upcp stalled(don't beat me, I tried to search)

[schwim]

Hi there guys,

Before you tar & feather me, the search system tells me that my search is under 3 words, regardless of the length of my search term.

I've waited in hopes that it would finally kill the process, but it's not. upcp has been in action for a very long time now. When looking at the processes, I find:

9806 (upcp) /usr/bin/perl / cPanel Update (upcp) - Master
9904 (upcp) /usr/bin/perl / cPanel Update (upcp) - Slave

1) Which process do I want to kill
2) How do I kill it?

Thanks in advance for any help you can provide, and I'm sorry for the question I'm sure has been asked, but I tried searching.

thanks,
json

[nyjimbo]

Do you see anything else running below it that looks weird, usually they are ganged pretty close on a "ps ax". I just did an upgrade to Stable 11 on a freebsd box and it sat on one thing for about 20 minutes but then took off running.

How long has it been just sitting?

[schwim]

Hi there Jimbo, and thanks for the reply.

Mine's been running for just over two weeks now.

thanks,
json

[nyjimbo]

Hi there Jimbo, and thanks for the reply.

Mine's been running for just over two weeks now.

thanks,
json

Your upcp update has been running for two weeks without ending ????

[schwim]

that is correct. Every night I get an email stating that it's already running, and I should let it finish or use --force to force a new update.

I just don't know how to resolve the issue.

thanks,
json

[Infopro]

From a post here somewhere.

Kill stopped jobs
Type 'jobs' to see a list of the jobs and 'kill %job-number' to stop them.
Or, kill fg %1

Or go to WHM, Show Current CPU usage, find 9904 and click it. Then click kill. Then do 9806

Should work.

The term kill process comes up with many results. If you use one word under 3 letters add a wild card.

kil*

Even gives me results on the forums.

Google, 2 mil.
http://www.google.com/search?hl=en&q...+process+linux

Only thing I'm not sure of is why you have two running and why its been running for weeks without a trouble ticket sent to support?

Good luck.

[schwim]

From a post here somewhere.

Kill stopped jobs
Type 'jobs' to see a list of the jobs and 'kill %job-number' to stop them.
Or, kill fg %1

Or go to WHM, Show Current CPU usage, find 9904 and click it. Then click kill. Then do 9806

Should work.

The term kill process comes up with many results. If you use one word under 3 letters add a wild card.

kil*

Hi there infopro, and thanks for the reply.

Unless I'm mistaken in regards to the implementation, I seem to have a problem printing the jobs currently running. Here is my output, which I think you'll agree is lacking

Even gives me results on the forums.

My search provides this error. As you can see, I've met the three word requirement.

Google, 2 mil.
http://www.google.com/search?hl=en&q...+process+linux

I can kill a process. I need to know which processes are the necessary process to kill, and was wondering if I had to do it a certain way for the WHM scripts.

Only thing I'm not sure of is why you have two running and why its been running for weeks without a trouble ticket sent to support?

Ours is not to wonder why, but to copulate and multiply.

thanks,
json

[schwim]

I killed both instances(master & slave) of the process, and will see how it goes.

thanks,
json

[Infopro]

In this image, http://personal.schwim.net/images/sh...nel_search.jpg

The word to is listed. ANY word with two letters like this, add a * That's the prob.

[dragon2611]

killall upcp

usally fixes that for me

[schwim]

Thanks very much for that infopro. That should solve a lot of my search problems, as I'd seen it more than once. I've used vbulletin a lot, and never came across that problem.

I killed the processes, and it did run last night, but It's got more errors than usual. Mail:SPF is normal, but now PerlMagick is failing it's install.

At the end of my upcp, was this:

Integrity check failed. Running auto update & repair.. please wait...Checking for a running upcp ... still running ... waiting 60 seconds ... ...Done

and not the normal end up upcp, so something's gone wrong there.

Could you tell me what I need to do to shut down all updates for cPanel & WHM?

thanks,
json

[koolcards]

that is correct. Every night I get an email stating that it's already running, and I should let it finish or use --force to force a new update.
I just don't know how to resolve the issue.
thanks,
json

Have you tried forcing an update?

/scripts/upcp --force

Could you tell me what I need to do to shut down all updates for cPanel & WHM?

You can set WHM Main >> Server Configuration >> Update Config to "Never Update"

[Infopro]

Instead, you should set to manual updates, and do them when you can sit there and keep an eye on the update as it happens. Via CLI or thru WHM. I've done it that way myself for years.

You might also want to check this list to make sure all is ready for cPane11.

http://forums.cpanel.net/showthread.php?t=66166



schwim at this point, you might do best to contact cPanel now and let the pros take a look for you. Sounds like you've got amess on your hands.

[Infopro]

About PerlMagick, check this thread. (note the search term for future ref)

http://forums.cpanel.net/showthread....ght=PerlMagick

[schwim]

You can set WHM Main >> Server Configuration >> Update Config to "Never Update"

Hi there koolcards, and thanks for the reply.

I set it to manual, per infopro's suggestion, but I'm wondering: If I want to update manually, do I now need to start updates via CLI, or is there a WHM panel for updating maually?

thanks,
json