hung server

hung server

[David Dougall]

I have a linux NFS server running base 2.4.18 with xfs patches from SGI.
Seeminly random times, the nfs server will become unresponsive.  On
a client, doing showmount -e [server] will timeout.  Looking on the
server shows that the nfs daemons will go into the DW state as shown
by ps:
root       123  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
root       124  0.2  0.0     0    0 ?        DW   Jul25   2:50 [nfsd]
root       127  0.2  0.0     0    0 ?        DW   Jul25   2:52 [nfsd]
root       128  0.2  0.0     0    0 ?        DW   Jul25   2:52 [nfsd]
root       129  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
root       130  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
root       131  0.3  0.0     0    0 ?        DW   Jul25   2:55 [nfsd]
root       132  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]

  The load will sit at about 8 which is how many nfsd's that I have
running.  I thought this was usually caused by a hardware problem, but I
am able to get the the local filesystems that are exported just fine.
Any help would be appreciated.
Thanks
--David Dougall



-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: hung server

[James Pearson]

This sounds the same as the problem I reported - see:

http://marc.theaimsgroup.com/?l=linux-nfs&m=102579958731319&w=2

and the thread at:

http://marc.theaimsgroup.com/?l=linux-xfs&m=102579948531207&w=2


I've only just upgraded to a 2.4.19-rc3 + XFS CVS kernel, so I don't
know yet if the problem has been resolved...

James Pearson

David Dougall wrote:
> 
> I have a linux NFS server running base 2.4.18 with xfs patches from SGI.
> Seeminly random times, the nfs server will become unresponsive.  On
> a client, doing showmount -e [server] will timeout.  Looking on the
> server shows that the nfs daemons will go into the DW state as shown
> by ps:
> root       123  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
> root       124  0.2  0.0     0    0 ?        DW   Jul25   2:50 [nfsd]
> root       127  0.2  0.0     0    0 ?        DW   Jul25   2:52 [nfsd]
> root       128  0.2  0.0     0    0 ?        DW   Jul25   2:52 [nfsd]
> root       129  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
> root       130  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
> root       131  0.3  0.0     0    0 ?        DW   Jul25   2:55 [nfsd]
> root       132  0.2  0.0     0    0 ?        DW   Jul25   2:53 [nfsd]
> 
>   The load will sit at about 8 which is how many nfsd's that I have
> running.  I thought this was usually caused by a hardware problem, but I
> am able to get the the local filesystems that are exported just fine.
> Any help would be appreciated.
> Thanks
> --David Dougall
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> NFS maillist  -  NFS@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nfs


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs