Re: 1000ms delay in networking stack or driver, new bug?

Sami Farin (sfarin@ratol.fi)
Wed, 22 Sep 1999 01:45:51 +0300 ( )

I experienced this, too... I was playing with ntpd 4.0.97.
When I was tweaking ntp configs, system time got changed many times
by ntpd... Now that ntp is setup ok and time doesn't change sporadically
anymore, I haven't noticed anything odd with 2.2.12 kernel in 12 days
(wrt ping times).

PING taimen.ratol.fi (192.26.135.84): 56 data bytes
64 bytes from 192.26.135.84: icmp_seq=0 ttl=252 time=4007.2 ms
64 bytes from 192.26.135.84: icmp_seq=1 ttl=252 time=4003.2 ms
64 bytes from 192.26.135.84: icmp_seq=2 ttl=252 time=4002.3 ms
64 bytes from 192.26.135.84: icmp_seq=3 ttl=252 time=4002.5 ms
64 bytes from 192.26.135.84: icmp_seq=4 ttl=252 time=6003.7 ms
64 bytes from 192.26.135.84: icmp_seq=5 ttl=252 time=6002.1 ms
64 bytes from 192.26.135.84: icmp_seq=6 ttl=252 time=6005.7 ms
64 bytes from 192.26.135.84: icmp_seq=7 ttl=252 time=6002.1 ms
64 bytes from 192.26.135.84: icmp_seq=8 ttl=252 time=5518.2 ms
64 bytes from 192.26.135.84: icmp_seq=9 ttl=252 time=5003.2 ms
64 bytes from 192.26.135.84: icmp_seq=10 ttl=252 time=5004.0 ms
64 bytes from 192.26.135.84: icmp_seq=11 ttl=252 time=5002.6 ms
64 bytes from 192.26.135.84: icmp_seq=12 ttl=252 time=5004.0 ms
64 bytes from 192.26.135.84: icmp_seq=13 ttl=252 time=5002.1 ms
64 bytes from 192.26.135.84: icmp_seq=14 ttl=252 time=6002.1 ms
64 bytes from 192.26.135.84: icmp_seq=15 ttl=252 time=6002.1 ms
64 bytes from 192.26.135.84: icmp_seq=16 ttl=252 time=6002.2 ms
64 bytes from 192.26.135.84: icmp_seq=17 ttl=252 time=6008.3 ms
64 bytes from 192.26.135.84: icmp_seq=18 ttl=252 time=6002.9 ms
64 bytes from 192.26.135.84: icmp_seq=19 ttl=252 time=6004.2 ms
64 bytes from 192.26.135.84: icmp_seq=20 ttl=252 time=6002.2 ms
64 bytes from 192.26.135.84: icmp_seq=21 ttl=252 time=6002.8 ms
64 bytes from 192.26.135.84: icmp_seq=22 ttl=252 time=6002.2 ms
64 bytes from 192.26.135.84: icmp_seq=23 ttl=252 time=6002.3 ms
64 bytes from 192.26.135.84: icmp_seq=24 ttl=252 time=6002.2 ms
64 bytes from 192.26.135.84: icmp_seq=25 ttl=252 time=6002.2 ms
64 bytes from 192.26.135.84: icmp_seq=26 ttl=252 time=6003.8 ms
64 bytes from 192.26.135.84: icmp_seq=27 ttl=252 time=6004.0 ms
64 bytes from 192.26.135.84: icmp_seq=28 ttl=252 time=6002.4 ms
64 bytes from 192.26.135.84: icmp_seq=29 ttl=252 time=5023.7 ms
64 bytes from 192.26.135.84: icmp_seq=30 ttl=252 time=5002.3 ms
64 bytes from 192.26.135.84: icmp_seq=31 ttl=252 time=4007.8 ms
64 bytes from 192.26.135.84: icmp_seq=32 ttl=252 time=4003.5 ms
64 bytes from 192.26.135.84: icmp_seq=33 ttl=252 time=3011.8 ms
64 bytes from 192.26.135.84: icmp_seq=34 ttl=252 time=3002.8 ms
64 bytes from 192.26.135.84: icmp_seq=35 ttl=252 time=4007.3 ms
64 bytes from 192.26.135.84: icmp_seq=36 ttl=252 time=5002.2 ms
64 bytes from 192.26.135.84: icmp_seq=37 ttl=252 time=6003.4 ms
64 bytes from 192.26.135.84: icmp_seq=38 ttl=252 time=6003.1 ms

--- taimen.ratol.fi ping statistics ---
45 packets transmitted, 39 packets received, 13% packet loss

ifdown/ifup didn't cure ping times, I had to reboot.
After reboot I got 2.8ms pings on average (don't remember loss-%).
I have x86+gcc-2.95.1. 3c509 card (15GB dl/19GB ul in 12 days).

-- 
Safari - safari@iki.fi - PGP key 0x443BD271 - http://www.iki.fi/safari/
  The UNIX Guru's View of Sex: "unzip ; strip ; touch ; finger ; \
  mount ; fsck ; more ; yes ; umount ; sleep"




-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/