Re: [CentOS] NFS options in fstab

2011-12-29 Thread Daniel Bird
On 29/12/2011 15:23, Marc Deop wrote:
> On Thursday 29 December 2011 14:35:34 Daniel Bird wrote:
>> retry=n   The  number of minutes to retry an NFS mount operation in the
>> foreground or background before giving up.  The default
>> value for forground mounts is 2 minutes.  The default
>> value for background mounts is 1 minutes, which is  roughly
>> one week.
>>
>> So retry=2 would give the same behavior on background mounts as
>> foreground mounts I reckon.
> Have you tried this? I'd love to know if it really works.
I've got it down to a reasonable timeout (1m 30 sec) now with these options

proto=udp,timeo=1,retrans=0,retry=1,intr,soft,bg

with proto=tcp it's about 4mins

>
> You could as well try automount with your nfs shares...
Yes, that is an option. We use it for some services.

Cheers

D

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] NFS options in fstab

2011-12-29 Thread Marc Deop
On Thursday 29 December 2011 14:35:34 Daniel Bird wrote:
> retry=n   The  number of minutes to retry an NFS mount operation in the
> foreground or background before giving up.  The default
> value for forground mounts is 2 minutes.  The default
> value for background mounts is 1 minutes, which is  roughly
> one week.
> 
> So retry=2 would give the same behavior on background mounts as
> foreground mounts I reckon.

Have you tried this? I'd love to know if it really works.

You could as well try automount with your nfs shares...


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] NFS options in fstab

2011-12-29 Thread Daniel Bird
On 29/12/2011 14:25, Daniel Bird wrote:
> Hi All,
> I've a quick question I'm hoping someone can help out on; One of our NFS
> servers has an intermittent hard lock problem under high load. This is a
> CentOS 5.7 box. It's due for replacement sometime next year. However...
>
> The issue is when this box dies the boxes that mount the NFS exports
> have trouble with the mount and some processes time out casing the
> system mounting the file system to become unresponsive also.  Of course
> un-monuting the file system with -l fixes it.
>
> This is an example of our current mount options in fstab
>
> hostname:/share1/ /mountpoint nfs rw,intr,soft,bg 0 0
>
> The question is, what (other) options should we be using to allow the
> mounting server to ignore/fail/timeout the mount, or at least continue
> normal processing, when the NFS server packs up?
Ahhh, I should read the man pages more carefully!!

Just found this

retry=n   The  number of minutes to retry an NFS mount operation in the
foreground or background before giving up.  The default
value for forground mounts is 2 minutes.  The default
value for background mounts is 1 minutes, which is  roughly
one week.

So retry=2 would give the same behavior on background mounts as
foreground mounts I reckon.

> FYI we can't move to NFS4 since we've some old boxes running Solaris 8 
> mounting this filesystem also, which cannot be upgraded Solaris 10 to
> support v4.
>
> All the best
>
> Dan
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] NFS options in fstab

2011-12-29 Thread Daniel Bird
Hi All,
I've a quick question I'm hoping someone can help out on; One of our NFS
servers has an intermittent hard lock problem under high load. This is a
CentOS 5.7 box. It's due for replacement sometime next year. However...

The issue is when this box dies the boxes that mount the NFS exports
have trouble with the mount and some processes time out casing the
system mounting the file system to become unresponsive also.  Of course
un-monuting the file system with -l fixes it.

This is an example of our current mount options in fstab

hostname:/share1/ /mountpoint nfs rw,intr,soft,bg 0 0

The question is, what (other) options should we be using to allow the
mounting server to ignore/fail/timeout the mount, or at least continue
normal processing, when the NFS server packs up?
FYI we can't move to NFS4 since we've some old boxes running Solaris 8 
mounting this filesystem also, which cannot be upgraded Solaris 10 to
support v4.

All the best

Dan

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos