jamf process using a lot of CPU

[PeterG]

I have a couple machines that the jamf process seems to use quite a bit of CPU cousing the user to complain.

In one case, i am seeing to jamf processes running. and it is bouncing betweeen 19% and 55% cpu usage. (this one has been running for 5 minutes)

Anyone see this before? Any Ideas on the cause? cure?

Peter

[tlarkin]

what does the jamf log say and what policies are you running on these machines? Do you have any "any trigger" policies running? Are they constantly recon'ing or updating inventory?

---

Thomas Larkin
TIS Department
KCKPS USD500
tlarki at kckps.org
blackberry: 913-449-7589
office: 913-627-0351

[#3]

We had a similar issue several weeks ago. After talking with JAMF support, we think that the server got to a point that there were so many clients checking in so frequently. We created a new every30 policy and changed all our every15's to every30. Tom is also right about the "any trigger", I was told that this is a legacy feature that is bad practice to use.

-Adam

[tlarkin]

I have just a bit over 6,000 clients checking in every 30 minutes randomized by every 15 minutes in my jamf framework. This allows the clients to sporadically space them self out over time so that I don't get too many checking in at once. We also had issues with our database getting really really huge and the mysql daemon would basically append the processors out at 200% (it was a dual core G5 xserve on my first JSS).

Then we migrated to a new intel Xserve running Xeons (dual quads) and I kept the same framework settings and my CPU usage hardly ever spikes over 100% and I have a total of 8 cores in that server.

---

Thomas Larkin
TIS Department
KCKPS USD500
tlarki at kckps.org
blackberry: 913-449-7589
office: 913-627-0351

[PeterG]

We only have about 280 clients in total..

...and it seems that it is only happening on two or three machines.

the logs show nothing unusual (she was complaining around 10:45)

Wed Jun 17 10:04:39 CST02063pb jamf[1154]: Checking for policies triggered by "every15"...
Wed Jun 17 10:11:47 CST02063pb jamf[1191]: Checking for policies triggered by "every15"...
Wed Jun 17 10:11:48 CST02063pb jamf[1191]: Executing Policy Update Inventory Policy...
Wed Jun 17 10:25:06 CST02063pb jamf[2876]: Checking for policies triggered by "logout" for user "jnardone"...
Wed Jun 17 10:26:07 CST02063pb jamf[205]: Checking for policies triggered by "startup"...
Wed Jun 17 10:26:08 CST02063pb jamf[205]: Executing Policy Enforce Management Framework...
Wed Jun 17 10:26:25 CST02063pb jamf[286]: Enforcing Management Framework...
Wed Jun 17 10:26:31 CST02063pb jamf[362]: Checking for policies triggered by "login" for user "jnardone"...
Wed Jun 17 10:45:07 CST02063pb jamf[460]: Checking for policies triggered by "every15"...
Wed Jun 17 10:45:07 CST02063pb jamf[460]: Executing Policy Update Inventory Policy...
Wed Jun 17 11:13:48 CST02063pb jamf[2463]: Checking for policies triggered by "every15"...
Wed Jun 17 11:30:20 CST02063pb jamf[2519]: Checking for policies triggered by "every15"...

Peter

[tlarkin]

well I think you need to duplicate the problem and see what else is running in the background of the machine. You are checking in every 15 minutes but what are the specs of your JSS? Remember that every client checked in is a unicast connection over ssh from the server to the client.