perfSONAR User Q&A and Other Discussion

[Pete Siemsen <>]

Thanks for all the replies. To answer a few questions...

> Just to check if this is perhaps OS specific, are you machines on centOS7 or still on SL6/or something else ?

I'm running CentOS 6.9.

> Do you have any suggestion about the rest of the processes?

perfsonar-1850$ ps -ef | wc -l

1124

perfsonar-1850$ ps -ef | grep owampd | wc -l

313

perfsonar-1850$ ps -ef | grep pscheduler | wc -l

282

perfsonar-1850$ ps -ef | grep powstream | wc -l

257

perfsonar-1850$ ps -ef | grep defunct | wc -l

254

perfsonar-1850$ ps -ef | grep http | wc -l

63

> Please share the type of pS installation you have / a test-point or a toolkit deployment

It's a toolkit, network install.

> What else is on the system? I assume esmond.

Just the vanilla install.

> Does it have maddash as well? That could also put some pressure on httpd.

No maddash.

From Marian's commands...

perfsonar-1850$ ps -eTf | grep 501 | grep "powstream" | wc -l

4

perfsonar-1850$ ps -eTf | grep 501 | grep httpd | wc -l

8

perfsonar-1850$ ps -eTf | grep 501 | grep owamp | wc -l

0

perfsonar-1850$ ps -eTf | grep 501 | grep runner | wc -l

1

perfsonar-1850$ ps -eTf | grep 501 | grep pscheduler | wc -l

4

perfsonar-1850$ ps -eTf | grep 501 | grep "\[run\]" | wc -l

1

perfsonar-1850$ ps -eTf | grep 501 | wc -l

15

The system is set to selfupdate, and is running 4.0.2. I suppose I should consider upgrading the OS to CentOS 7.

I rebooted again, and the warning about too many threads is gone. I expect it to reappear in a few days. Perhaps 2000 threads is reasonable, and I should just raise the Nagios threshold to, say, 3000? Before I rebooted, I noticed that the Nagios alarm message said the thread count was 2074.

-- Pete

On Thu, Dec 7, 2017 at 1:36 PM, Mark Feit <> wrote:

Marian Babik writes:

     …as stopping pscheduler runner leaves quite a number of powstream processes around…

4.0.2 includes some changes to the way the daemons and forked programs are started and stopped that should improve that situation.  Runners started by older versions won’t get swept up during upgrades because those versions don’t include the clean-ups, so you may have to go in and kill off the stragglers manually or just reboot.

If you continue to see that on 4.0.2 systems after a flush-out of the old processes, drop me a line and we can have a look at it.

--Mark