VMware Cloud Community
sblandford
Contributor
Contributor
‎02-19-2009 08:27 AM

Availability on/off in 10 minute cycles at 30% average

I have one server running CentOS 5.x which is reporting intermittent availability. All the other metrics report OK except availability. If I look at the graph it is available 30% of the time in a 10 minute cycle.

There seems to be a log of chatter about re-enabling and disabling services in the agent log file (attached).

I have many other servers connected with no problems. The server in question is not using much swap memory, so I don't think it is memory-squashed. The server also seems to be running fine in every other respect.

I have had this problem with server and agent version 4.0.1 and also after updating them to 4.0.3. I have tried completely removing the agent, deleting the /opt directory and re-installing it. No change in symptom.
client.log
0 Kudos
3 Replies
MarieMcGarry
Enthusiast
Enthusiast
‎02-23-2009 08:26 AM

Hey Simon, Sorry you're seeing a problem. I've asked QA to take a look at this. We will get back to you ASAP.

m
sblandford
Contributor
Contributor
‎02-23-2009 08:32 AM

Thanks. Much appreciated.
0 Kudos
MarieMcGarry
Enthusiast
Enthusiast
‎02-25-2009 09:14 AM

Hey Simon,

I found a support case that has a solution to your problem. The problem was seen in environment where agent host memory for overloaded.

I'm pasting in the support engineer's recommendation here:

In Agent's bin directory execute run the agent script with the "setup" option. This should make the agent repeat the configuration dialog, so you can re-enter agent-server connection settings.

hq-agent.sh setup

If this does not solve the problem, try this:
stop the Agent
delete the data directory
restart the Agent

Simon, will you try this and see if it solves your problem? thanks.

m
0 Kudos