| Summary: | X uses 90% cpu when rpmdrake is run from drakconf | ||
|---|---|---|---|
| Product: | Mageia | Reporter: | Barry Jackson <zen25000> |
| Component: | RPM Packages | Assignee: | Thierry Vignaud <thierry.vignaud> |
| Status: | RESOLVED DUPLICATE | QA Contact: | |
| Severity: | normal | ||
| Priority: | Normal | CC: | marja11 |
| Version: | Cauldron | ||
| Target Milestone: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Source RPM: | Suspect drakxtools rev. 171084 | CVE: | |
| Status comment: | |||
| Bug Depends on: | |||
| Bug Blocks: | 5000 | ||
| Attachments: |
All updates prior to problem
systemactivity after attempting to use rpmdrake - after 30 mins+ systemactivity after closing rpmdrake |
||
|
Description
Barry Jackson
2011-11-26 21:54:26 CET
what's the link with rev 171084? I think we're just trying to update the progress bar too often. Are you in a VM? What's the output of 'lspcidrake -v |fgrep Card' Keywords:
(none) =>
NEEDINFO (In reply to comment #1) > what's the link with rev 171084? The only link with rev 171084 was the time this started to happen. I had never seen this before that date and it was within a hour or so of an update that included that revision. > I think we're just trying to update the progress bar too often. The progress bar has gone when this starts. The increased CPU usage starts just after the dialog box (with the progress bar) closes. (In reply to comment #2) > Are you in a VM? No - I noticed this on my regular Cauldron installation on real hardware. Tested again in a clean install - again on the same hardware. > What's the output of 'lspcidrake -v |fgrep Card' [baz@jackodesktop ~]$ lspcidrake -v |fgrep Card Card:NVIDIA GeForce FX series: nVidia Corporation|NV31 [GeForce FX 5600 Ultra] [DISPLAY_VGA] (vendor:10de device:0311 subv:1462 subd:9311) (rev: a1) Created attachment 1137 [details]
All updates prior to problem
This is the *full* list of updates just prior to this issue starting.
Created attachment 1175 [details]
systemactivity after attempting to use rpmdrake - after 30 mins+
This attachment shows the system activity which has been the same for over 30 mins from trying to open rpmdrake. The high cpu usage of X starts immediately after the "finding available packages" dialog closes.
Created attachment 1176 [details]
systemactivity after closing rpmdrake
This shows the system activity immediately after the above, but after closing rpmdrake with the window header close button.
Raising this to release blocker as without rpmdrake there is no distribution.
There are no issues using urpmi.
I have tested this in two totally separate mga2 installations (on the same hardware)
The only common feature is that whichever is running has /var/cache/urpmi/rpms mounted to a common partition. This works perfectly with urpmi.
Raising this to release blocker. Priority:
Normal =>
release_blocker Also since last night I have been using nouveau not nvidia as the new X11 is not supported for my card - so it's not the proprietary driver causing this. I must reiterate that it started with the updates listed in the attachment above on Wed Nov 23 - never an issue prior to then. Cannot be release critical since it's not reproductible and only affect you Priority:
release_blocker =>
Normal (In reply to comment #9) > Cannot be release critical since it's not reproductible and only affect you OK. Further testing in a fully updated Cauldron indicates that this also affects system updates when run from mcc but *not* when started from the red systray icon (is it mageiaupdate?). The effect is just the same as in rpmdrake. Removing NEEDINFO, because there is no missing requested information Keywords:
NEEDINFO =>
(none) Should be fixed since rpmdrake-5.28 *** This bug has been marked as a duplicate of bug 2775 *** Status:
NEW =>
RESOLVED \o/ WFM Thanks - back to sanity :)
Marja Van Waes
2012-03-17 22:07:05 CET
Blocks:
(none) =>
5000 |