 144 316.34 T (16) 225 316.34 T (400) 288 316.34 T (3com, ACC, Proteon, W) 360 316.34 T (ell\337eet, UMD) 457.48 316.34 T 0 F -0 (For more information on the interoperability testing, the following people can be contacted: Fred ) 72 291.01 P (Baker [fbaker], Rob Coltun [rcoltun], Dino Farinacci [dino], Jonathan Hsu [jhsu], John Moy ) 72 277.01 T ([jmoy], and W) 72 263.01 T (illiam Streilein [bstreile].) 141.8 263.01 T 3 F (7.1 T) 72 229.67 T (esting methodology) 104.53 229.67 T 0 F -0.07 (In the interoperability tests, the routers have been interconnected using ethernet, serial lines \050PPP ) 72 203.01 P (and proprietary\051, X.25 and 802.5 token ring. Monitoring of the routers has been done through ) 72 189.01 T -0.48 (connecting terminals \050either directly or via telnet\051 to the router consoles. Each implementation has ) 72 175.01 P -0.37 (a dif) 72 161.01 P -0.37 (ferent user interface, which makes monitoring somewhat dif) 93.06 161.01 P -0.37 (\336cult. As explained earlier in this ) 379.34 161.01 P (document, there is now an OSPF MIB, which in the future will enable a common monitoring ) 72 147.01 T (interface to all implementations.) 72 133.01 T FMENDPAGE %%EndPage: "23" 24 %%Page: "24" 24 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 24]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (In general, each implementation has an error logging capability) 72 676 T (, and this is often how problems ) 375.33 676 T -0.28 (are discovered. LAN protocol analyzers are also used to capture OSPF protocol packet exchanges ) 72 662 P -0.04 (that are causing problems. These packet traces are available for analysis either during of after the ) 72 648 P (testing sessions.) 72 634 T (V) 72 608 T (eri\336cation of routing was done through visual inspection of implementations\325 routing table and ) 79.33 608 T (link state databases \050via the console interface\051, and through network debugging tools such as ) 72 594 T (\322ping\323 and \322traceroute\323.) 72 580 T 3 F (7.2 First r) 72 546.67 T (ound \050Pr) 134.7 546.67 T (oteon, 9/25/90 - 9/29/90\051) 187.7 546.67 T 0 F (The \336rst round of OSPF protocol testing took place at Proteon Inc.\325) 72 520 T (s W) 395.14 520 T (estborough facility) 413.16 520 T (, the ) 502.66 520 T (week of September 25, 1990. Three implementations participated, from the vendors 3com, Pro-) 72 506 T (teon and W) 72 492 T (ell\337eet.) 126.33 492 T -0.11 (There were two 3com routers, two W) 72 466 P -0.11 (ell\337eet routers and two Proteon routers available for testing. ) 249.88 466 P (These routers were interconnected with ethernets and serial lines. External routes were imported ) 72 452 T -0.3 (from the Proteon company internet. In addition, during of) 72 438 P -0.3 (f hours we were able to connect the rout-) 345.88 438 P (ers under test to the Proteon company internet, forming one fairly lar) 72 424 T (ge OSPF system.) 402.21 424 T (The testing at Proteon proceeded as follows:) 72 398 T (\245) 72 378 T (All routers were connected to a single ethernet. Then, as routers were taken up and down, the ) 85.54 378 T (Designated Router election algorithm and the Database Description process were tested. Also ) 85.54 364 T (OSPF\325) 85.54 350 T (s reliable \337ooding algorithm was tested in this con\336guration.) 117.53 350 T (\245) 72 330 T (T) 85.54 330 T (wenty to thirty external routes were imported into the OSPF system by a Proteon router ) 92.02 330 T (\050which was simultaneously running RIP\051. It was then veri\336ed that these external routes were ) 85.54 316 T (installed into the router) 85.54 302 T (\325) 197.58 302 T (s routing tables.) 200.91 302 T (\245) 72 282 T (One of the 3com routers was con\336gured to originate an OSPF default route. This tested OSPF ) 85.54 282 T -0.3 (default route processing, and also tested the behavior of the system when multiple routers were ) 85.54 268 P (importing external routes.) 85.54 254 T (\245) 72 234 T -0.35 (The OSPF system was split into areas. Both regular OSPF areas \050non-stub\051 and stub areas were ) 85.54 234 P (tested.) 85.54 220 T (\245) 72 200 T (The six routers under test were connected to the Proteon company internet \050which was also ) 85.54 200 T (running OSPF\051, forming an OSPF system of eighteen routers. This con\336guration was short-) 85.54 186 T (lived, due to a disagreement between the 3com and Proteon routers concerning how to repre-) 85.54 172 T (sent an OSPF default route.) 85.54 158 T (Unfortunately) 72 132 T (, incomplete records were kept of this testing, so that no maps of the testing con\336g-) 138.5 132 T (urations can be reproduced for this document.) 72 118 T FMENDPAGE %%EndPage: "24" 25 %%Page: "25" 25 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 25]) 493.7 73 T 72 108 540 684 R 7 X V 6 F 0 X (7.2.1 Pr) 72 676 T (oblems found in the First r) 114.42 676 T (ound testing) 251.13 676 T 0 F (A couple of OSPF protocol/speci\336cation problems were uncovered in the \336rst round of testing. ) 72 650 T -0.02 (First, it was noticed that there was a window in the Database Description process where concur-) 72 636 P -0.07 (rently \337ooded MaxAge advertisements could prevent database synchronization from completing. ) 72 622 P (This required a change in the speci\336cation\325) 72 608 T (s handling of MaxAge LSAs.) 278.51 608 T (Secondly) 72 582 T (, it was found that the OSPF speci\336cation did not specify how the Network Mask \336eld ) 115.86 582 T (should be set in external LSAs that were advertising the DefaultDestination. This was a minor ) 72 568 T (problem, but caused dif) 72 554 T (\336culties because of assumptions made in one implementation on how the ) 185.04 554 T (mask should be set.) 72 540 T 3 F (7.3 Second r) 72 506.67 T (ound \050SURAnet, 12/17/90 - 12/21/90\051) 148.71 506.67 T 0 F (The second round of OSPF protocol testing took place at SURAnet\325) 72 480 T (s College Park facility) 397.8 480 T (, the ) 503.95 480 T (week of December 12, 1990. Four implementations participated, from the vendors 3com, ACC, ) 72 466 T (Proteon and W) 72 452 T (ell\337eet.) 142.99 452 T (There were two 3com routers, two ACC routers, two W) 72 426 T (ell\337eet routers and four Proteon routers ) 338.51 426 T (available for testing. These routers were interconnected with ethernets, serial lines and token ) 72 412 T (rings. External routes were imported from SURAnet by one of the Proteon routers.) 72 398 T (The testing at SURAnet proceeded as follows. Initially nine routers were con\336gured as a single ) 72 372 T (backbone area, with six of the routers connected to a single ethernet. This is pictured in Figure 4. ) 72 358 T (In this con\336guration, the ) 72 344 T 6 F (Designated Router transition ) 192.93 344 T 0 F (and ) 344.5 344 T 6 F (database synchr) 364.82 344 T (onization) 447.56 344 T 0 F ( process ) 494.88 344 T (were tested. Ninety-six external routes were imported from SURAnet to stress the \337ooding algo-) 72 330 T -0.36 (rithm. By restarting the router that was importing the routes, the) 72 316 P 6 F -0.36 ( \337ushing of advertisements) 375.55 316 P 0 F -0.36 ( from ) 511.4 316 P (the routing domain was tested. Additionally) 72 302 T (, ) 281.77 302 T 6 F (variable-length subnets) 287.76 302 T 0 F ( and OSPF\325) 407.37 302 T (s optional ) 462.68 302 T 6 F (T) 512.66 302 T (OS ) 520.45 302 T (r) 72 288 T (outing) 77.11 288 T 0 F ( capability were tested in this con\336guration.) 109.77 288 T (Next the routers were con\336gured into four separate ) 72 262 T 6 F (OSPF ar) 319.8 262 T (eas) 364.55 262 T 0 F (, with each area directly con-) 380.54 262 T (nected to the OSPF backbone \050which was a single ethernet\051. There were no virtual links in this ) 72 248 T (con\336guration. Inter) 72 234 T (-area routing was tested, including having AS boundary routers internal to a ) 164.36 234 T -0.09 (non-backbone area. Also tested was the case where a single router was both an area border router ) 72 220 P (and an AS boundary router) 72 206 T (.) 201.24 206 T -0.16 (For more details of the testing, see the \322Of) 72 180 P -0.16 (\336cial report of the Second Round T) 274.71 180 P -0.16 (esting\323 listed below) 442.82 180 P -0.16 (.) 537 180 P FMENDPAGE %%EndPage: "25" 26 %%Page: "26" 26 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 26]) 493.7 73 T 72 108 540 684 R 7 X V 6 F 0 X (7.3.1 Of\336cial r) 72 433 T (eport of the Second r) 148.07 433 T (ound testing) 255.12 433 T 4 F (The following r) 72 407 T (eport was sent to the ospf, ospf-tests, and r) 145.53 407 T (outer) 350.32 407 T (-r) 375.4 407 T (equir) 383.61 407 T (ements mailing lists after ) 408.49 407 T (the second r) 72 393 T (ound of inter) 130.18 393 T (operability tests:) 191.71 393 T 0 F -0.37 ( The second round of OSPF multi-vendor testing was done in College Park, Maryland the week of ) 72 367 P (12/17/90. The facilities were provided by SURAnet. Four major router vendors were present, ) 72 353 T -0.09 (Advanced Computer Communications \050ACC\051, Proteon, 3Com, and W) 72 339 P -0.09 (ell\337eet. A press conference ) 406.83 339 P (and presentation was provided for 3 dif) 72 325 T (ferent data communication magazine representatives.) 260.31 325 T -0.13 ( Each vendor provided at least 2 routers. Each vendor had a device connected to a common Ether-) 72 299 P (net. This Ethernet was con\336gured as the OSPF backbone area. The rest of the routers were ) 72 285 T (attached to the various backbone routers via Ethernet, T) 72 271 T (oken Ring, proprietary serial line, PPP ) 339.29 271 T (serial line, and X.25 type media. The following test scenarios were performed and completed in ) 72 257 T (the following order:) 72 243 T (\245) 72 223 T (Intra-area routing. All routers were con\336gured to reside in the backbone area. Designated ) 85.54 223 T (Router election was performed various number of times so each vendor could be designated ) 85.54 209 T (router for a period of time. Packet data was captured on a Snif) 85.54 195 T (fer for later observation.) 383.42 195 T (\245) 72 175 T -0.3 (V) 85.54 175 P -0.3 (ariable Length Subnet Masks. Some of the serial lines in the con\336guration were con\336gured to ) 92.86 175 P (be on the same IP network but with dif) 85.54 161 T (ferent subnet masks. It was observed that all routers ) 271.87 161 T (stored routes for the dif) 85.54 147 T (ferent length subnets.) 197.9 147 T 72 108 540 684 C 72 441 540 684 C 70.88 441 541.12 684 R 7 X 0 K V 200.71 602.17 227.71 629.17 R V 1 H 2 Z 0 X N 0 F (ACC) 212.05 548.33 T 263.71 647.17 290.71 674.17 R 7 X V 0 X N 137.71 494.17 164.71 521.17 R 7 X V 0 X N 208.71 494.17 236.71 521.17 R 7 X V 0 X N 209.71 539.17 236.71 566.17 R 7 X V 0 X N 137.71 539.17 164.71 566.17 R 7 X V 0 X N 79.88 585 502.88 585 2 L 4 H N 137.71 638.17 326.71 638.17 2 L N 375.21 540 402.21 567 R 7 X V 1 H 0 X N 438.21 540 465.21 567 R 7 X V 0 X N 375.21 603 402.21 630 R 7 X V 0 X N (WF) 379.55 611.16 T (WF) 380.55 550.16 T (P) 447.55 550.82 T (P) 274.05 658.99 T (P) 211.05 612.99 T (3C) 143.05 548.85 T (3C) 145.05 505.86 T (ACC) 211.83 551.17 T (ACC) 210.83 503.17 T 388.33 567.33 388.33 584 2 L N 451.67 567.33 451.67 584.83 2 L N 388.33 603.17 388.33 584.83 2 L N 276.67 647.33 276.67 639 2 L N 214.17 629 214.17 638.17 2 L N 222.5 521.5 222.5 539 2 L N 152.5 521.5 151.67 539 2 L N 224.17 566.5 224.17 584 2 L N 150.83 566.5 150.83 584 2 L N 6 F (Figur) 164.17 458.17 T (e 4: Initial con\336guration for the Second r) 192.6 458.17 T (ound testing) 401.94 458.17 T 214.33 602 214.33 587 2 L N 72 108 540 684 C 0 0 612 792 C FMENDPAGE %%EndPage: "26" 27 %%Page: "27" 27 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 27]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T (Import SURAnet routes. The Proteon router was listening for RIP routes generated by the ) 85.54 676 T -0.37 (SURAnet routers. These routes were imported into our OSPF test system. 96 external link state ) 85.54 662 P -0.02 (advertisements were generated as a result. Many scaling type implementation problems sur-) 85.54 648 P (faced for each vendor during this exercise. ) 85.54 634 T (\245) 72 614 T -0.28 (T) 85.54 614 P -0.28 (ype of Service generation. While the test setup was still con\336gured as a single area, the 3Com ) 92.02 614 P (router generated T) 85.54 600 T (ype of Service link state advertisements. It was observed how the other ven-) 173.28 600 T (dor implementations reacted to it. Some problems were found. ) 85.54 586 T (\245) 72 566 T (Inter) 85.54 566 T (-area routing. Multiple areas were con\336gured. Common non-backbone areas were shared ) 107.94 566 T (by Proteon and W) 85.54 552 T (ell\337eet and by ACC and 3Com. It was observed that the correct Intra-area ) 171.52 552 T (and Inter) 85.54 538 T (-area routes appeared in each router) 128.26 538 T (\325) 300.19 538 T (s routing table. At this point in the test setup, the ) 303.53 538 T (Proteon router regenerated the 96 SURAnet routes into the con\336guration. It was observed that ) 85.54 524 T (the routes were learned and propagated over area boundaries. Some problems occur at this ) 85.54 510 T (point. More emphasis on this scenario will occur at the next round of testing.) 85.54 496 T (\245) 72 476 T (OSPF over X.25. A point-to-point link was connected between the Proteon router and the ) 85.54 476 T (3Com router) 85.54 462 T (. The X.25 packet level was con\336gured to run over the link. OSPF was enabled ) 145.84 462 T -0.2 (over the link to verify that multi-vendor OSPF over X.25 was performed. Both of these routers ) 85.54 448 P (were in the same area.) 85.54 434 T (\245) 72 414 T (MaxAge advertisements. Link state advertisements were \337ushed from the routing domain ) 85.54 414 T (using the MaxAge procedure. W) 85.54 400 T (e veri\336ed that all routers removed the advertisements from ) 241.45 400 T (their databases, after they were properly acknowledged by the \337ooding procedure. Some prob-) 85.54 386 T (lems were found in this test, although not nearly as many as in the \336rst round of testing.) 85.54 372 T -0.41 ( Each vendor agreed that this sort of testing was extremely valuable and that it should occur again. ) 72 346 P (3Com has of) 72 332 T (fered for the third round of testing to occur in Santa Clara sometime in February) 133.08 332 T (. ) 517.36 332 T -0 (3Com will encourage other OSPF implementations to join in the testing. Items that will be tested ) 72 318 P (are: ) 72 304 T (\245) 72 284 T (Intra-area routing with loops \050as well as equal-cost multipath\051.) 85.54 284 T (\245) 72 264 T (Inter) 85.54 264 T (-area testing including Stub and T) 107.94 264 T (ransit area support, with both Intra-area and Inter) 270.41 264 T (-area ) 505.98 264 T (loops.) 85.54 250 T (\245) 72 230 T (V) 85.54 230 T (irtual link testing in the looped Inter) 93.48 230 T (-area con\336guration.) 266.47 230 T (\245) 72 210 T (RIP/OSPF route interchange including testing forwarding address capability in external link ) 85.54 210 T (state advertisements.) 85.54 196 T (\245) 72 176 T (EGP/OSPF router interchange including the use of the route tag \336eld in external link state ) 85.54 176 T (advertisements.) 85.54 162 T (\245) 72 142 T -0.22 (More than two routers connected to an X.25 network. W) 85.54 142 P -0.22 (e would like to test OSPF\325) 354.08 142 P -0.22 (s non-broad-) 479.27 142 P (cast multi-access capabilities by attaching more than two vendor) 85.54 128 T (\325) 395.76 128 T (s routers to an X.25 packet ) 399.09 128 T (switch.) 85.54 114 T FMENDPAGE %%EndPage: "27" 28 %%Page: "28" 28 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 28]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T (Several vendors running OSPF and RIP simultaneously) 85.54 676 T (. This will further test the OSPF/RIP ) 351.27 676 T (interactions.) 85.54 662 T (\245) 72 642 T (T) 85.54 642 T (est processing of links with cost LSIn\336nity) 92.02 642 T (. These links should be treated as unreachable.) 297.14 642 T -0.14 ( Furthermore, we hope that in future rounds of testing an OSPF MIB would allow us to also use a ) 72 616 P (network management station to gather test data.) 72 602 T -0.14 ( In summary) 72 576 P -0.14 (, the stability of implementations were better this time more so than the \336rst round of ) 131.56 576 P (testing. No problems with the protocol design were encountered. The exchange of ideas and the ) 72 562 T (cooperation among implementors that occurred during this test ef) 72 548 T (fort, continues the spirit that ) 385.57 548 T (OSPF is truly an open protocol. ) 72 534 T 6 F (7.3.2 Pr) 72 502 T (oblems found in the Second r) 114.42 502 T (ound testing) 263.14 502 T 0 F (No problems were found in the OSPF protocol during the second round of testing.) 72 476 T 3 F (7.4 Third r) 72 442.67 T (ound \0503com, 2/4/91 - 2/8/91\051) 140.94 442.67 T 0 F (The third round of OSPF protocol testing took place at 3com\325) 72 416 T (s Santa Clara facility) 367.16 416 T (, the week of ) 466.64 416 T (February 4, 1991. Five implementations participated, from the vendors 3com, ACC, Proteon and ) 72 402 T (W) 72 388 T (ell\337eet and the publicly available University of Maryland implementation \050running on a SUN ) 82.36 388 T (workstation\051.) 72 374 T -0.35 (There were \336ve 3com routers, four ACC routers, three W) 72 348 P -0.35 (ell\337eet routers, three Proteon routers and ) 343.28 348 P -0.2 (the UMD Sun workstation available for testing \050giving a total of 16 routers available\051. These rout-) 72 334 P -0.14 (ers were interconnected with ethernets, serial lines and X.25. External routes were imported from ) 72 320 P (BARRNet by one of the 3com routers.) 72 306 T (The initial testing con\336guration is shown in Figure 5. Three areas were con\336gured, along with a ) 72 280 T -0.12 (non-contiguous backbone. The backbone was then joined by con\336guring two virtual links. In this ) 72 266 P (con\336guration the following OSPF functionality was tested: ) 72 252 T 6 F (inter) 358.17 252 T (-ar) 382.38 252 T (ea r) 397.48 252 T (outing) 416.91 252 T 0 F ( and ) 449.56 252 T 6 F (virtual links) 472.88 252 T 0 F (.) 535.2 252 T -0.01 (The system was then recon\336gured so that twelve of the routers were connected to a single ether-) 72 226 P (net. This con\336guration is pictured in Figure 6. By bringing routers up and down, this con\336gura-) 72 212 T -0.37 (tion tested) 72 198 P 6 F -0.37 ( Designated Router election) 121.27 198 P 0 F -0.37 (, ) 261.05 198 P 6 F -0.37 (database synchr) 266.68 198 P -0.37 (onization) 349.05 198 P 0 F -0.37 ( and ) 396.36 198 P 6 F -0.37 (r) 418.94 198 P -0.37 (eliable \337ooding) 424.05 198 P 0 F -0.37 (. T) 501.32 198 P -0.37 (o see ) 513.43 198 P (how this functionality) 72 184 T (, and also the implementations, scale, 400 external routes were imported ) 176.5 184 T (from BARRNet.) 72 170 T FMENDPAGE %%EndPage: "28" 29 %%Page: "29" 29 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 29]) 493.7 73 T 72 108 540 684 R 7 X V 6 F 0 X (7.4.1 Of\336cial r) 72 266.03 T (eport of the Third r) 148.07 266.03 T (ound testing) 248.46 266.03 T 4 F (The following r) 72 240.03 T (eport was sent to the ospf, ospf-tests, and r) 145.53 240.03 T (outer) 350.32 240.03 T (-r) 375.4 240.03 T (equir) 383.61 240.03 T (ements mailing lists after ) 408.49 240.03 T (the thir) 72 226.03 T (d r) 106.54 226.03 T (ound of inter) 119.76 226.03 T (operability tests:) 181.28 226.03 T 0 F -0.19 (The third round of OSPF interoperability testing was held at 3com Corporation in Santa Clara the ) 72 200.03 P (week of February 4-8. Four router vendors came to the testing: 3com, ACC, Proteon and W) 72 186.03 T (ell-) 510.73 186.03 T -0.29 (\337eet. In addition, Rob Coltun brought the University of Maryland implementation, which was run ) 72 172.03 P (on a Sun W) 72 158.03 T (orkstation.) 127.34 158.03 T 72 108 540 684 C 72 274.03 540 684 C 70.88 274.03 541.12 684 R 7 X 0 K V 90.01 339 226.01 428 R 3 X V 1 H 2 Z 7 X N 376.88 416 529.88 578 R 3 X V 7 X N 196.88 497 331.88 551 R 3 X V 7 X N 106.88 650 322.88 650 2 L 4 H 0 X N 103.88 596 517.88 596 2 L N 385.88 542 520.88 542 2 L N 502.88 524 502.88 434 2 L N 457.88 497 484.88 524 R 7 X V 1 H 0 X N 457.88 443 484.88 470 R 7 X V 0 X N 403.88 497 430.88 524 R 7 X V 0 X N 403.88 560 430.88 587 R 7 X V 0 X N 457.88 560 484.88 587 R 7 X V 0 X N 331.88 614 358.88 641 R 7 X V 0 X N 196.88 614 223.88 641 R 7 X V 0 X N 277.88 533 304.88 560 R 7 X V 0 X N 133.88 560 160.88 587 R 7 X V 0 X N 97.88 506 124.88 533 R 7 X V 0 X N 160.88 506 187.88 533 R 7 X V 0 X N 214.88 533 241.88 560 R 7 X V 0 X N 210.88 614 210.88 596 2 L N 211.01 641 212.01 651 2 L N 345.01 614 345.01 598 2 L N 293.01 560 293.01 597 2 L N 418.01 587 419.01 596 2 L N 473.01 587 473.01 596 2 L N 474.01 560 474.01 544 2 L N 419.01 560 419.01 544 2 L N 419.01 542 419.01 524 2 L N 474.01 542 474.01 525 2 L N 485.01 511 502.01 511 2 L N 485.01 458 503.01 458 2 L N 90 450 44 40.5 312.88 411.5 A 212.01 515 322.01 515 2 L 4 H N 292.01 533 292.01 517 2 L 1 H N 229.01 533 230.01 517 2 L N 151.88 587 151.88 596 2 L N 150.01 559 174.01 533 2 L N 142.01 561 110.01 533 2 L N 110.01 505 110.01 481 2 L N 188.01 519 215.01 545 2 L N 460.01 498 344.01 440 2 L N 458.01 455 355.01 425 2 L N 251.01 478 252.01 339 2 L 4 H N 98.01 395 212.01 395 2 L N 472.01 443 471.01 429 471.01 351 253.01 353 4 L 1 H N 205.88 452 232.88 479 R 7 X V 0 X N 138.88 410 165.88 437 R 7 X V 0 X N 110.88 352 137.88 380 R 7 X V 0 X N 167.88 353 194.88 380 R 7 X V 0 X N 233.01 466 252.01 466 2 L N 153.01 410 153.01 395 2 L N 180.01 379 180.01 395 2 L N 124.01 380 124.01 395 2 L N 195.01 365 252.01 365 2 L N 0 F (X.25) 299.01 408 T (WF) 210.01 462 T (WF) 462.01 569 T (WF) 282.01 542 T (ACC) 136.01 569 T (ACC) 163.01 516 T (ACC) 100.01 516 T (ACC) 217.01 542 T 84.01 481 133.01 481 2 L 4 H N (3C) 203.01 624 T (3C) 408.01 570 T (3C) 412.01 504 T (3C) 463.01 505 T (3C) 175.01 361 T (P) 467.01 454 T (P) 121.01 364 T (P) 341.01 625 T (UMD) 140.01 418 T (BARRNet) 180.01 664 T 130.48 665.49 119.01 669 130.6 672.1 130.54 668.79 4 Y V 175.01 668 129.54 668.81 2 L 0.5 H N 442.75 550.79 434 559 445.77 556.68 444.26 553.73 4 Y V 493.97 478.97 485.99 470.01 488.01 481.84 490.99 480.4 4 Y V 490.99 480.41 M 498.6 499.44 500.8 523.73 486.01 538.1 D 474.28 549.49 458.36 548.76 444.3 553.76 D 1 H 0 Z N 495.62 555.84 486 563 497.95 562.02 496.79 558.93 4 Y V 497.71 464.62 486 462.01 494.54 470.43 496.12 467.53 4 Y V 496.18 467.46 M 522.4 486.33 524.46 531.39 503.42 554 D 501.42 556.15 499.16 557.72 496.81 558.95 D N (VL) 505.01 551 T (VL) 482.01 526 T 6 F (Figur) 187.01 302 T (e 5: Third r) 215.45 302 T (ound testing: Initial con\336guration) 274.86 302 T 72 108 540 684 C 70.88 274.03 541.12 684 R 4 H 0 Z 7 X 0 K N 0 0 612 792 C FMENDPAGE %%EndPage: "29" 30 %%Page: "30" 30 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 30]) 493.7 73 T 72 108 540 684 R 7 X V 0 X -0.37 (T) 72 363.3 P -0.37 (esting was performed over ethernet, point-to-point networks \050using PPP\051 and X.25. In all we had ) 78.49 363.3 P (16 routers available: \336ve 3com routers, four ACC routers, three Proteon routers, three W) 72 349.3 T (ell\337eet ) 496.72 349.3 T (routers and Rob\325) 72 335.3 T (s SUN. W) 151.95 335.3 T (e also were able to import external routes from BARRNet.) 199.96 335.3 T (Speci\336c tests performed included the following:) 72 309.3 T (\245) 72 289.3 T (Initially we con\336gured the routers into three separate ) 85.54 289.3 T ( areas and a physically disconnected ) 342.67 289.3 T (backbone. The ) 85.54 275.3 T -0.1 ( backbone was then reconnected through con\336guration of ) 159.14 275.3 P -0.1 ( several virtual links. ) 436.8 275.3 P (These tests veri\336ed the ) 85.54 261.3 T -0.24 ( generation and processing of summary link advertisements, ) 198.79 261.3 P -0.24 ( as well as ) 488.33 261.3 P (the operation of virtual links.) 85.54 247.3 T (\245) 72 227.3 T (W) 85.54 227.3 T (e connected multiple routers to an X.25 packet switch, ) 95.9 227.3 T ( testing OSPF\325) 360.04 227.3 T (s non-broadcast net-) 430.02 227.3 T (work capability) 85.54 213.3 T (. OSPF ) 159.7 213.3 T ( was successfully run over the X.25 network, using routers ) 197.36 213.3 T ( that were ) 481.82 213.3 T (both DR eligible and DR ineligible. Some ) 85.54 199.3 T ( problems were encountered, but they all involved ) 290.1 199.3 T (running ) 85.54 185.3 T ( IP over X.25 \050i.e., they were not X.25 speci\336c\051.) 125.85 185.3 T (\245) 72 165.3 T (W) 85.54 165.3 T (e also connected a 3com router) 95.9 165.3 T (, Proteon router) 244.29 165.3 T (, and Rob\325) 318.76 165.3 T (s ) 368.4 165.3 T ( SUN to an ethernet, and then ) 376.06 165.3 T (treated the ethernet as ) 85.54 151.3 T ( a non-broadcast network. This allowed us to connect Rob\325) 193.44 151.3 T (s ) 475.91 151.3 T ( SUN into ) 483.57 151.3 T (the rest of the routing domain without installing ) 85.54 137.3 T -0.49 ( the IP multicast modi\336cations to the SUN ker-) 318.08 137.3 P (nel. It ) 85.54 123.3 T ( also further tested the OSPF\325) 116.51 123.3 T (s non-broadcast network ) 258.1 123.3 T ( capability) 379 123.3 T (.) 428.52 123.3 T 72 108 540 684 C 72 371.3 540 684 C 70.99 371.3 541.01 684 R 7 X 0 K V 105.44 654.71 321.44 654.71 2 L 4 H 2 Z 0 X N 106.44 600.71 520.44 600.71 2 L N 394.99 558 421.99 585 R 7 X V 1 H 0 X N 232.99 558 259.99 585 R 7 X V 0 X N 330.44 618.71 357.44 645.71 R 7 X V 0 X N 195.44 618.71 222.44 645.71 R 7 X V 0 X N 106.99 558 133.99 585 R 7 X V 0 X N 209.44 618.71 209.44 600.71 2 L N 209.56 645.71 210.56 655.71 2 L N 343.56 618.71 343.56 602.71 2 L N 0 F (3C) 201.56 628.71 T (P) 339.56 629.71 T (BARRNet) 178.56 668.71 T 129.03 670.2 117.56 673.71 129.15 676.81 129.09 673.51 4 Y V 173.56 672.71 128.09 673.52 2 L 0.5 H N 106.99 612 133.99 639 R 7 X V 1 H 0 X N 142.99 612 169.99 639 R 7 X V 0 X N 142.99 558 169.99 585 R 7 X V 0 X N 277.99 558 304.99 585 R 7 X V 0 X N 90 450 44 40.5 285.99 472.5 A (X.25) 272.12 469 T 214.99 540 322.99 540 2 L 4 H N 340.99 558 367.99 585 R 7 X V 1 H 0 X N 439.99 558 466.99 585 R 7 X V 0 X N 484.99 558 511.99 585 R 7 X V 0 X N 385.99 540 475.99 540 2 L 4 H N 421.99 504 520.99 504 2 L N 439.99 459 466.99 486 R 7 X V 1 H 0 X N 493.99 459 520.99 486 R 7 X V 0 X N 376.99 495 403.99 522 R 7 X V 0 X N 394.99 621 421.99 648 R 7 X V 0 X N 124.99 612 124.99 603 2 L N 158.33 611.67 158.33 602.5 2 L N 121.66 585 121.66 600 2 L N 156.66 584.17 156.66 600.83 2 L N 246.66 585 246.66 601.67 2 L N 293.33 585 293.33 600 2 L N 246.66 558.33 246.66 540 2 L N 293.33 558.33 293.33 541.67 2 L N 354.99 585 354.99 600 2 L N 353.33 558.33 314.99 503.33 2 L N 439.99 470.83 329.99 472.5 2 L N 453.33 485.83 453.33 503.33 2 L N 508.33 485.83 508.33 504.17 2 L N 498.33 558.33 498.33 505.83 2 L N 389.99 521.67 389.99 540.83 2 L N 409.99 558.33 409.99 540 2 L N 454.99 558.33 454.99 539.17 2 L N 408.33 585 408.33 600 2 L N 454.16 585 454.16 600.83 2 L N 498.33 585 498.33 600.83 2 L N 407.49 620.83 407.49 601.67 2 L N (ACC) 144.16 622.5 T (ACC) 107.99 568.67 T (ACC) 144.33 565.67 T (ACC) 108.16 622.67 T (3C) 445.83 568.33 T (3C) 490.49 567.83 T (3C) 446.83 469 T (3C) 500.66 469.33 T (WF) 399.99 630.83 T (WF) 237.16 567.83 T (WF) 282.66 568.17 T (P) 349.99 569.17 T (P) 403.83 567.83 T (UMD) 377.49 505 T 6 F (Figur) 176.99 401.67 T (e 6: Third r) 205.43 401.67 T (ound testing: second con\336guration) 264.84 401.67 T 72 108 540 684 C 0 0 612 792 C FMENDPAGE %%EndPage: "30" 31 %%Page: "31" 31 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 31]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T -0.03 (W) 85.54 676 P -0.03 (e then recon\336gured the OSPF system so that all but ) 95.9 676 P -0.03 ( three of the routers were connected to a ) 344.78 676 P (single ethernet. ) 85.54 662 T ( This tested the Designated Router functionality \05012 ) 161.81 662 T ( routers were synchroniz-) 413.65 662 T (ing with the DR\051. W) 85.54 648 T (e then also ) 182.85 648 T -0.03 ( tested the DR election algorithm, by selectively ) 237.15 648 P -0.03 ( restarting the ) 471.12 648 P (DR, or sometimes both the DR and the ) 85.54 634 T ( Backup DR. This also tested OSPF\325) 275.09 634 T (s Database ) 450 634 T (Description ) 85.54 620 T ( process.) 144.5 620 T (\245) 72 600 T (In this con\336guration, we then imported 400 external ) 85.54 600 T ( routes from BARRNet \050one of the 3com ) 338.37 600 T (routers ran both ) 85.54 586 T ( OSPF and EGP\051. Some problems were encountered in ) 164.48 586 T ( implementations\325 ) 430.3 586 T (buf) 85.54 572 T (fer allocation strategies, and ) 101.31 572 T ( problems were also found in the way implementations ) 239.53 572 T ( avoid ) 505.69 572 T (IP fragmentation. But overall, this system was ) 85.54 558 T ( fairly stable.) 310.4 558 T (The following problems we found in the OSPF speci\336cation:) 72 532 T (\245) 72 512 T (The speci\336cation should say that the \322Network mask\323 \336eld ) 85.54 512 T ( should not be veri\336ed in OSPF ) 371.67 512 T (Hellos received over virtual ) 85.54 498 T ( links.) 222.09 498 T (\245) 72 478 T -0.24 (The speci\336cation should say that on multi-access networks ) 85.54 478 P -0.24 ( neighbors are identi\336ed by their IP ) 368.78 478 P (address, and on ) 85.54 464 T -0.15 ( point-to-point networks and virtual links by their OSPF ) 162.81 464 P -0.15 ( Router ID. This elim-) 433.66 464 P (inates confusion when, for example, ) 85.54 450 T ( a router is restarted and comes up with the same IP ) 262.4 450 T (address ) 85.54 436 T ( but a dif) 124.5 436 T (ferent Router ID.) 167.26 436 T -0.08 (Thanks to 3com for providing the testing facility) 72 410 P -0.08 (, cables. terminals, X.25 switch. etc. Also thanks ) 304.16 410 P (to V) 72 396 T (ince Fuller of BARRNet for helping us import the BARRNet routes. ) 92.27 396 T 6 F (7.4.2 Pr) 72 364 T (oblems found in the Third r) 114.42 364 T (ound testing) 256.48 364 T 0 F -0.05 (A couple of speci\336cation/protocol problems were found in the third round of interoperability test-) 72 338 P (ing. First, it was noticed that the speci\336cation did not specify the setting of the Network Mask ) 72 324 T (\336eld in Hellos sent over virtual links. This caused some initial dif) 72 310 T (\336culty in bringing up virtual ) 385.6 310 T (links between routers belonging to dif) 72 296 T (ferent vendors. Secondly) 254 296 T (, it was noticed that the speci\336ca-) 372.8 296 T (tion was not strict enough in de\336ning how OSPF neighbors are identi\336ed on multi-access net-) 72 282 T -0.2 (works. This caused dif) 72 268 P -0.2 (\336culties in one implementation when another vendor) 179.78 268 P -0.2 (\325) 432.21 268 P -0.2 (s router was restarted ) 435.55 268 P (with the same IP address but a dif) 72 254 T (ferent OSPF Router ID. This is discussed more fully in the ) 234.01 254 T (above \322Of) 72 240 T (\336cial Report of the Third Round T) 121.4 240 T (esting\323.) 285.79 240 T 3 F (7.5 Overall: Featur) 72 206.67 T (es tested) 189.88 206.67 T 0 F (All signi\336cant protocol features and mechanisms have been tested in the three rounds of interop-) 72 180 T (erability testing. In particular) 72 166 T (, the following basic pieces of the protocol have been tested:) 211.42 166 T (\245) 72 146 T 6 F (Designated Router election.) 85.54 146 T 0 F ( W) 226.43 146 T (ith as many as thirteen routers attached to a single LAN, the ) 240.27 146 T (election of Backup Designated Router and Designated router was veri\336ed by bringing routers ) 85.54 132 T (up and down, singly and in pairs.) 85.54 118 T FMENDPAGE %%EndPage: "31" 32 %%Page: "32" 32 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 32]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T 6 F (Adjacency bringup) 85.54 676 T 0 F (. The Database Description process was veri\336ed, with databases having ) 183.82 676 T (over 400 LSAs. Adjacency bringup was also veri\336ed in times when \337ooding was taking place ) 85.54 662 T (simultaneously) 85.54 648 T (.) 157.39 648 T (\245) 72 628 T 6 F (Reliable \337ooding) 85.54 628 T 0 F (. It was veri\336ed that OSPF\325) 171.84 628 T (s \337ooding algorithm maintains database synchro-) 302.44 628 T (nization, both in the presence of loops in the topology) 85.54 614 T (, and with lar) 343.26 614 T (ge databases \050over 400 ) 406.34 614 T (LSAs\051.) 85.54 600 T (\245) 72 580 T 6 F (Flushing advertisements fr) 85.54 580 T (om r) 222.57 580 T (outing domain) 246.67 580 T 0 F (. OSPF\325) 320.98 580 T (s procedure for \337ushing old or ) 358.97 580 T (unreachable LSAs from the routing domain was veri\336ed, both in the presence of topology ) 85.54 566 T (loops and with many LSAs being \337ushed at once. This is also referred to as OSPF\325) 85.54 552 T (s MaxAge ) 482.63 552 T (procedure.) 85.54 538 T (\245) 72 518 T 6 F -0.25 (OSPF r) 85.54 518 P -0.25 (outing hierar) 124.05 518 P -0.25 (chy) 191.22 518 P -0.25 (. ) 208.55 518 P 0 F -0.25 (The OSPF four level routing hierarchy: intra-area, inter) 214.3 518 P -0.25 (-area. type 1 ) 478.13 518 P (externals and type 2 externals was tested.) 85.54 504 T (\245) 72 484 T 6 F (Import of external r) 85.54 484 T (outing information) 188.24 484 T 0 F (. The importing of external routes has been tested, ) 285.19 484 T (with as many as 400 imported at once. Also, the varying options in external LSAs has been ) 85.54 470 T (tested: type 1 or type 2 metrics and forwarding addresses.escribe all options. In addition, test ) 85.54 456 T (setups were utilized having AS boundary routers both internal to non-backbone areas and also ) 85.54 442 T (being simultaneously area border routers.) 85.54 428 T (\245) 72 408 T 6 F (Running pr) 85.54 408 T (otocol over various network types. ) 144.98 408 T 0 F (In the test setups, OSPF has been run over ) 323.52 408 T (ethernet, point-to-point serial lines \050both PPP and proprietary\051) 85.54 394 T 6 F (, ) 383.34 394 T 0 F (802.5 token ring and X.25.) 389.34 394 T (\245) 72 374 T 6 F (Non-br) 85.54 374 T (oadcast, multi-access networks. ) 122.63 374 T 0 F (OSPF has been tested over X.25. Some testing was ) 286.19 374 T (also done treating ethernet as a non-broadcast network. T) 85.54 360 T (wo separate situations were tested: ) 359.48 360 T -0.2 (when all routers attached to the non-broadcast network were DR-eligible, and when only some ) 85.54 346 P (of them were.) 85.54 332 T (\245) 72 312 T 6 F -0.54 (Authentication. ) 85.54 312 P 0 F -0.54 (OSPF\325) 167.62 312 P -0.54 (s authentication procedure was tested for the two current authentication ) 199.62 312 P (types.) 85.54 298 T (\245) 72 278 T 6 F (Equal-cost multipath. ) 85.54 278 T 0 F (Much of the testing was done in con\336gurations with redundant paths, ) 199.82 278 T (and equal-cost multipath was veri\336ed through examination of implementations\325 routing tables.) 85.54 264 T (\245) 72 244 T 6 F (V) 85.54 244 T (ariable-length subnet masks. ) 93.09 244 T 0 F (It was veri\336ed that implementations paid attention to the net-) 243.02 244 T (work mask \336eld present in OSPF LSAs.) 85.54 230 T (T) 72 204 T (esting was also performed on the following pieces of OSPF\325) 78.49 204 T (s Area functionality:) 367.3 204 T (\245) 72 184 T 6 F (Extent of advertisements.) 85.54 184 T 0 F ( It was veri\336ed that all advertisements except external LSAs were ) 215.77 184 T (\337ooded throughout a single area only) 85.54 170 T (.) 263.64 170 T (\245) 72 150 T 6 F (Inter) 85.54 150 T (-ar) 111.07 150 T (ea r) 126.17 150 T (outing. ) 145.6 150 T 0 F (The generation and processing of summary link LSAs was tested. Also ) 184.25 150 T (tested were con\336gurations having multiple area border routers attaching to a single area.) 85.54 136 T 6 F (\245) 72 116 T (V) 85.54 116 T (irtual links. ) 93.75 116 T FMENDPAGE %%EndPage: "32" 33 %%Page: "33" 33 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 33]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (The following OSPF options were also tested:) 72 676 T (\245) 72 656 T 6 F (T) 85.54 656 T (OS r) 93.32 656 T (outing. ) 117.43 656 T 0 F (The interplay between T) 156.08 656 T (OS-capable and non-T) 273.44 656 T (OS-capable routers was tested, ) 381.8 656 T (by con\336guring T) 85.54 642 T (OS-speci\336c metrics in the only implementation \0503com\051 supporting T) 165.94 642 T (OS rout-) 494.86 642 T (ing.) 85.54 628 T (\245) 72 608 T 6 F (Stub ar) 85.54 608 T (eas. ) 123.64 608 T 0 F (OSPF\325) 145.62 608 T (s stub area functionality was veri\336ed.) 177.62 608 T 3 F (7.6 T) 72 574.67 T (esting conclusions) 104.53 574.67 T 0 F (The interoperability testing has proven to be very valuable. Many bugs were found \050and \336xed\051 in ) 72 548 T -0.02 (the implementations. Some protocol problems were found \050and \336xed\051, and gray areas of the spec-) 72 534 P (i\336cation were cleared up. Implementations have also been \322bullet-proofed\323 in order to deal with ) 72 520 T -0.03 (the unexpected behavior of other implementations. All participants in the testing now understand ) 72 506 P (the maxim \322be conservative in what you generate, and liberal in what you accept\323 \050if they didn\325) 72 492 T (t ) 531.1 492 T (already\051.) 72 478 T 3 F (7.7 Futur) 72 444.67 T (e work) 131.21 444.67 T 0 F (The one thing that has gone mostly untested at the interoperability sessions is the interaction ) 72 418 T (between OSPF and other routing protocols \050such as RIP and EGP\051. Each interoperability session ) 72 404 T -0.22 (generally had a router running multiple routing protocols in order to import external routing infor-) 72 390 P (mation into the OSPF system. However) 72 376 T (, simultaneously running multiple routing protocols ) 262.08 376 T (between dif) 72 362 T (ferent vendors\325 routers has not been tested.) 128.07 362 T (Each vendor has developed a slightly dif) 72 336 T (ferent architecture for the exchange of routing informa-) 266.98 336 T -0.12 (tion between dif) 72 322 P -0.12 (fering routing protocols. As the OSPF \336eld testing has also shown, this exchange ) 149.49 322 P (of routing information is an area of ongoing work and a candidate for future standardization.) 72 308 T FMENDPAGE %%EndPage: "33" 34 %%Page: "34" 34 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 34]) 493.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (8.0 Simulation) 72 673.33 T 0 F -0.17 (The OSPF protocol has been simulated by the Distributed Systems Research Group at the Univer-) 72 646 P (sity of Maryland Baltimore County \050UMBC\051. The two principal investigators for the OSPF simu-) 72 632 T (lation project are Dr) 72 618 T (. Deepinder P) 168.26 618 T (. Sidhu of UMBC and Rob Coltun. They have been aided by ) 232.54 618 T (three graduate students: S. Abdallah, T) 72 604 T (. Fu and R. Nair) 257.98 604 T (. This section attempts to summarize their ) 334.61 604 T (simulation setup and results. For more information, contact the Distributed Systems Research ) 72 590 T (Group at the following address:) 72 576 T (Dr) 216 557.33 T (. Deepinder P) 227.99 557.33 T (. Sidhu) 292.28 557.33 T (Department of Computer Science) 216 541.33 T (University of Maryland Baltimore County) 216 524 T (Baltimore, MD 21228) 216 506 T (email: sidhu@umbc3.umbc.edu) 216 488 T -0.02 (A demo was given of their OSPF simulation at the March 4-8, 1991 IETF in St. Louis. Details of ) 72 462 P (the demo should be available in the IETF proceedings.) 72 448 T 3 F (8.1 Simulator setup) 72 414.67 T 0 F (The Distributed System Research Group uses a signi\336cantly enhanced version of the MIT net-) 72 388 T (work simulator) 72 374 T (. The simulator is event driven, and contains support for both point-to-point net-) 144.3 374 T (works and ethernet links. It can simulate characteristics of both packet switches and hosts, and ) 72 360 T (can simulate internet behavior under various types of data traf) 72 346 T (\336c load \050e.g., Poisson, normal, ) 369.22 346 T -0 (exponential and uniform distributions\051. This latter ability could be used, for example, to simulate ) 72 332 P (how a routing protocol works in a congested internet. Speci\336c network topologies can be input ) 72 318 T (into the simulator) 72 304 T (, or pseudo-random network topologies can be generated. Packet loss rates can ) 156.15 304 T (also be simulated.) 72 290 T (T) 72 264 T (o simulate OSPF) 78.49 264 T (, Rob Coltun\325) 158.83 264 T (s OSPF implementation was incorporated into the simulator) 223.81 264 T (, ) 510.84 264 T (essentially unchanged.) 72 250 T -0.29 (The output of the simulator can be displayed in a graphical manner \050it uses X windows\051. Any vari-) 72 224 P -0.23 (able in the implementation under test can be monitored. In addition, statistical reports can later be ) 72 210 P (produced from logging \336les produced during the simulation runs.) 72 196 T 3 F (8.2 Simulation r) 72 162.67 T (esults) 171.26 162.67 T 0 F (The OSPF simulation has been run using the following topologies:) 72 136 T FMENDPAGE %%EndPage: "34" 35 %%Page: "35" 35 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 35]) 493.7 73 T 72 108 540 684 R 7 X V 0 X (\245) 72 676 T -0.38 (The two sample topologies in the OSPF speci\336cation \050Figure 2 and Figure 6 in [2]\051. The \336rst of ) 85.54 676 P (these topologies shows an Autonomous System without areas, the second the same AS with ) 85.54 662 T (three areas and a virtual link con\336gured.) 85.54 648 T (\245) 72 628 T (The 19-node hub topology from [7].) 85.54 628 T (\245) 72 608 T (A lar) 85.54 608 T (ge network of over 50 nodes, all attached to a single ethernet.) 109.63 608 T (\245) 72 588 T -0.04 (A lar) 85.54 588 P -0.04 (ge network of over 50 nodes, containing both ethernets and serial lines, pseudo randomly ) 109.59 588 P (generated.) 85.54 574 T (In these topologies, the correctness of the OSPF database synchronization was veri\336ed. This was ) 72 548 T (done through examination of the nodes\325 OSPF link state databases and the nodes\325 routing tables. ) 72 534 T (The implementation of multiple OSPF areas was also tested. Also, database conver) 72 520 T (gence time ) 470.52 520 T (was analyzed as a function of the network components\325 link speeds.) 72 506 T (Also, some formal analysis of the OSPF protocol was undertaken. The neighbor and interface ) 72 480 T -0.05 (state machines were analyzed. In addition, the Designated Router election algorithm was veri\336ed ) 72 466 P (for certain sets of initial conditions.) 72 452 T FMENDPAGE %%EndPage: "35" 36 %%Page: "36" 36 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 36]) 493.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (9.0 Refer) 72 673.33 T (ence Documents) 137.87 673.33 T 0 F (The following documents have been referenced by this report:) 72 646 T ([1]) 72 626 T (Moy) 108 626 T (, J., \322The OSPF Speci\336cation\323, RFC 1) 129.88 626 T (131, October 1989.) 313.34 626 T ([2]) 72 608 T (Moy) 108 608 T (, J., \322OSPF V) 129.88 608 T (ersion 2\323, RFC 1247, July 1991.) 193.85 608 T ([3]) 72 590 T (Coltun, R. and Baker) 108 590 T (, F) 209.13 590 T (., \322OSPF V) 220.83 590 T (ersion 2 Management Information Base\323, RFC 1248, ) 274.14 590 T (July 1991.) 108 576 T ([4]) 72 558 T (Reynolds, J. and Postel, J., \322Assigned Numbers\325, RFC1060, March 1990.) 108 558 T ([5]) 72 540 T (Corporation for National Research Initiatives, \322Proceedings of the Thirteenth Internet ) 108 540 T (Engineering T) 108 526 T (ask Force\323, Cocoa Beach, Florida, April 1) 176.11 526 T (1-14, 1989.) 377.85 526 T ([6]) 72 508 T (Corporation for National Research Initiatives, \322Proceedings of the Sixteenth Internet ) 108 508 T (Engineering T) 108 494 T (ask Force\323, Florida State University) 176.11 494 T (, February 6-9, 1990.) 348.22 494 T ([7]) 72 476 T (Gardner) 108 476 T (, M., et al., \322T) 146.81 476 T (ype-of-service routing: modeling and simulation,\323 Report 6364, ) 213.59 476 T (BBN Communications Corporation, January 1987.) 108 462 T ([8]) 72 444 T (Corporation for National Research Initiatives, \322Proceedings of the Seventeenth Internet ) 108 444 T (Engineering T) 108 430 T (ask Force\323, Pittsbur) 176.11 430 T (gh Supercomputing Center) 270.83 430 T (, May 1-4, 1990.) 399.61 430 T ([9]) 72 412 T (Corporation for National Research Initiatives, \322Proceedings of the Eighteenth Internet ) 108 412 T (Engineering T) 108 398 T (ask Force\323, University of British Columbia, July 30-August 3, 1990.) 176.11 398 T FMENDPAGE %%EndPage: "36" 37 %%Page: "37" 37 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 37]) 493.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (10.0 People) 72 673.33 T 0 F -0.09 (The following people have contributed information to this report and can be contacted for further ) 72 646 P (information:) 72 632 T 72 596.01 540 604 C 72 601.98 540 601.98 2 L 0.5 H 0 Z 0 X 0 K N 0 0 612 792 C 5 F 0 X 0 K (T) 72 607.33 T (ABLE 5. People r) 77.93 607.33 T (efer) 152.71 607.33 T (ences in this r) 169.17 607.33 T (eport) 227.56 607.33 T (Refer) 72 585.34 T (ence T) 95.68 585.34 T (ag) 122.79 585.34 T (Name) 144 585.34 T (Af\336liation) 234 585.34 T (email) 396 585.34 T 1 F (bstreile) 72 568.34 T (W) 144 568.34 T (illiam Streilein) 153.04 568.34 T (W) 234 568.34 T (ell\337eet Communications) 242.63 568.34 T (bstreile@well\337eet.com) 396 568.34 T (dino) 72 552.34 T (Dino Farinacci) 144 552.34 T (3com Corporation) 234 552.34 T (dino@buckeye.esd.3com.com) 396 552.34 T (fbaker) 72 536.34 T (Fred Baker) 144 536.34 T (Advanced Computer Communications) 234 536.34 T (fbaker@acc.com) 396 536.34 T (jef) 72 520.34 T (f) 82.36 520.34 T (Jef) 144 520.34 T (frey Bur) 155.47 520.34 T (gan) 188.88 520.34 T (Sterling Software) 234 520.34 T (jef) 396 520.34 T (f@nsipo.nasa.gov) 406.36 520.34 T (jhsu) 72 504.34 T (Jonathan Hsu) 144 504.34 T (W) 234 504.34 T (ell\337eet Communications) 242.63 504.34 T (jhsu@well\337eet.com) 396 504.34 T (jmoy) 72 488.34 T (John Moy) 144 488.34 T (Proteon, Inc.) 234 488.34 T (jmoy@proteon.com) 396 488.34 T (kannan) 72 472.34 T (Kannan V) 144 472.34 T (aradhan) 183.69 472.34 T (OARnet) 234 472.34 T (kannan@oar) 396 472.34 T (.net) 446.28 472.34 T (medin) 72 456.34 T (Milo Medin) 144 456.34 T (Sterling Software) 234 456.34 T (medin@nsipo.nasa.gov) 396 456.34 T (rcoltun) 72 440.34 T (Rob Coltun) 144 440.34 T (University of Maryland) 234 440.34 T (rcoltun@umd5.umd.edu) 396 440.34 T (rrv) 72 424.34 T (Ross V) 144 424.34 T (each) 172.04 424.34 T (University of Illinois) 234 424.34 T (rrv@seka.cso.uiuc.edu) 396 424.34 T (vaf) 72 408.34 T (V) 144 408.34 T (ince Fuller) 150.62 408.34 T (BARRNet) 234 408.34 T (vaf@valinor) 396 408.34 T (.stanford.edu) 445.73 408.34 T FMENDPAGE %%EndPage: "37" 38 %%Page: "38" 38 612 792 0 FMBEGINPAGE 72 702 540 720 R 7 X 0 K V 0 F 0 X (RFC 1246) 72 712 T (Experience with OSPF) 251.03 712 T (July 1991) 493.02 712 T 72 69.05 540 81 R 7 X V 0 X ([Moy]) 72 73 T ([Page 38]) 493.7 73 T 72 108 540 684 R 7 X V 2 F 0 X (Security Considerations) 72 673.33 T 0 F (The OSPF protocol\325) 72 646 T (s security architecture is described in Section 4.0.) 168.62 646 T 2 F (Author) 72 605.33 T (\325) 122.04 605.33 T (s Addr) 126.77 605.33 T (ess) 173.13 605.33 T 0 F (John Moy) 72 578 T (Proteon Inc.) 72 564 T (2 T) 72 550 T (echnology Drive) 87.48 550 T (W) 72 536 T (estborough, MA 01581) 82.36 536 T (Phone: \050508\051 898-2800) 72 510 T (Email: jmoy@proteon.com) 72 496 T FMENDPAGE %%EndPage: "38" 39 %%Trailer %%BoundingBox: 0 0 612 792 %%Pages: 38 1 %%DocumentFonts: Times-Roman %%+ Times-Bold %%+ Times-Italic