NPM Dying

Question

Hi All,

I have routinely been having my orion server just stop checking and alerting.  When I attempt to log into the server via rdp, I get an access denied error.  I have to reboot the box to get things back up and running.  I see the following several times in the event log:

Windows cannot determine the user or computer name. (An operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full. ). Group Policy processing aborted.

This one refers to an error with group policy, but there are several programs that cannot function with the same error of "n operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full."

I have never seen this error on a server before.  Is Orion filling up "buffers" on the server?  any ideas?

Mars · Answer

Hi,

Thanks for this tip my syslog service kept on dying a few times a day.
We did have the /3GB switch applied 2 weeks ago, by switching it off again we solved the problem!!

Greetings,

rhltechie · Answer

Hello,

Thanks for your reply.  I am not actually using the /3GB switch.  I contacted supported a day ago and sent them my diags from the box.  They seemed to think it was APM using all of the memory and causing this.  I had not applied the APM SP1..so I did so.  Crossing my fingers that I do not experience this again.

Thanks

aLTeReGo · Answer

This error can occur if OS is running out of non-paged pool. Are you using the /3GB switch on the machine? If so please turn it off and reboot.