Date: Mon, 9 Dec 2002 14:50:18 +0100 From: "Koepp, Karsten" <Karsten.Koepp@lambdanet.net>
can anyone of you point me to statistics about the frequency of BGP updates measured at some point in the internet? A customer complained about av. 10 updates per second which we consider normal, but we'd like to see values for comparison.
Karsten, thats really an interesting question: ACOnet receives a full routing-table from SPRINT/AS1239 and from GEANT/AS20965. Comparing the current BGP session counters: neighbor SPRINT, uptime ~20weeks, Received 7926940 messages neighbor GEANT, uptime ~ 7weeks, Received 29597194 messages which gives a long-term average of 419 BGP update messages per minute from GEANT but only 39 BGP update messages per minute from SPRINT afaik GEANT does no BGP flap damping, SPRINT however does, according to http://www.sprint.net/policy/bgpdamp.html I don't know if damping is the only reason for the factor >10 between them, there might be other reasons as well, e.g. SPRINT backbone is Cisco based, GEANT is Juniper based and they might be using completely different IGPs and iBGP concepts. I am also suspecting some improperly behaving BGP implementations beeing around, e.g. at the VIX we are having peering neighbors who are sending superproportional more BGP updates than others with the majority of the messages beeing withdrawals for prefixes which they do not own nor announce. Btw, Lambdanet is sending us a long-term average of 2 BGP update messages per minute at the VIX. Kind regards CP --- ---------------------------------------------------------------------- --- --- Christian Panigl : Vienna University Computer Center - ACOnet --- --- VUCC - ACOnet - VIX : -------------------------------------------- --- --- Universitaetsstrasse 7 : Mail: Panigl@CC.UniVie.ac.at (CP8-RIPE) --- --- A-1010 Vienna / Austria : Tel: +43 1 4277-14032 (Fax: -9140) --- --- ---------------------------------------------------------------------- ---
At 04:06 PM 11-12-02 +0100, you wrote:
Date: Mon, 9 Dec 2002 14:50:18 +0100 From: "Koepp, Karsten" <Karsten.Koepp@lambdanet.net>
can anyone of you point me to statistics about the frequency of BGP updates measured at some point in the internet? A customer complained about av. 10 updates per second which we consider normal, but we'd like to see values for comparison.
Karsten,
thats really an interesting question:
ACOnet receives a full routing-table from SPRINT/AS1239 and from GEANT/AS20965. Comparing the current BGP session counters:
neighbor SPRINT, uptime ~20weeks, Received 7926940 messages neighbor GEANT, uptime ~ 7weeks, Received 29597194 messages
I get: Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd 62.40.103.225 4 20965 5185632 50759 14159389 0 0 5w0d 117711 I get 102/min from Geant. But going to Oregon route-views: Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd 4.0.4.90 4 1 190667 7057 1058412 0 0 4d21h 115692 62.164.11.10 4 8782 9569 7058 1058412 0 0 4d21h 2841 64.50.224.5 4 4181 189574 7058 1058412 0 0 4d21h 116139 64.166.72.140 4 65533 0 0 0 0 0 never Active 64.200.199.3 4 7911 276874 7056 1058412 0 0 4d21h 116131 64.200.199.4 4 7911 286984 7057 1058412 0 0 4d21h 116132 65.106.7.139 4 2828 232735 7057 1058412 0 0 4d21h 115893 66.185.128.48 4 1668 221156 7057 1058412 0 0 4d21h 116882 129.250.0.6 4 2914 257151 7057 1058412 0 0 4d21h 100326 129.250.0.11 4 2914 212586 7057 1058412 0 0 4d21h 100331 130.217.2.25 4 681 7404 7057 1058412 0 0 4d21h 863 134.55.20.229 4 293 264268 7057 1058412 0 0 4d21h 117149 141.142.12.1 4 1224 394416 14087 1058412 0 0 4d21h 118853 144.228.241.81 4 1239 174356 7056 1058412 0 0 4d21h 113516 154.11.63.86 4 852 176176 7056 1058412 0 0 4d21h 117555 154.11.98.18 4 852 175166 7056 1058412 0 0 4d21h 117551 155.229.0.36 4 4565 8311 7056 1058412 0 0 4d21h 603 157.22.9.7 4 715 0 0 0 0 0 never Active 157.130.182.254 4 19092 3630791 7056 1058412 0 0 4d21h 116131 157.130.185.17 4 701 0 0 0 0 0 never Active 167.142.3.6 4 5056 221492 7057 1058412 0 0 4d21h 115884 192.12.65.1 4 13 7080 7056 1058412 0 0 4d21h 14 192.121.154.25 4 1755 0 0 0 0 0 never Active 192.205.31.33 4 7018 185694 7056 1058412 0 0 4d21h 115560 193.0.0.56 4 3333 194057 7056 1058412 0 0 4d21h 117778 193.251.128.22 4 5511 215679 7056 1058412 0 0 4d21h 115891 194.85.4.249 4 3277 0 0 0 0 0 never Active 195.66.224.82 4 4513 332053 7056 1058412 0 0 4d21h 116670 195.66.232.239 4 5459 146403 7056 1058412 0 0 4d21h 43052 195.66.232.254 4 5459 96665 7056 1058412 0 0 4d21h 43052 195.219.96.239 4 8297 148848 313 0 0 0 1d18h Active 196.7.106.72 4 2905 7222 7050 1058412 0 0 1d14h 153 198.32.8.252 4 11537 43213 7056 1058412 0 0 4d21h 7280 198.32.162.1 4 3582 7074 7056 1058412 0 0 4d21h 1 202.232.1.91 4 2497 247209 7056 1058412 0 0 4d21h 116032 202.249.2.86 4 7500 252732 7057 1058412 0 0 4d21h 117750 203.62.252.26 4 1221 170186 7056 1058412 0 0 4d21h 116474 203.181.248.233 4 7660 470644 7023 1058412 0 0 1d05h 103885 203.194.0.5 4 9942 191459 7043 1058412 0 0 2d14h 115852 204.29.239.1 4 6066 0 0 0 0 0 never Active 204.42.253.253 4 267 121171 7056 1058412 0 0 4d21h 100284 204.152.184.126 4 3557 0 0 0 0 0 never Active 204.212.44.131 4 234 17468 7056 1058412 0 0 4d21h 9315 205.189.32.153 4 6509 0 0 0 0 0 never Active 205.215.45.50 4 4006 0 0 0 0 0 never Active 206.186.255.223 4 2493 161467 7056 1058412 0 0 4d21h 115746 206.220.240.95 4 10764 510164 7056 1058412 0 0 4d21h 117276 207.45.223.244 4 6453 571078 7059 1058412 0 0 1d09h 115827 207.126.96.1 4 6461 249283 7057 1058412 0 0 4d21h 116166 207.172.6.173 4 6079 247608 7056 1058412 0 0 4d21h 108573 207.172.6.221 4 6079 261547 7056 1058412 0 0 4d21h 108573 207.246.129.6 4 11608 163496 7056 1058412 0 0 4d21h 116083 207.246.129.14 4 11608 0 0 0 0 0 never Active 208.51.113.253 4 3549 565481 14089 1058412 0 0 2d14h 116007 208.172.146.2 4 3561 182408 7056 1058412 0 0 4d21h 115733 208.186.154.35 4 5650 261002 14078 1058412 0 0 1d03h 115992 208.186.154.36 4 5650 257134 14077 1058412 0 0 2d04h 115992 209.10.12.28 4 4513 328739 7056 1058412 0 0 4d21h 116669 209.10.12.125 4 4513 509182 7056 1058412 0 0 4d21h 116670 209.10.12.156 4 4513 344348 7056 1058412 0 0 4d21h 116670 209.19.207.70 4 2548 11766 7055 1058412 0 0 4d21h 1079 209.123.12.4 4 8001 369722 7057 1058412 0 0 4d21h 116628 209.161.175.4 4 14608 227675 7056 1058412 0 0 4d21h 116014 209.244.2.115 4 3356 190592 7056 1058412 0 0 4d21h 115720 212.4.193.253 4 8918 0 0 0 0 0 never Active 213.140.32.144 4 12956 1602548 7056 1058412 0 0 4d21h 116414 213.200.87.254 4 3257 377670 14087 1058412 0 0 4d21h 116225 216.18.63.137 4 6539 377725 7061 1058412 0 0 2d10h 116044 216.140.2.62 4 6395 537616 14087 1058412 0 0 4d21h 115873 216.140.8.59 4 6395 526492 14087 1058412 0 0 4d21h 115874 216.140.14.186 4 6395 364923 14087 1058412 0 0 4d21h 115872 217.75.96.60 4 16150 916050 7090 1058412 0 0 20:57:29 116697 which shows about 25-35/min on avg for full feeds. Clearly there is something wrong with the Geant feed. I get lots of output on 'sho ip bgp flap-sta' from AS3549 (GBLX) and AS1299 (Telia) as seen from Geant. Should we take this to the APM list? -Hank
which gives a long-term average of
419 BGP update messages per minute from GEANT but only 39 BGP update messages per minute from SPRINT
afaik GEANT does no BGP flap damping, SPRINT however does, according to http://www.sprint.net/policy/bgpdamp.html
I don't know if damping is the only reason for the factor >10 between them, there might be other reasons as well, e.g. SPRINT backbone is Cisco based, GEANT is Juniper based and they might be using completely different IGPs and iBGP concepts.
I am also suspecting some improperly behaving BGP implementations beeing around, e.g. at the VIX we are having peering neighbors who are sending superproportional more BGP updates than others with the majority of the messages beeing withdrawals for prefixes which they do not own nor announce.
Btw, Lambdanet is sending us a long-term average of 2 BGP update messages per minute at the VIX.
Kind regards CP
--- ---------------------------------------------------------------------- --- --- Christian Panigl : Vienna University Computer Center - ACOnet --- --- VUCC - ACOnet - VIX : -------------------------------------------- --- --- Universitaetsstrasse 7 : Mail: Panigl@CC.UniVie.ac.at (CP8-RIPE) --- --- A-1010 Vienna / Austria : Tel: +43 1 4277-14032 (Fax: -9140) --- --- ---------------------------------------------------------------------- ---
Christian wrote: .
From: "Koepp, Karsten" <Karsten.Koepp@lambdanet.net>
can anyone of you point me to statistics about the frequency of BGP updates measured at some point in the internet? A customer complained about av. 10 updates per second which we consider normal, but we'd like to see values for comparison.
You might find RIS at http://www.ripe.net/ris/ quite interresting :) Greets, Jeroen
participants (3)
-
Christian Panigl, ACOnet/VIX/UniVie -
Hank Nussbacher -
Jeroen Massar