Bug 470752
| Summary: | ACPID is claiming power button even when gnome-power-manager is running | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Martin Ebourne <fedora> |
| Component: | acpid | Assignee: | Zdenek Prikryl <zprikryl> |
| Status: | CLOSED RAWHIDE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | rawhide | CC: | martin, zprikryl |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2008-11-12 06:33:44 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
I can confirm that /etc/acpi/actions/power.sh does not correctly detect a running kpowersave (the g-p-m for KDE4). My power.sh script does detect the correct uid, but it fails to match the output of ps, as "ps axo uid,cmd" outputs "/usr/bin/kpowersave" and not "kpowersave". Maybe this also goes for gnome-power-manager? On this box I'm running Fedora 9, 2.6.26.6-79.fc9.x86_64., pm-utils-1.1.0-7.fc9.x86_64. In /etc/acpi/actions/power.sh, if I change "kpowersave" to "/usr/bin/kpowersave", I get the wanted behaviour. The man page of "ps" states that "cmd" means the "simple name of executable", which I guess means the full path including the name, but not the option list. MartinG, that is clearly a different but very much related bug. My issue is due to a change in Fedora 10 that breaks the UID check. It was working on F9 and it works on F10 once I fix the check. I am using gnome though. Looks like in the KDE case it's already broken on F9. Not currently at my machine so can't check to see if there's any change in that for F10. Is anyone looking at this bug? Can we get this fixed before F10 final? From my experience having the power button shut the machine down when the user only expected it to suspend to RAM is very unpleasant. (In reply to comment #0) > fixes the problem. Could be more fancy and check for either I guess. I think check for "unix-user" is sufficient. (In reply to comment #1) > In /etc/acpi/actions/power.sh, if I change "kpowersave" to > "/usr/bin/kpowersave", I get the wanted behaviour. I'll fix the script in other way, instead "==" I'll use "~". This solution catches either kpowersave or /usr/bin/kpowersave. Also I'll try to get this update to the F-10 final. Great - thanks :) |
Description of problem: If I press the laptop power button to suspend to RAM (as configured in g-p-m and as was working for F9) the machine shuts down. Version-Release number of selected component (if applicable): acpid-1.0.6-8.fc10.x86_64 ConsoleKit-0.3.0-2.fc10.x86_64 How reproducible: Every time Steps to Reproduce: 1. Press power button Actual results: Machine shuts down Expected results: Action configured as per g-p-m should occur. Additional info: Looks like ConsoleKit has changed the output of ck-list-sessions: # ck-list-sessions Session2: unix-user = '500' realname = 'Martin Ebourne' seat = 'Seat1' .... UID is 'unix-user' whereas /etc/acpi/actions/power.sh is looking for 'uid': ck-list-sessions | \ awk ' /^Session[0-9]+:$/ { uid = active = x11 = "" ; next } { gsub(/'\''/, "", $3) } $1 == "uid" { uid = $3 } Changing this to: $1 == "unix-user" { uid = $3 } fixes the problem. Could be more fancy and check for either I guess.