From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i586; en-US; rv:0.9.9) Gecko/20020513
Description of problem:
nfs no longer works. I bet if I restart it, it will work, but I have not yet
done this. I have data in /var/log/messages about this. I can ping the "failed"
machine from the other one on my LAN, but df does not see the failed machine.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Everyday at a little after 1AM, my other machine runs (under cron) a backup of
the two machines on my LAN. This morning, the backup failed to pick up anything
on this machine, complaining that it could not stat the files on this machine (I
The complaints are in /var/log/messages (to be attached).
I suppose it is remotely possible that this is a hardware problem, but I have
not seen one on this machine in several years, and that one was fixed by
replacing the cache in about 1998.
Created attachment 61165 [details]
/var/log/messages data in support of this bug.
kernel on this machine is:
touchl:jdbeyer[~]$ rpm -q kernel
touchl:jdbeyer[~]$ uname -r
I tried to do /etc/rc.d/init.d/nfs restart and it did not work; trouble shutting
down the nfsd processes, it seemed. I could not get rid of them with kill -9
(they got into the D state), so I rebooted the machine.
I have found the same problem on a production Redhat Linux 7.3 server running
The messages file output is the same
It is imperative that this gets fixed as this is crashing my production server
which supports 100 various UNIX clients.
this is assumed to be fixed in the 2.4.18-17.7.x erratum kernel
I understand that the bug is fixed now.
In which case, how can I find more about what was wrong so that I can perhaps
minimise the impact of the bug on my system until I can upgrade the kernel?
Well there are quite a few code changes.
If you want those you end up with a new kernel anyway...