HI zenoss-users, I have been trying to figure this out for quite some time - we have a bunch of servers running Solaris 10/SPARC and the ucd-based SNMP agent with several large (>1.1TB) filesystems on each for which ZenOSS keeps generating "weird" alerts as a result of apparent miscalculation of the filesystem's size. A couple of examples:
Filesystem AA on server A has a size of 2.0TB of which 1.3TB is in use, for a utilisation of 67% (df -h output) ZenOSS reports the same to be a 10.6GB filesystem with 25.6GB in use, for a utilisation of 242% Filesystem BB on server B has a size of 1.5TB of which 29G is in use, for a utilisation of 2%. ZenOSS reports the same to be a 68.5KB filesystem with 29.3GB, for a utilisation of 44782503%...! After a few tests with CLI SNMP commands I figured out that these sizes are being misreported by the SNMP agent. Googling around I read some posts (e.g. http://www.mail-archive.com/[email protected]/msg10881.html ) claiming that this is a bug in the MIB although there may be a few (inelegant) workarounds. Reading the posts here and in the -dev list/forum I have not managed to figure out what the state of play is for ZenOSS regarding monitoring of large filesystems - does anyone here have an idea on how this can be done? Grateful for any responses, of course. Thanks, Apollon -------------------- m2f -------------------- Read this topic online here: http://forums.zenoss.com/viewtopic.php?p=37836#37836 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
