I try to start crond but when i run crond status I received

crond dead but subsys locked

I deleted /var/lock/subsys/crond but still the same

In the /var/log/message I found Mar 21 18:09:15 simco3 squid[4797]: error: bind: Address already in use
Mar 21 18:09:15 simco3 squid[4797]: fatal: Bind to port 22 failed: Transport endpoint is not connected.
Mar 21 18:09:15 simco3 crond: crond startup succeeded


Any suggestion?

From what it looks like, you have multiple things going on.

1. Cron is running, short of the 4th line saying cron died, but no 4th line present.

2. Squid is trying to do something that is already being used, my guess would be by sshd.

If you're not doing squid proxy type stuff, then I'd suggest turning that service off.

I deleted squid

Now in the /var/log/messages i have only:
Mar 21 18:09:15 simco3 crond: crond startup succeeded

but /etc/init.d/status return
crond dead but subsys locked

I don't know what's happent.
The server worked ok for more then a half of year but Monday morning i received all that problem.

Instead of relying on whatever it is you're relying on ( i have no idea what /etc/init.d/status is ), ps auxw | grep cron

If you see it, it's running.

You might also want to tail -f /var/log/cron as well as tail -f /var/log/messages when you go to start-up/restart cron and provide some more information.

Sorry I mean:
/etc/init.d/crond status (get the crond staus) ant it return "crond dead but subsys locked"

I run ps auxw | grep cron and no process running.

I run tail -f /var/log/messages when I tried to start crond but everithing looks ok:
Mar 22 16:24:42 simco3 crond: crond startup succeeded

tail -f /var/log/crond/info , tail -f /var/log/crond/error , tail -f /var/log/crond/error , don't return anything.

I run top when I tried to start crond and I see the process running for a few seconds.

Everithing is OK but It don't work.

My only guess is that it's dying because it can't write to the logs that it needs to write to, that's why you're also not seeing any of the error logs.

Check permissions on the logs, I'm guessing that cron should be run by root, since I haven't run Mandrake in a really long time, I don't remember.

Another possibility might be looking at the startup script and how it's called...start crond on the command line and see if you get any error messages that way.

I solved the problem:
I don't know what was wrong. The logs was OK and when I tried to start manually still no error.

In this case I compared the crond application from the server with problem with crond application from another server. I saw that the size and last modified date was different.

On the server with problem
*crond | 12000 | Feb 28 10:00 (the hour and size are from my memory)

On the good server
*crond | 23160 | Feb 13 2003

The problem appeared on Mar 21 maybe when I restarted the server.
I think that the problem comes when I made an system update and when I restarted the system I see it.

Thank you Technoslave for all your support.

When I try ro restart squid it displays the following:

squid dead but subsys locked
2005/04/21 12:51:29 | parse ConfigFile: line 2174 unrecognized: 'delay_pools 0'
squid: ERROR: no running copy

What can I do to fix this?

It looks like you have some problem in your squid configuration. If you made any recent changes look at them and try to find where the problem is.
Usually when a problem appear you must look at the most recent changes.

That's why I'm asking here. By the way I had already solved my own problem I deleted the delay pools on the last line of squid.conf.tnx