Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 11 Mar 2006 12:37:03 +0300
From:      "Eugene" <genie@geniechka.ru>
To:        "Kris Kennaway" <kris@obsecurity.org>
Cc:        freebsd-fs@freebsd.org
Subject:   Re: Dump/snapshot problems
Message-ID:  <001e01c644ef$68a76070$1b12c055@genie>
References:  <004c01c644e8$c05c8c70$1b12c055@genie> <20060311090532.GA43378@xor.obsecurity.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi!

----- Original Message ----- 
From: "Kris Kennaway" <kris@obsecurity.org>
To: "Eugene" <genie@geniechka.ru>
Cc: <freebsd-fs@freebsd.org>
Sent: Saturday, March 11, 2006 12:05
Subject: Re: Dump/snapshot problems

> > 1) System hangs or at least not accessible over the network and has to 
> > be
> > physically reset. Last message on console
> > Collecting pv entries -- suggest increasing PMAP_SHPGPERPRO?
> > maxproc limit exceeded by id(0), please see tuning(7) and login.conf(5)
> > Question: what exactly should I tune?
>
> The value of PMAP_SHPGPERPROC in your kernel config, and the maxproc
> limit as described in those two manpages.

Well, I can read -) I would like to know why it fails in virtually unloaded 
system (except for dump) and what parameters/values _in particular_ should 
be tweaked.

> > 2) Dump process dies with message
> > mksnap_ffs: Cannot create //.snap/dump_snapshot: Resource temporarily
> > unavailable
> > dump: Cannot create //.snap/dump_snapshot: No such file or directory
> >
> > dmesg message
> > fsync: giving up on dirty: 0xc57b5d68: tag devfs, type VCHR, usecount 
> > 60263,
> > writecount 0, refcount 960, flags (VV_OBJBUF), lock type devfs: EXCL
> > (count 1) by thread 0xc610f7d0 (pid 87575)
> > dev ad4s1a
> >
> > What is wrong?  What is '0xc57b5d68' (number is repeating)?

> You can ignore the fsync message, it's a side-effect of the snapshot
> operation not succeeding.  Just retry the dump.

That's very comforting but why is it not succeding? And I would really like 
the dump to work automatically (as it did for more than a year before these 
problems started to appear recently).

Eugene 




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?001e01c644ef$68a76070$1b12c055>