send mail problem..

[Varghese Mathew]

Hi Everyone,

It seems our V2500 server is running with lot of sendmail processes and somehow lot of "0"byte files with sticky bit enabled are being created in the directory /var/spool/sockets/pwgr .Approx 7 files are being created per one minute. Can anyone tell me what is this all about.?

If i restart this sendmail will this affect anything else in the system like telnetd or the Informix database running in the system.As far as i know we don't use the sendmail for any purpose.

Thanx in advance
Cheers !!!
Mathew

[Herve BRANGIER]

Hi,

1) This files are not sticky bit enabled. The first
letter (s in your case) get the file's type.
s is for sockets file.

2) I don't know why your sendmail seems to fork
every time ... But if you don't want to receive
or send mail outside your server you can stop
sendmail. Local mail already works... (modifiy
/etc/rc.config.d/mailsrvs and put sendmail_server=0).

HTH

Herv?

[Sachin Patel]

Hi Varghese,
I don't know th answer of first question. For second question if you are not sending any mail from that V-class system you can shutdown sendmail. System can still receive the mail.

Sachin

[Bill Thorsteinson]

Check the option that controls how often you
retry the mail queue. If it is too frequent, then
you end up with a bunch of sendmail processes
trying to send mail at the same time. They quickly
get tied up waiting for locks.

The command mailq will tell you how many messages
are waiting to be be sent.

If you have a big backlog, try turning off automatic
retrys and flushing the que mainually.

If this doesn't solve your problem try check the crontabs
for entries starting "* *". They will run once a minute.

[Ralf Hildebrandt]

Sendmail is am major security hole.

A) Consider the latest patches. What you're seeing might be the effect of a root exploit
B) Consider removeing sendmail altogether: www.postfix.org
Actually, that is what HP themselves do. Only the customers are left suffering (HP uses Postfix exclusively on their MX'es!)