Hi! Here is a daisy of a feature :) .. According to a sys admin, Solaris 10 reports memory used by the filesystem cache (swap space) as kernel memory which, even though it could be deallocated in a heartbeat, means Spectrum polls an OID for memory used and counts both physical and logical RAM. So any system with enough storage accesses over time will eventually show memory full. While this isn't a Spectrum issue, the server admin doesn't have an already-existing SNMP oid to replace the "buggy" oid.
Has anyone else ever run into this issue and how did you fix it? I guess we could cron a script to deallocate, but was looking for something a little more NMS oriented. ________________________________ CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it. --- To unsubscribe from spectrum, send email to [email protected] with the body: unsubscribe spectrum [email protected]
