SYSERR(root): readqf: cannot open ./df*

Jim Levie jim at ENTROPHY-FREE.NET
Thu Dec 12 15:49:35 GMT 2002


On Thu, 2002-12-12 at 06:04, Kevin Spicer wrote:
> I've been having this problem too, but on an older version of Mailscanner
> (3.23).  I didn't think it was a problem until I started getting reports
> from users of mail delivered with '<<<No Message Collected>>>' instead of
> the body.  The ID's of these messages tied up with those in the error log.
> I also noted error messages like the following...
>
>
>
>
> Dec  9 10:53:59 mailhost sendmail[19633]: gB9Arxid019633: SYSERR(root):
> cannot flock (/etc/mail/access.db, fd=7, type=1, omode=100000, euid=0): No
> locks available
>
Is the failed lock always for access.db?
>
> I've been doing a bit of research into this and I _thought_ I'd found out
> what causes it.  According to the release notes for sendmail 8.12.5 (
> http://www.sendmail.org/8.12.5.html ) there is an issue with file locking
> under 2.4 linux kernels.  Upgrading to sendmail 8.12.6 did not help, so I
> recompiled sendmail from the source rpm with -DHASFLOCK=0 - this means that
> sendmail uses fcntl rather than flock for locking.  It appears that the
> Mandrake rpms don't do this by default (perhaps no-one noticed because
> postfix is the default?.
>
Well there could be a problem with flock on a 2.4 kernel, but it sounds
like something else is going on since changing to fcntl didn't fix it.
I've not see any problems like this on any of my RedHat mail servers,
you'd think that one that  handles some 250K messages per day would
encounter the problem at some point.

What other services does this box provide? What kind of mail load does
it see? Are there any automated processes futzing with access.db?
--
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=
The instructions said to use Windows 98 or better, so I installed RedHat
   Jim Levie                                 email:
jim at entrophy-free.net



More information about the MailScanner mailing list