2.5->2.4 nfs corrupts
nfs server: 2.4.21-pre2
nfs client: 2.5.67
Every now and then I have problems copying data to and from the nfs mount. It can either result in aborted writes like this:
15 [11:23:59] >> mv maestro.mpeg /share/funny
mv: closing `/share/funny/maestro.mpeg': Input/output error 15 [11:24:24] >> cp maestro.mpeg /share/funny
(it succeeded on the 2nd go)
or reads where unzip failed to unzip a file across the nfs link. there's nothing in dmesg on either side of the fence to indicate problems and while there was no message in the logs today, yesterday when I was writing to the nfs server and got some corrupt files tyhis was logged:
messages:Apr 9 11:50:19 theirongiant kernel: nfs: server bob not responding, timed out messages:Apr 9 13:15:29 theirongiant kernel: nfs: server bob not responding, timed out
But this is a local lan and the pc was never unplugged nor do problems come up in anything else at any other time. The eepro driver didn't detect a disconnection either.
Hope this helps. I'll do some more digging if someone can tell me what to dig with and how. This is all I can think of atm.
Signature · 6 lines
Martin's distress was in contrast to the bitter satisfaction of some of his fellow marines as they surveyed the scene. "The Iraqis are sick people and we are the chemotherapy," said Corporal Ryan Dupre. "I am starting to hate this country. Wait till I get hold of a friggin' Iraqi. No, I won't get hold of one. I'll just kill him." - http://www.informationclearinghouse.info/article2479.htm
Quoted 2 lines · CaT
nfs server: 2.4.21-pre2
nfs client: 2.5.67
Quite a few people (myself included) are seeing this.
Quoted 1 line
From the reports I've seen so far, it looks like it only
happens when the client is a faster box than the server. In my case, I have a 2.8Ghz P4 as the client, hammering a poor defenceless 1Ghz VIA C3.
This is probably worth creating a bugzilla entry for.
Dave
Quoted 9 lines · Dave Jones
Quoted 2 lines · CaT
nfs server: 2.4.21-pre2
nfs client: 2.5.67Quite a few people (myself included) are seeing this.
Quoted 1 line
From the reports I've seen so far, it looks like it only
happens when the client is a faster box than the server. In my case, I have a 2.8Ghz P4 as the client, hammering a poor defenceless 1Ghz VIA C3.
This is the case here also: a p3 700 hammering a k6 300.
Quoted 1 line
This is probably worth creating a bugzilla entry for.
Would this be something I'd do or someone else?
Signature · 6 lines
Martin's distress was in contrast to the bitter satisfaction of some of his fellow marines as they surveyed the scene. "The Iraqis are sick people and we are the chemotherapy," said Corporal Ryan Dupre. "I am starting to hate this country. Wait till I get hold of a friggin' Iraqi. No, I won't get hold of one. I'll just kill him." - http://www.informationclearinghouse.info/article2479.htm
Petition against EU software patents
In reply to Dave JonesQuoted 11 lines · Dave Jones
Quoted 2 lines · CaT
nfs server: 2.4.21-pre2
nfs client: 2.5.67Quite a few people (myself included) are seeing this. From the reports I've seen so far, it looks like it only happens when the client is a faster box than the server. In my case, I have a 2.8Ghz P4 as the client, hammering a poor defenceless 1Ghz VIA C3.
This is probably worth creating a bugzilla entry for.
It may be worth noting that I'm not seeing any problems between an AMD XP 1800+ client running 2.5.66-mm3 and an AMD K6-2 300 server running SuSE's 2.4.18-4GB kernel. SuSE applies quite a lot of NFS-related patches to their kernels, if I'm not mistaken.
I'm mounting /home, /var/spool/mail and /mp3 over NFS, so I think I should have noticed problems if there were any ;-)
Signature · 1 line
Erik Hensema <[email protected]>
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
" " == cat <[email protected]> writes:
> nfs server: 2.4.21-pre2 nfs client: 2.5.67
> Every now and then I have problems copying data to and from the > nfs mount. It can either result in aborted writes like this:
> 15 [11:23:59] >> mv maestro.mpeg /share/funny mv: closing > `/share/funny/maestro.mpeg': Input/output error 15 [11:24:24] > >> cp maestro.mpeg /share/funny
Unless you are using soft mounts (in which case problem == user error) I suggest you try the latest (as of last night - probably late afternoon Oz-time) bitkeeper. It contains a patch that has been posted several times on this list against a 2.5. NFS client read corruption.
Cheers,
Trond