We are having hgh cpu utilization and I could see tat are two gwpoa running with deifferent PID´s
Cybersecurity
DevOps Cloud
IT Operations Cloud
We are having hgh cpu utilization and I could see tat are two gwpoa running with deifferent PID´s
One post office has two processes , one master and second is spawned automatically for soap communication.
David
since today gw24.2 is released today! (only for information)
why do you run 2 gwpoa and only 1 post office?
do you have a secondary domain on your server? this is necessary if your first domain crashed!
i am on oes24.1.1 with my groupwise server and there is no trouble/problem with the release build of gw24.2
this only for information, but why 2 gwpoa for 1 po? Thats my question!
as per David, two instances is normal.
as for high CPU, there are a number of things that it could be that I've seen
- Tight on memory, so it is busy swapping out all the time. GroupWise really benefits by having sufficient memory to cache the active database files. I watch the swap levels, and if swap is being fully consumed, it likely benefits from adding memory to this system. If it takes a few weeks to use it up without the high cpu, then we can play with swappiness some. Details assuming on SLES/OES based on the PID reference, but version would still be a good reference here.
- Drive access issues. I've seen where an over busy SAN (DR replication issue) can show as high CPU. Make sure your storage system is healthy, and that you generally have over 10% free on all mount points.
- GWChecks (schedule or not GW maintenance) can really push a system. You can see if any are running on the POA http status page, and I have my recommended GW practices page you can check out as well. After the above-mentioned SAN issues, the bigger GWChecks (Full Contents Analyze/Fix) would run way longer than their normal weekend runs.
________________________
Andy of KonecnyConsulting.ca in Toronto
Please use the "Like" and/or "Verified Answers" as appropriate as that helps us all.
Tks, so one down, very good information
Greiner,
As davidkrotil said into a reply of this post:
So that is normal
Shame on me, very important things to check.
I will check those things out and reply here
Thank you so much
to see if the memory is enough activate nagios service on your sles15 or oes-server and install scatools, then you become a html as output and you see any important informations of your server!