A few questions I can't find in archive...

[MailScanner]

OK, I did a cull of the outbound queue by killing all the NDR's with a
retry count over four (see other thread for script). This had little or
no effect on the CPU utilisation.

Since it was very quiet (Sunday morning) I stopped MailScanner and did a
debug run:

Starting MailScanner...
In Debugging mode, not forking...
debug: Score set 0 chosen.
debug: running in taint mode? no
debug: ignore: test message to precompile patterns and load modules
debug: using "/usr/share/spamassassin" for default rules dir
debug: using "/etc/mail/spamassassin" for site rules dir
debug: using "/etc/MailScanner/spam.assassin.prefs.conf" for user prefs
file
debug: bayes: 16862 tie-ing to DB file R/O
/root/.spamassassin/bayes_toks
debug: bayes: 16862 tie-ing to DB file R/O
/root/.spamassassin/bayes_seen
debug: bayes: found bayes db version 2
debug: Score set 3 chosen.
debug: Initialising learner
debug: is Net::DNS::Resolver available? yes
debug: trying (3) google.com...
debug: looking up MX for 'google.com'
debug: MX for 'google.com' exists? 1
debug: MX lookup of google.com succeeded => Dns available (set
dns_available to hardcode)
debug: is DNS available? 1
debug: all '*From' addrs: ignore at compiling.spamassassin.taint.org
debug: running header regexp tests; score so far=0
debug: running body-text per-line regexp tests; score so far=2.077
debug: bayes corpus size: nspam = 1080753, nham = 99884
debug: uri tests: Done uriRE
debug: tokenize: header tokens for *F = "U*ignore
D*compiling.spamassassin.taint.org D*spamassassin.taint.org D*taint.org
D*org"
debug: tokenize: header tokens for *m = " 1078048471 13698
spamassassin_spamd_init "
debug: bayes token 'N:H*m:NNNNN' => 0.0309530650921124
debug: bayes token 'somewhat' => 0.0463561582202703
debug: bayes token 'N:H*m:NNNNNNNNNN' => 0.1269607132118
debug: bayes token 'H*F:D*org' => 0.150210214552688
debug: bayes: score = 0.00387031687925626
debug: Syncing Bayes journal and expiring old tokens...
debug: lock: 16862 created
/root/.spamassassin/bayes.lock.NMIBWKMS1.16862
debug: lock: 16862 trying to get lock on /root/.spamassassin/bayes with
0 retries
debug: lock: 16862 link to /root/.spamassassin/bayes.lock: link ok
debug: bayes: 16862 tie-ing to DB file R/W
/root/.spamassassin/bayes_toks
debug: bayes: 16862 tie-ing to DB file R/W
/root/.spamassassin/bayes_seen
debug: bayes: found bayes db version 2
debug: synced Bayes databases from journal in 1 seconds: 945 unique
entries (1050 total entries)
debug: bayes: 16862 untie-ing
debug: bayes: 16862 untie-ing db_toks
debug: bayes: 16862 untie-ing db_seen
debug: bayes: files locked, now unlocking lock
debug: unlock: 16862 unlink /root/.spamassassin/bayes.lock
debug: lock: 16862 created
/root/.spamassassin/bayes.lock.NMIBWKMS1.16862
debug: lock: 16862 trying to get lock on /root/.spamassassin/bayes with
0 retries
debug: lock: 16862 link to /root/.spamassassin/bayes.lock: link ok
debug: bayes: 16862 tie-ing to DB file R/W
/root/.spamassassin/bayes_toks
debug: bayes: 16862 tie-ing to DB file R/W
/root/.spamassassin/bayes_seen
debug: bayes: found bayes db version 2
debug: bayes: expiry check keep size, 75% of max: 112500
debug: bayes: token count: 12247148, final goal reduction size: 12134648
debug: bayes: First pass?  Current: 1078048472, Last: 1073760779, atime:
43200, count: 108579, newdelta: 386, ratio: 111.758701037954
debug: bayes: something fishy, calculating atime (first pass)

It hung on this line and had to be killed manually. Further
investigation on google showed that this is a SA problem
(unsurprisingly).

We will bring a second server on-line for redundancy soon, at which
point I will most likely upgrade this one to Fedora Core-1 and rebuild
the SA components from scratch.

Bart...

-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK] On
Behalf Of MailScanner
Posted At: 27 February 2004 22:24
Posted To: MailScanner
Conversation: A few questions I can't find in archive...
Subject: Re: A few questions I can't find in archive...


 
All the trimmings: tmpfs, caching DNS, local lookups.

We do have a large (20k) number of messages in the outbound queue.

Bart...

-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK] On
Behalf Of Raymond Dijkxhoorn Posted At: 27 February 2004 22:07 Posted
To: MailScanner
Conversation: A few questions I can't find in archive...
Subject: Re: A few questions I can't find in archive...


Hi!

> Any (dis)advantages in HyperThreading? We're running a fairly busy 
> (72k
> msg/day) dual Xeon 2.4 2GB RAM with HT on RH9 / MS 4-26.7 / SA 2.63 / 
> Clam 0.65
>
> In top the MailScanner process continuously has near 100% of CPU 2. 
> The idle percentage hunts around between one of the virtual CPU's at 
> near 0% and the rest roughly equal. Sometimes all four are roughly 
> equal. Our CPU utilisation as showing in Mailscanner-MRTG (older
> version) is higher than we would like it to be.

72K is not that many for a dual xeon, sounds strange. Are you using
tmpfs also ? Do you have a RBL lookups on a local server or all on the
remote RBL servers? Do you run a caching DNS on the box itself ?

Bye,
Raymond.