Bug 1110835
| Summary: | [scale] balloon stats retrieval should be done outside of the actual API call | |||
|---|---|---|---|---|
| Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Michal Skrivanek <michal.skrivanek> | |
| Component: | vdsm | Assignee: | Francesco Romani <fromani> | |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Yuri Obshansky <yobshans> | |
| Severity: | high | Docs Contact: | ||
| Priority: | unspecified | |||
| Version: | 3.4.0 | CC: | aberezin, bazulay, cpelland, danken, dkuznets, emarcian, fromani, gklein, iheim, lpeer, mavital, michal.skrivanek, oourfali, pkliczew, sherold, ybronhei, yeylon, yobshans | |
| Target Milestone: | --- | Keywords: | ZStream | |
| Target Release: | 3.5.0 | Flags: | mavital:
needinfo+
|
|
| Hardware: | x86_64 | |||
| OS: | Linux | |||
| Whiteboard: | virt | |||
| Fixed In Version: | vdsm-4.13.2-0.18.el6ev | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | ||
| Clone Of: | 1102147 | |||
| : | 1110845 1110848 (view as bug list) | Environment: | ||
| Last Closed: | 2015-01-13 11:42:11 UTC | Type: | Bug | |
| Regression: | --- | Mount Type: | --- | |
| Documentation: | --- | CRM: | ||
| Verified Versions: | Category: | --- | ||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | ||
| Cloudforms Team: | --- | Target Upstream Version: | ||
| Embargoed: | ||||
| Bug Depends On: | 1102147 | |||
| Bug Blocks: | 1110845, 1110848, 1142923, 1156165 | |||
|
Description
Michal Skrivanek
2014-06-18 14:46:54 UTC
*** Bug 1102814 has been marked as a duplicate of this bug. *** Bug verification failed. I ran simple script which perform command getAllVmStats and measure response time See results: getAllVmStats response time w/o patch min 37.319 sec average 46.398 sec max 52.005 sec getAllVmStats response time with patch min 33.058 sec average 48.268 sec max 56.701 sec RHEVM - 3.5.0-0.17.beta.el6ev VDSM - vdsm-4.16.7.1-1.el6ev (In reply to Yuri Obshansky from comment #7) > Bug verification failed. > I ran simple script which perform command getAllVmStats > and measure response time > See results: > getAllVmStats response time w/o patch > min 37.319 sec > average 46.398 sec > max 52.005 sec > > getAllVmStats response time with patch > min 33.058 sec > average 48.268 sec > max 56.701 sec Inital observation suggests that something else is eating the gains obtained with this patch. Difference is likely little enough to get lost in variation. The times are too high anyway. not failed, still in investigation of whether it improved things or not, so moving bug back to appropriate state Need to figure out a better verification test or find the source of high latency in general. May need an artificial test...but I'm not sure it's worth the effort. There are likely other reasons for ~45s average and that's something we should understand... Michal, could you please suggest how we should test it, or maybe point us to other info we can provide to help you figure this out? as I said, perhaps only by artificial test, run the getallvmstats in a loop as fast as you can with and without the patch, you *should* see a real difference; again, not that it's that important to prove it as the benefit is obvious from the flow perspective...it may be just negligent/hidden by other stuff easily Michal, based on comment #7 we already tested your suggested flow, and I can't say we see real difference. How do you suggest we act on this BZ? it was not really representative, the conditions are specific and i suggested artificial test. But IMHO not worth it and you can close the bug Closing based on comment #14 |