bayes.mutex

Steve Campbell campbell at cnpapers.com
Tue Nov 23 21:38:09 GMT 2004


    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "US-ASCII" character set.  ]
    [ Some characters may be displayed incorrectly. ]

Matt,

Thanks,

Very well spoken. I did discover why I had to run "spamassassin -p ...." as
the user that normally runs spamassassin due to this problem. I had no idea
what a mutex was, but figured it was some kind of lock. It all turns our
that the permissions on this file were improper due to the --linting test I
was doing. After chmod-ding the file, all returned to normal.

Lesson learned

Thanks much

Steve Campbell

----- Original Message -----
From: "Matt Kettler" <mkettler at EVI-INC.COM>
To: <MAILSCANNER at JISCMAIL.AC.UK>
Sent: Tuesday, November 23, 2004 4:13 PM
Subject: Re: bayes.mutex


> At 03:04 PM 11/23/2004, Steve Campbell wrote:
> >I seem to be having a problem with sa-learn due to a file called
bayes.mutex
> >in my bayes directory. Can someone explain what this file is? The
archives
> >has plenty of stuff listing it, but I didn't see anything saying what it
is.
>
> It's a mutex.
>
> Ok, perhaps that's not so helpful if you're not familar with multithreaded
> programing and access locking.
>
> In short, a mutex is a MUTual EXclusion. It's used to lock access to some
> piece of data so you don't run into consistency problems where two
> different threads are both trying to update the same data and one winds up
> stomping on the changes of the other.
>
> (ie: if two threads decide to increment a counter at the same time, both
> read a value of 1, both increment it to 2 and then write back, you get 2.
> If you have locking, then the two threads operate one after the other, and
> you get the right answer of 3.)
>
> In your case, MailScanner probably timed out on a SA when it was trying to
> do an opportunistic expiry and made it leave the mutex laying around.
> sa-learn refuses to update the bayes database because it sees the mutex
and
> thinks that SA is currently working on updating the DB. Check your mail
> logs for "SpamAssassin timed out and was killed" messages. They are evil,
> and should be regarded as a problem and are not a healthy thing for SA's
> bayes subsystem.
>
> In the short term, jack up your spamassassin timeout in mailscanner.conf,
> stop mailscanner, wack the bayes.mutex with rm, run
sa-learn --force-expire
> then restart mailscanner. That should clear up the need for expiry, and
> make it less likely to be killed again.
>
> For a longer discussion of more detailed tricks to stop timeouts, here's
> one of my old posts:
> ----------------------------
>
> In the longer term, here's some suggestions I use on my own MailScanner
> server: (I use all of these together)
>
> 1) Increase the spamassassin timeout in MailScanner.conf. Bring it to 60
> seconds at least, I have mine set to 120.
>
> 2) Set the "Rebuild Bayes Every" parameter in MailScanner.conf. 86400
> seconds is a good start. This makes MailScanner invoke SA's bayes
> housekeeping directly, rather than during a scan of a message.
>
> 3) in /etc/mail/spamassassin/local.cf set: bayes_auto_expire 0. This will
> keep SA from trying to run bayes expires (long and slow) during message
> handling, but relies on #2 above to allow expiry to occur.
>
> 4) I also have a sa-learn --force-expire running as a daily cronjob. I
have
> tested the setup without this measure, and #2 is sufficient to cause
expiry
> to occur. Really this is just a fail-safe to allow expiry to occur even if
> MailScanner's calls fail to run it properly for some reason.
>
> ------------------------ MailScanner list ------------------------
> To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
> 'leave mailscanner' in the body of the email.
> Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
> the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).
>
> Support MailScanner development - buy the book off the website!

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the MAQ (http://www.mailscanner.biz/maq/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!




More information about the MailScanner mailing list