Your last line of vmstat shows a high number of context switches and interrupts . Most of the time is accounted for in the idle time column.
mpstat output would probably be more useful, perhaps you have lock contention. /sG/ ----- Original Message ----- Hi! I encounter the following problem, which I can reproduce. But I have no clue, why this happens: System: uname -a SunOS frodo 5.11 oi_151a i86pc i386 i86pc Solaris Hardware: Dell PE T710, 8GB RAM When the box is getting low on memory and it starts to page out, it usually locks up without any log entries. It just freezes. Only a hard reboot gets it back running. Even though I have no problems with the mass storage, I read that this kind of problem might occur in conjunction with the controller and driver of the HDDs. The DELL PE T7100 has a SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] and I have the mpt_sas driver installed that seems to ship with OpenIndiana. The swap accessed through this controller. Further more I ran vmstat the other day which produced the following output (I made a screenshot with my camera ;-)) http://www.flickr.com/photos/71479913@N03/6461877513/ The last line shows a huge number of faults but I am not able to interpret that. Does anyone have a hint how to gat rid of this problem? Achim _______________________________________________ OpenIndiana-discuss mailing list [email protected] http://openindiana.org/mailman/listinfo/openindiana-discuss _______________________________________________ OpenIndiana-discuss mailing list [email protected] http://openindiana.org/mailman/listinfo/openindiana-discuss
