Messages in this thread | | | Date | Fri, 9 Jul 1999 11:07:16 -0600 (MDT) | From | Jason Gunthorpe <> | Subject | Ethernet slowdown problem |
| |
I just got hit by that weird ethernet slowdown problem that some people have reported, here is the results of a ping over ethernet:
--- blue.priv.deltatee.com ping statistics --- 1165 packets transmitted, 1159 packets received, 0% packet loss round-trip min/avg/max = 1796.0/4706.5/7994.9 ms
The very odd part is that telnet sessions out of this machine are very responsive still - I am typing this message over telnet. The nfs client is -extremely- slow and so is the (user land) NFS server that this machine runs. DNS has also become quite slow.
I fired up tcpdump on my DNS server and did a lookup, here is the trace:
10:56:51.387148 172.16.1.131.1035 > 172.16.1.1.53: 9218+ (29) 10:56:51.387148 172.16.1.1.53 > 172.16.1.131.1035: 9218 NXDomain* 0/1/0 (102) 10:56:54.386925 172.16.1.131.1035 > 172.16.1.2.53: 9218+ (29) 10:56:54.396925 172.16.1.2.53 > 172.16.1.131.1035: 9218 NXDomain* 0/1/0 (102)
I also straced the 'host' command and noticed that it sent a total of 4 messages using send and in this instance it wasn't able to recieve any messages, it reported a DNS timeout error.
The kernel is 2.2.10 with no patches, the network card is a Tulip using the tulip driver (Dlink DE-530):
tulip.c:v0.89H 5/23/98 becker@cesdis.gsfc.nasa.gov eth0: Digital DC21041 Tulip at 0x6400, 21041 mode, 00 80 c8 7d e6 28, IRQ 10. eth0:21041 Media information at 30, default media 0800 (Autosense). eth0: 21041 media #0, 10baseT. eth0: 21041 media #0, 10baseT. eth0: 21041 media #0, 10baseT.
Hrm, just as I finished writing this the problem went away (which I have never seen it do before).
Hope this helps, Jason
- 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/
| |