Menu
▾
▴
[Sblim-issues] [ sblim-Bugs-1585628 ] gather: Invalid value of CEC CPU usage
|
From: SourceForge.net <no...@so...> - 2006-10-27 08:13:38
|
Bugs item #1585628, was opened at 2006-10-27 09:57 Message generated for change (Comment added) made by mihajlov You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=712784&aid=1585628&group_id=128809 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Provider Group: Function Status: Open Resolution: None Priority: 5 Private: No Submitted By: Viktor Mihajlovski (mihajlov) Assigned to: Viktor Mihajlovski (mihajlov) Summary: gather: Invalid value of CEC CPU usage Initial Comment: When we gathered CEC CPU usage by using SBLIM , we encountered invalid data of "TotalCPUTime" "ExternalViewTotalCPUPercentage" "UserModeTime" and "ExternalViewUserModePercentage" in zCECMetricValue. It seems to happen when shutting down and rebooting many linux. ---------------------------------------------------------------------- >Comment By: Viktor Mihajlovski (mihajlov) Date: 2006-10-27 10:13 Message: Logged In: YES user_id=1198711 The attached patch fixes the miscalculation of CPU utilization after the deactivation of one or more LPARs: - gatherer plugin is providing a CPU time adjustment for all detected inactive LPARs - repository plugin is applying time adjustment if necessary Related problems fixed with this patch: - take into account the per LPAR overhead for the LPAR kernel mode time - make sure that single LPAR data is consistent after a deactivate/re-activate ---------------------------------------------------------------------- Comment By: Viktor Mihajlovski (mihajlov) Date: 2006-10-27 10:01 Message: Logged In: YES user_id=1198711 The problem is that the CPU counters for the LPARs that have been deactivated will be reset. The outcome can be that the cumulated CPU times of the interval end are lower than at the interval start. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=712784&aid=1585628&group_id=128809 |