We're noticing that the SolarWinds agent is consuming a significant amount of RAM on our servers.
In this instance, it's using over 3GB.
Restarting the agent does temporarily alleviate the issue. I'm thinking of running a scheduled task on all my servers to restart the agent daily, but that seems to be masking the problem.
Does anyone have any ideas on how to resolve this?
We have published a KB article with a temporary workaround to this issue while we continue to develop a more permanent fix. It is available at the link below.
aLTeReGo When I go to this link https://support.solarwinds.com/Success_Center/Orion_Platform/Knowledgebase_Articles/Orion_agent_memo... it asks me to login and it is not accepting my credentials. Do you have another link to that page? I was unable to find it when I search in the Success Center.
So we fix one issue, which exposes another issue...
Please help me understand this;
Seriously, am I missing something here?
Yesterday, a server admin was complaining about the Solarwinds Agent using so much memory. I uninstalled the agent with the buddy fix and I watched the used memory drop 9 Gigs. I then installed the agent with out the Cortex option and the used memory barely increased. I checked it again today and the used memory was nothing like it was yesterday. The server admin claimed he saw an immediate increase in server response after I uninstalled the agent software with the ServiceHost_220.127.116.116.apkg plugin, and he's still happy.
Can someone please provide an update on when we can expect an actual hotfix to address this issue?
You are correct that the KB as published has not worked for all customers so we have made a buddy drop available for those customers affected with open support cases. This buddy drop will effectively replace/overwrite the offending plugin with a dormant stub until a proper fix that addresses the leak is available.
The article says "Note this issue means a loss of functionality to perform Cortex Real-time Polling against those agents".
What functionality would be lost? I'm trying to figure out if it's better to disable this plugin or to continue to mask the issue by restarting the agent service.
My support case for this issue was just closed with the following:
We are writing to inform you that we are closing this case. Our Engineering team are currently unable to implement a fix for your issue at this time.
A fix may be implemented in the future.
That's completely unacceptable. My case is still open (case 00203794). Try responding back and referencing the case numbers in this thread.
For the life of me I can't understand why they would do that.
At this point i can definitively say this buddy drop did not resolve our with the Solarwinds.Servicehost.Process memory leak. We're updating out ticket now.
See screenshots of 2 different systems. You can clearly see when the patch was install point, the memory usage dropping and then climbing back up. By the end of the week, I'm forecasting that the Solarwinds.ServiceHost.Process memory usage will be the same if not greater than when we installed the buddy drop.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process. Learn more today by joining now.