Hello all,

I'm using autofs+nfs combo on all my workstations.
These machine are also used for software testing and crash a -lot-.
Problem is - both nfs and autofs are notorious for going zombie once the
host server dies.

Any idea how I can get the nfs-client (and/or the autofs) to time-out
gracefully / error out, when the nfs server dies instead of just hanging
till the end of time?

- Gilboa


--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, 2008-08-14 at 16:40 +0300, Gilboa Davara wrote:
> Hello all,
>
> I'm using autofs+nfs combo on all my workstations.
> These machine are also used for software testing and crash a -lot-.
> Problem is - both nfs and autofs are notorious for going zombie once the
> host server dies.
>
> Any idea how I can get the nfs-client (and/or the autofs) to time-out
> gracefully / error out, when the nfs server dies instead of just hanging
> till the end of time?
----
what's the host server? RHEL? CentOS?

Is this your issue?
https://bugzilla.redhat.com/show_bug.cgi?id=453094

Craig

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, Aug 14, 2008 at 07:44:30AM -0700, Craig White wrote:
> Subject: Re: Autofs timeout?
> From: Craig White <craigwhite@azapple.com>
> To: For users of Fedora <fedora-list@redhat.com>
> Date: Thu, 14 Aug 2008 07:44:30 -0700
> Reply-To: For users of Fedora <fedora-list@redhat.com>
> Delivered-To: niftyfedora@niftyegg.com
> Return-Path: fedora-list-bounces@redhat.com
> Message-Id: <1218725070.6782.114.camel@lin-workstation.azapple.com>
> Sender: fedora-list-bounces@redhat.com
>
> On Thu, 2008-08-14 at 16:40 +0300, Gilboa Davara wrote:
> > Hello all,
> >
> > I'm using autofs+nfs combo on all my workstations.
> > These machine are also used for software testing and crash a -lot-.
> > Problem is - both nfs and autofs are notorious for going zombie once the
> > host server dies.
> >
> > Any idea how I can get the nfs-client (and/or the autofs) to time-out
> > gracefully / error out, when the nfs server dies instead of just hanging
> > till the end of time?
> ----
> what's the host server? RHEL? CentOS?
>
> Is this your issue?
> https://bugzilla.redhat.com/show_bug.cgi?id=453094

And also does the fstab and autofs mount line contain the intr flag?

intr If an NFS file operation has a major timeout and it is
hard mounted, then allow signals to interupt the file
operation and cause it to return EINTR to the calling
program. The default is to not allow file operations to
be interrupted.

see also "bg" and "soft".

Depending on what software testing is testing these might help.


--
T o m M i t c h e l l
Got a great hat... now what.

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, 2008-08-14 at 12:20 -0700, Nifty Fedora Mitch wrote:
> On Thu, Aug 14, 2008 at 07:44:30AM -0700, Craig White wrote:
> > Subject: Re: Autofs timeout?
> > From: Craig White <craigwhite@azapple.com>
> > To: For users of Fedora <fedora-list@redhat.com>
> > Date: Thu, 14 Aug 2008 07:44:30 -0700
> > Reply-To: For users of Fedora <fedora-list@redhat.com>
> > Delivered-To: niftyfedora@niftyegg.com
> > Return-Path: fedora-list-bounces@redhat.com
> > Message-Id: <1218725070.6782.114.camel@lin-workstation.azapple.com>
> > Sender: fedora-list-bounces@redhat.com
> >
> > On Thu, 2008-08-14 at 16:40 +0300, Gilboa Davara wrote:
> > > Hello all,
> > >
> > > I'm using autofs+nfs combo on all my workstations.
> > > These machine are also used for software testing and crash a -lot-.
> > > Problem is - both nfs and autofs are notorious for going zombie once the
> > > host server dies.
> > >
> > > Any idea how I can get the nfs-client (and/or the autofs) to time-out
> > > gracefully / error out, when the nfs server dies instead of just hanging
> > > till the end of time?
> > ----
> > what's the host server? RHEL? CentOS?
> >
> > Is this your issue?
> > https://bugzilla.redhat.com/show_bug.cgi?id=453094
>
> And also does the fstab and autofs mount line contain the intr flag?
>
> intr If an NFS file operation has a major timeout and it is
> hard mounted, then allow signals to interupt the file
> operation and cause it to return EINTR to the calling
> program. The default is to not allow file operations to
> be interrupted.
>
> see also "bg" and "soft".
>
> Depending on what software testing is testing these might help.
----
in light of the current regression bug in RHEL and NFS, it would seem
that adding the 'intr' option would be useful.

How/where would one add mount options to LDAP auto mounts?

Craig

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, 2008-08-14 at 07:44 -0700, Craig White wrote:
> On Thu, 2008-08-14 at 16:40 +0300, Gilboa Davara wrote:
> > Hello all,
> >
> > I'm using autofs+nfs combo on all my workstations.
> > These machine are also used for software testing and crash a -lot-.
> > Problem is - both nfs and autofs are notorious for going zombie once the
> > host server dies.
> >
> > Any idea how I can get the nfs-client (and/or the autofs) to time-out
> > gracefully / error out, when the nfs server dies instead of just hanging
> > till the end of time?
> ----
> what's the host server? RHEL? CentOS?
>
> Is this your issue?
> https://bugzilla.redhat.com/show_bug.cgi?id=453094
>
> Craig
>

Sorry for the late reply.
Problem with both RHEL, CentOS and Fedora.

Kill the host, the autofs/nfs clients go zombie.

- Gilboa

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list

On Thu, 2008-08-14 at 12:20 -0700, Nifty Fedora Mitch wrote:
> On Thu, Aug 14, 2008 at 07:44:30AM -0700, Craig White wrote:
> > Subject: Re: Autofs timeout?
> > From: Craig White <craigwhite@azapple.com>
> > To: For users of Fedora <fedora-list@redhat.com>
> > Date: Thu, 14 Aug 2008 07:44:30 -0700
> > Reply-To: For users of Fedora <fedora-list@redhat.com>
> > Delivered-To: niftyfedora@niftyegg.com
> > Return-Path: fedora-list-bounces@redhat.com
> > Message-Id: <1218725070.6782.114.camel@lin-workstation.azapple.com>
> > Sender: fedora-list-bounces@redhat.com
> >
> > On Thu, 2008-08-14 at 16:40 +0300, Gilboa Davara wrote:
> > > Hello all,
> > >
> > > I'm using autofs+nfs combo on all my workstations.
> > > These machine are also used for software testing and crash a -lot-.
> > > Problem is - both nfs and autofs are notorious for going zombie once the
> > > host server dies.
> > >
> > > Any idea how I can get the nfs-client (and/or the autofs) to time-out
> > > gracefully / error out, when the nfs server dies instead of just hanging
> > > till the end of time?
> > ----
> > what's the host server? RHEL? CentOS?
> >
> > Is this your issue?
> > https://bugzilla.redhat.com/show_bug.cgi?id=453094
>
> And also does the fstab and autofs mount line contain the intr flag?
>
> intr If an NFS file operation has a major timeout and it is
> hard mounted, then allow signals to interupt the file
> operation and cause it to return EINTR to the calling
> program. The default is to not allow file operations to
> be interrupted.
>
> see also "bg" and "soft".
>
> Depending on what software testing is testing these might help.

Thanks!

I'll give it a shot.

- Gilboa

--
fedora-list mailing list
fedora-list@redhat.com
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list