Sudden jump in CPU???

[Uncle Michael]

I happened to notice the CPU usage of my HS box was over 50% and holding steady, which is quite unusual. It's usually well below 10% and oscillates up and down with a fairly regular pattern.

When I investigated further I discovered that it had risen suddenly at about 5:03 this morning. No other measures changed, and HS did not appear to be doing anything special when the jump occurred. (We and the house were in asleep mode.) Here's the HS CPU usage (in %):

9/25/2012 4:48:36 AM 3.98
9/25/2012 4:50:37 AM 4.15
9/25/2012 4:52:42 AM 3.60
9/25/2012 4:54:47 AM 3.65
9/25/2012 4:56:52 AM 5.09
9/25/2012 4:58:57 AM 3.59
9/25/2012 5:01:05 AM 5.16
9/25/2012 5:03:30 AM 51.63
9/25/2012 5:05:46 AM 51.00
9/25/2012 5:07:36 AM 52.41
9/25/2012 5:09:49 AM 51.22
9/25/2012 5:11:59 AM 52.47
9/25/2012 5:13:51 AM 50.56
9/25/2012 5:16:31 AM 51.37
9/25/2012 5:18:13 AM 52.58

When I attempted to restart HS it hung and I had to force it closed with Task Manager. When I restarted it it was clearly not behaving normally and triggered it's own restart routine. That also failed to get it running properly, and it restarted on its own again. After the third restart it seemed to be back to normal and appears to be running normally since.

Any ideas would be appreciated.

[Wadenut]

I've noticed this behavior with 2.5.0.58, but you appear to be on 2.4.0.11 according to your sig.

I've also noticed, that if I leave things alone, so long as events and such are operating properly, the CPU often returns to normal within about 24 hours.

[Uncle Michael]

I've noticed this behavior with 2.5.0.58, but you appear to be on 2.4.0.11 according to your sig.

Yes, I've been using 2.4.0.11 since it was introduced. This is the first time I've seen this behavior.

I've also noticed, that if I leave things alone, so long as events and such are operating properly, the CPU often returns to normal within about 24 hours.

Interesting. I intervened after about 9 hours. I was noticing some execution delays, so I don't know if I'd be up for waiting a full 24 hr to see what happened. If it happens again I'll see if I can be that patient. Thanks.

[Wadenut]

If you have delays there's no reason to wait. I don't. If HS isn't doing its job, it needs a good (re)boot.
Have you looked at the Event Execution Delay script in the library? It might give you an early warning.

[Uncle Michael]

If you have delays there's no reason to wait. I don't. If HS isn't doing its job, it needs a good (re)boot.
Have you looked at the Event Execution Delay script in the library? It might give you an early warning.

Yes, in fact I have, but couldn't think of a reason to try it. I have some classes of events that sporadically execute slowly (possibly a plug-in conflict), but I don't believe your script will detect that kind of behavior.

However, if this high CPU condition becomes a recurring problem I probably will give it a try.

[Rotech]

I had that issue not so long ago, I thought it was HS, but it ended up being one of my sims memory cards (Ram) on mother brd going bad. Took it to the shop and they found the problem.

[Wadenut]

Yes, in fact I have, but couldn't think of a reason to try it. I have some classes of events that sporadically execute slowly (possibly a plug-in conflict), but I don't believe your script will detect that kind of behavior.

However, if this high CPU condition becomes a recurring problem I probably will give it a try.

In fact it won't. It attempts to measure the time between when the script should have been triggered and the time it actually begins execution. More a measure of how HS is performing in general.

[Pete]

Works for me here.