SNMP Changes, time outs and alerting

[Chris Fink]

I changed my snmp community strings over the weekend. Both at the IM7 server and on the managed servers. Not all managed servers took the change and have since gone critical. I expected this. Earlier today I started manually changing the servers to match the new strings, and they started coming back as accessible.

However, there is 1 that is giving me grief. It keeps timing out, I think, and paging out. The server is up, hasn't been rebooted and basically is ok. It's located in a remote site, along with about 25 other servers with the correct strings and they are not timing out. This server goes out from anywhere between 30 minutes to an hour, same with staying accessible. Nothing is consistent.

In order to keep it quite for the night, I have broken the strings again, to keep it in a critical state until I can figure out why it's doing this.

Any ideas would help. It could be a time out issue, but why aren't the other servers effected, and it wasn't doing this before changing the strings.

Thanks,
Chris

[Scott Shaffer]

Chris, what type of link is between the sites - that might help to understand what options you have.

But just basically, you have 2 things you can do. First, you can modify the timeout and rety settings for that system. Select it from within IM7 and then select SNMP Communication Settings. On this page you can modify the timeout and retry values for just this device.

Second, you could create a customer polling interval for just this class of system. You'd have to create a query that defined your remote servers, then reate a Status Polling task that polls those servers on an interval that is more likely to succeed. Given that you only have 1 server in this state, this probably isn't the way to go.