DA turns off every night

[kamsel]

Hi,

For three last days DA turns off by itself every night. Apache, exim, proftpd etc. are working but....DA panel just turns off and I am not able to access it through http://domain.com:2222.
Any ideas why? I need to reboot server to make it work next day..
Please replay.

Thanks

[kamsel]

Also, this won't turn on DA also:

kam# pwd
/usr/local/directadmin/scripts
kam# ./directadmin start
Starting DirectAdmin: [ OK ]

And this is nto working too:

kam# pwd
/usr/local/etc/rc.d
kam# ./directadmin start
Starting DirectAdmin: [ OK ]
kam# ./directadmin
Usage: {start|stop|status|reload|restart}
kam# ./directadmin status
directadmin (pid 188 4392 4398 4412 4415 4420 ) is running...

[thuskey]

Typically if the status command shows it running, it means it is running. Did you check "netstat -a" to see if the service was listening for connections?

tcp 0 0 *:2222 *:* LISTEN

and did you do "top -c" to make sure directadmin wasn't defunct or runnig 100% before resetting the server?

[kamsel]

I do not understand this part:


and did you do "top -c" to make sure directadmin wasn't defunct or runnig 100% before resetting the server?

Now I guess it is too late to check it. There is nothing about it in README files or whatever that I need to do something before rebooting. I was doing reboots without any problems before, but for few days from now it keeps turning off.

[thuskey]

Sorry, top is just a command that outputs similar information to the ps command, but in a continuous loop. It shows more information about the status of the running processes than ps, and is a great tool for finding out whom is eating up all the server resources.

[kamsel]

Oh my god, I just saw that:


6894 root 54 0 2308K 1648K RUN 108:34 98.78% 98.78% directadmin


What is going on and why ???

[thuskey]

It's not a big issue that it hits 90+% unless it stays there and never falls. I've seen two other threads I think in the last week of people having an issue of directadmin pegging out at 90+% and never falling back down. I'm not sure if a cause or solution has yet been identified.

Rather than rebooting the server tomorrow, you may try to "kill -9 {pid}" the pid of the directadmin service. To find that pid, you would do a "ps -e | grep directadmin", which should result in something like:

30566 ? 00:00:01 directadmin
3200 ? 00:00:00 directadmin
3201 ? 00:00:00 directadmin
3249 ? 00:00:00 directadmin
3252 ? 00:00:00 directadmin
3253 ? 00:00:00 directadmin

usually killing off the top one will kill them all off. So in this case, I would only do a "kill -9 30566"

Then you should be able to restart it via your normal /etc/rc.d/init.d/directadmin start command (on RedHat).

[kamsel]

Hi,

Thanks for replay.

Look at this:



Processor Name Intel(R) Pentium(R) 4 CPU 2.80GHz (2813.54-MHz 686-class CPU)
Total Memory 1019.07 MB
Free Memory 38.4766 MB
System Uptime 0 Days, 19 Hours and 57 Minutes

_______________________________
kam# uptime
2:36PM up 19:56, 2 users, load averages: 1.00, 1.00, 0.98
kam# ps -e | grep directadmin
kam#


And here is after reboot:



kam# uptime
2:43PM up 3 mins, 2 users, load averages: 0.04, 0.09, 0.04
kam# ps -e | grep directadmin


after executing top command:
188 root 10 0 2292K 1108K wait 0:00 0.00% 0.00% directadmin





Seems to be ok, but it will start again within few minutes or hours - the eating memory thing.

[DirectAdmin Support]

Hello,

Next time you see it doing that, take note of the process number (ie: 188) and run the following:


kill -USR1 188

That sends a signal to the process to dump it's current location into the /usr/local/directadmin/error.log file.

John