qmail not sending mail and queuing up

[sillyhead]

Hi, thanks for this forum. I'm new here. I've searched and have not found my problem yet.

Here's the background:

I started working on this system about 1.5 years ago. Qmail was already installed, and has been working fine until now. I'm able to add aliases, virtual domains, etc, and it has been chugging along fine. I don't really mess with the machine other than to add an occasional alias. Noone else has access to it.

A couple of days ago, mail started queueing up, and wouldn't go out. I tried restarting it, but no joy.

I didn't really look at the queue then, because I thought that a certain user may be having his own problems, and I created an alias for myself, tested it, and it went through okay.

Later, the certain user started getting his mail, from July on. Yay, miller time, right? Nope.

Today, I checked on the system and it was queueing mail again, and no mail will go through at all. I thought it might be a dns problem, but I'm able to surf from the box, using domain names and IP addresses, and I can get to it from anywhere.

I'm at a loss. I'm kind of rusty, because I haven't done a lot of this work lately, but I've searched everywhere for the answer, and can't find it.

I use qmailctl restart to restart it, but I've noticed that I get the repeating @4000000044d55b6531921234 alert: cannot start: qmail-send is already running
errors in my log file, whether I use that or the svc -d/-u /service/qmail-send command.

I know that I cannot ever get it to completely shut down at any point, because I tried to, in order to try to use the qmHandle script.

So, bottom line, it's one of those "it's suddenly not working" things, only this time I have noone to ask repeatedly, "You sure you didn't change anything?" because I haven't even been on the box in months and noone else has access to it.

Any clues for me?
Gratefully,
C Dale

 

[sillyhead]

More information:

I'm not able to receive mail at that server, either, I don't think, or at least not consistently. I sent a test mail to see if it would register in the logs, but, nothing.

I'd checked stats earlier, and it was up, but I'm getting this right now:

[root@mailbox qmail-send]# qmailctl restart
Restarting qmail:
* Stopping qmail-smtpd.
* Sending qmail-send SIGTERM and restarting.
* Restarting qmail-smtpd.

waited a few minutes

[root@mailbox qmail-send]# qmailctl stat
/service/qmail-send: up (pid 7590) 484 seconds
/service/qmail-send/log: down 2690 seconds, normally up
/service/qmail-smtpd: up (pid 7087) 624 seconds
/service/qmail-smtpd/log: up (pid 7088) 624 seconds
messages in queue: 10
messages in queue but not yet preprocessed: 0

So, it looks like the logging isn't starting, which is why I'm not seeing anything go through, however, when I check the queue, the mail I've sent myself from outside the network is not there. However, there are some mails that are queuing up very sporadically, it seems, from inside the network, and out.

I'm about ready to do a complete reinstall or something drastic, because I can't figure out what's going on.
Any help would be tremendously appreciated.
Thanks,
C Dale

 

[sillyhead]

More info:

I can telnet to port 25 of the server, and it takes all the commands till I get to the end, and does not respond to "." I don't know if this makes a difference or not.

 

[sillyhead]

Still more info:

qmail-showctl shows:

qmail home directory: /var/qmail.
user-ext delimiter: -.
paternalism (in decimal): 2.
silent concurrency limit: 120.
subdirectory split: 23.
user ids: 501, 502, 503, 0, 504, 505, 506, 507.
group ids: 501, 502.

badmailfrom: (Default.) Any MAIL FROM is allowed.

bouncefrom: (Default.) Bounce user name is MAILER-DAEMON.

bouncehost: (Default.) Bounce host name is xxxx.com.

concurrencylocal: (Default.) Local concurrency is 10.

concurrencyremote: (Default.) Remote concurrency is 20.

databytes: (Default.) SMTP DATA limit is 0 bytes.

defaultdomain: Default domain name is xxxx.com.

defaulthost: (Default.) Default host name is xxxx.com.

doublebouncehost: (Default.) 2B recipient host: xxxx.com.

doublebounceto: (Default.) 2B recipient user: postmaster.

envnoathost: (Default.) Presumed domain name is xxxx.com.

helohost: (Default.) SMTP client HELO host name is xxxx.com.

idhost: (Default.) Message-ID host name is xxxx.com.

localiphost: (Default.) Local IP address becomes xxxx.com.

locals:
Messages for xxxx.com are delivered locally.
others listed

me: My name is xxxx.com.

percenthack: (Default.) The percent hack is not allowed.

plusdomain: Plus domain name is xxxx.com.

qmqpservers: (Default.) No QMQP servers.

queuelifetime: (Default.) Message lifetime in the queue is 604800 seconds.

rcpthosts:
SMTP clients may send messages to recipients at xxxx.com.
others listed

morercpthosts: (Default.) No effect.

morercpthosts.cdb: (Default.) No effect.

smtpgreeting: (Default.) SMTP greeting: 220 anonyms.com.

smtproutes: (Default.) No artificial SMTP routes.

timeoutconnect: (Default.) SMTP client connection timeout is 60 seconds.

timeoutremote: (Default.) SMTP client data timeout is 1200 seconds.

timeoutsmtpd: (Default.) SMTP server data timeout is 1200 seconds.

virtualdomains:
lists them

concurrencyincoming: I have no idea what this file does.

defaultdelivery: I have no idea what this file does.

 

[sillyhead]

And STILL more info.

Since I worked as tech support for a while, and know that They Always Lie (kinda like House says, but I knew it before that show ever came out), I kept asking myself, "What did I do that may have messed things up?" Well, before the problem started, the answer is "NOTHING," but I did reboot the machine today, which is the first time I've done that since I started working on it.

I wonder if it's possible that there is something that was run before I started working there that kept the logs going, that isn't in the init scripts. I tried starting it using some lines from the LWQ site:
svc -u /service/qmail-send /service/qmail-send/log

But that didn't help either. It's still queueing mail, and I'm still seeing nothing in my logs, except the repeating
@4000000044d586783940269c alert: cannot start: qmail-send is already running

svc -u /service/qmail-send /service/qmail-send/log
[root@mailbox bin]# qmailctl stat
/service/qmail-send: up (pid 12659) 1 seconds
/service/qmail-send/log: up (pid 12647) 3 seconds
/service/qmail-smtpd: up (pid 11729) 1917 seconds
/service/qmail-smtpd/log: up (pid 11731) 1917 seconds
messages in queue: 31
messages in queue but not yet preprocessed: 0

-gurgle-

Oh well, if nothing else, I have documented what all I'm doing here, so I can go back and look at it and bonk myself on the head when the small and very obvious thing I'm overlooking comes to light.

 

[sillyhead]

Okay, final info. There were two instances of qmail running, and starting each time I used qmailctl or rebooted the machine, because two init scripts were in /service. The queue problem is because there are a lot of bad addresses in a script which runs from another machine to send out weekly bulletins. So, if you're googling around and having the same confusing problems I had, check /service and make sure you don't have two sets of scripts initiating qmail.

Wee!