Kernel 2.4.21 or RHEL AS3 Bug under Taroon? 2006-02-07 - By Justin Piszcz
Back
*
I have a quad-cpu server running RHEL AS 3, Taroon that acts as an NFS
server
for some other machines.
It had a 390 uptime when the problem occured.
Essentially, the directory paths it serves out stopped working, attempting
to
mount the directory on two client boxes results in:
box1# mount rhelas3_box:/share1 /mnt/share1
mount: RPC: Program not registered
box2# mount rhelas3_box:/share2 /mnt/share2
mount: RPC: Program not registered
My guess as to what happened is portmap died on the server and then the rpc
commands could not be established between the server and the clients.
The options used for the mount are as follows:
rhelas3_box:/share1 /mnt/share1 nfs
ro,tcp,nfsvers=3,rsize=8192,timeo=11,hard,intr,defaults
rhelas3_box:/share2 /mnt/share2 nfs
ro,tcp,nfsvers=3,rsize=8192,timeo=11,hard,intr,defaults
On the server in /etc/exports, the shares are below:
/mnt/share1 1.2.3.4(rw,no_root_squash,sync)
/mnt/share2 1.2.3.4(rw,no_root_squash,sync)
Is this more of a kernel issue or specifically the portmap daemon? I have
not
seen this error in the past where NFS mounts randomly die, disappear or fail
to
work.
Other than checking the logs during the time of the failure, what is the
best
way to begin troubleshooting this issue?
Thanks,
Justin.
*
<b><font size="2">
<p>I have a quad-cpu server running RHEL AS 3, Taroon that acts as an NFS
server <br>for some other machines.<br><br>It had a 390 uptime when the problem
occured.<br><br>Essentially, the directory paths it serves out stopped working,
attempting to
<br>mount the directory on two client boxes results in:<br><br>box1# mount
rhelas3_box:/share1 /mnt/share1<br>mount: RPC: Program not registered<br><br
>box2# mount rhelas3_box:/share2 /mnt/share2<br>mount: RPC: Program not
registered
<br><br>My guess as to what happened is portmap died on the server and then the
rpc <br>commands could not be established between the server and the clients.
<br><br>The options used for the mount are as follows:<br>rhelas3_box:/share1
/mnt/share1 nfs
<br>ro,tcp,nfsvers=3,rsize=8192,timeo=11,hard,intr,defaults<br>rhelas3_box:
/share2 /mnt/share2 nfs <br>ro,tcp,nfsvers=3,rsize=8192,timeo=11
,hard,intr,defaults<br><br>On the server in /etc/exports, the shares are below:
<br>
/mnt/share1 1.2.3.4(rw,no_root_squash,sync)<br>/mnt/share2 1.2.3.4
(rw,no_root_squash,sync)<br><br>Is this more of a kernel issue or specifically
the portmap daemon? I have not <br>seen this error in the past where NFS
mounts randomly die, disappear or fail to
<br>work.<br><br>Other than checking the logs during the time of the failure,
what is the best <br>way to begin troubleshooting this issue?<br><br>Thanks,<br
><span></span><span class="sg"><br><font color="#000000">Justin.
</font></span></p></font></b>
--
Taroon-list mailing list
Taroon-list@(protected)
https://www.redhat.com/mailman/listinfo/taroon-list
|
|
