PDA

View Full Version : Packet loss just 3 hops away


Dahan
02-02-2006, 08:37 PM
I noticed that interactive ssh sessions to my VPS would often stall for a few seconds, and after doing some pings, it seems there's sometimes quite a bit of packet loss, even to nearby routers. Here's a recent example pinging gnax-atl-KalahariLimited-dsl-wan1-gnax.net (209.51.131.101), which is hop #3 according to a traceroute:

traceroute: Warning: google.com has multiple addresses; using 64.233.187.99
traceroute to google.com (64.233.187.99), 30 hops max, 38 byte packets
1 169.254.1.1 (169.254.1.1) 0.411 ms 0.176 ms 0.137 ms
2 207.210.78.1 (207.210.78.1) 0.555 ms 0.395 ms 0.286 ms
3 gnax-atl-KalahariLimited-dsl-wan1-gnax.net (209.51.131.101) 1.020 ms 0.907 ms 0.753 ms
[ ... ]

64 bytes from 209.51.131.101: icmp_seq=4018 ttl=253 time=11.2 ms
64 bytes from 209.51.131.101: icmp_seq=4019 ttl=253 time=1.6 ms
64 bytes from 209.51.131.101: icmp_seq=4020 ttl=253 time=4.6 ms
64 bytes from 209.51.131.101: icmp_seq=4021 ttl=253 time=787420.0 ms
64 bytes from 209.51.131.101: icmp_seq=4022 ttl=253 time=786410.1 ms
64 bytes from 209.51.131.101: icmp_seq=4023 ttl=253 time=785400.1 ms
64 bytes from 209.51.131.101: icmp_seq=4024 ttl=253 time=784390.1 ms
64 bytes from 209.51.131.101: icmp_seq=4025 ttl=253 time=783380.1 ms
64 bytes from 209.51.131.101: icmp_seq=4026 ttl=253 time=782370.1 ms
64 bytes from 209.51.131.101: icmp_seq=4027 ttl=253 time=781360.1 ms
64 bytes from 209.51.131.101: icmp_seq=4028 ttl=253 time=780350.1 ms
64 bytes from 209.51.131.101: icmp_seq=4029 ttl=253 time=779340.1 ms
64 bytes from 209.51.131.101: icmp_seq=4030 ttl=253 time=778330.1 ms
64 bytes from 209.51.131.101: icmp_seq=4031 ttl=253 time=777320.1 ms
64 bytes from 209.51.131.101: icmp_seq=4032 ttl=253 time=776309.2 ms
64 bytes from 209.51.131.101: icmp_seq=4033 ttl=253 time=775300.1 ms
64 bytes from 209.51.131.101: icmp_seq=4034 ttl=253 time=774290.2 ms
64 bytes from 209.51.131.101: icmp_seq=4035 ttl=253 time=773280.2 ms
64 bytes from 209.51.131.101: icmp_seq=4036 ttl=253 time=772270.2 ms
64 bytes from 209.51.131.101: icmp_seq=4037 ttl=253 time=771260.2 ms
64 bytes from 209.51.131.101: icmp_seq=4038 ttl=253 time=770250.2 ms
64 bytes from 209.51.131.101: icmp_seq=4039 ttl=253 time=769240.2 ms
64 bytes from 209.51.131.101: icmp_seq=4040 ttl=253 time=768230.2 ms
64 bytes from 209.51.131.101: icmp_seq=4041 ttl=253 time=767220.2 ms
64 bytes from 209.51.131.101: icmp_seq=4042 ttl=253 time=766210.2 ms
64 bytes from 209.51.131.101: icmp_seq=4043 ttl=253 time=765200.2 ms
64 bytes from 209.51.131.101: icmp_seq=4044 ttl=253 time=764190.2 ms
64 bytes from 209.51.131.101: icmp_seq=4801 ttl=253 time=1.1 ms
64 bytes from 209.51.131.101: icmp_seq=4802 ttl=253 time=0.9 ms
64 bytes from 209.51.131.101: icmp_seq=4803 ttl=253 time=1.0 ms
64 bytes from 209.51.131.101: icmp_seq=4804 ttl=253 time=1.1 ms
64 bytes from 209.51.131.101: icmp_seq=4805 ttl=253 time=0.9 ms
[ ... ]
--- 209.51.131.101 ping statistics ---
9565 packets transmitted, 7507 packets received, 21% packet loss
round-trip min/avg/max = 0.6/2483.7/787420.0 ms

I don't mind occasional 12+ minute outages, but there seems to be constant low-level packet loss (1 to 2%), which seems a bit on the high side given how close
209.51.131.101 is. I'd just like for my ssh session to not stall for a few seconds every 10 minutes...

--- 209.51.131.101 ping statistics ---
9769 packets transmitted, 9605 packets received, 1% packet loss
round-trip min/avg/max = 0.6/3.3/86.1 ms

--- 209.51.131.101 ping statistics ---
4958 packets transmitted, 4877 packets received, 1% packet loss
round-trip min/avg/max = 0.6/4.7/306.5 ms