Collector Service is using more than 4 GB of RAM (5.7 GB)

I have collected diagnostic logs and create an Orion Insight Report. while reviewing the report, I got one error that is related to Collector Service which is utilizing high memory usage. I tried to restart the SolarWinds service. it got fixed but again it is utilizing more RAM.

Thanks in advanced.

Find more posts tagged with

Accepted answers

All comments

vinay.by

You can always repair the same, but I would suggest you to raise a vendor support case before you proceed with this.

Repair a failing Job Engine and Collector Components

gaba8696

I opened a case with SolarWinds vendor, and they suggested to repair the SolarWinds installer on the Primary Server. I tried to failover the primary server to secondary due to some reason unable to failover it. then I just restarted the Services on the Primary server, and it got fixed. gradually it starts increasing again.

vinay.by

did they suggest repair as mentioned in the link provided by me ? Restart might have fixed it temporarily - the details mentioned on the link will not break anything as i have used it in the past, but its good to send SolarWinds support this link and ask them if this can be done.

vinay.by

And with respect to failover , stop HA on web gui, reboot your secondary server, switch on HA on web gui again and then try failing over, if everything is intact it should failover this time

gaba8696

No, they have mention different steps to repair complete SolarWinds installer, below are the steps provided by vendor.

Resolution:

1. RDP into Server hosting Main poller.
2. Open RUN and type "Appwiz.cpl".
3. Search SolarWinds Platform and right click to repair.
4. Run the repair and complete the configuration wizard.

Note: Configuration wizard will prompt automatically in the middle of repairing.

gaba8696

I followed the same procedure. however, I did not perform a reboot of the secondary server. I disabled HA from the console and stopped all services on the primary server. After waiting for approx two minutes to allow failover to the secondary server, I manually started all services on the secondary server but was unsuccessful.

The following day, I stopped the Cortex services on the primary server, which immediately triggered a failover to the secondary server. Based on this observation, I suspect that HA was unable to initiate failover manually to the secondary node during the initial attempt.

gaba8696

I have reviewed the link, you have provided. based on the steps, I can confirm that no downtime would be needed to follow these steps.