From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; nb-NO; rv:1.7.10) Gecko/20050720 Fedora/1.0.6-1.1.fc4 Firefox/1.0.6 Description of problem: The ACPI battery status suddenly stops reporting data. [oyvind@arwen ~]$ cat /proc/acpi/battery/BAT0/{state,info} present: yes ERROR: Unable to read battery status present: yes ERROR: Unable to read battery information I became aware of this because the battery status applet suddenly reported "No battery present", for no apparent reason. The battery in my laptop works fine, and ACPI battery monitoring has always worked on previous kernels. Version-Release number of selected component (if applicable): kernel-2.6.12-1.1435_FC4 How reproducible: Didn't try Additional info: /var/log/messages contains a whole bunch of lines like these: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BIF] (Node dee678c0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.AC__._PSR] (Node dee67aa0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BST] (Node dee678a0), AE_OWNE R_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0._BIF] (Node dee678c0), AE_OWNE Will attach 'lspci -vv' and dmesg output at boot time.
Created attachment 118134 [details] dmesg at boot time
Created attachment 118135 [details] lspci -vv
Related to this update, perhaps (?): http://ftp.kernel.org/pub/linux/kernel/people/lenb/acpi/patches/release/2.6.12/broken-out/acpica-unix-20050729.patch
Can you try the latest errata kernel from updates-testing please ? That has an ACPI update which could fix this.
kernel-2.6.12-1.1435_FC4 seems to be the latest in updates-testing, at the moment, which is the one showing the problem. Did you mean rawhide/development (which currently seems to be kernel-2.6.12-1.1517_FC5) ? Perhaps interesting to note are these messages from kernel log, which I just got: ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 Battery status monitoring still works, though. But I don't know for how long. There is nothing suspicious in /var/log/acpid. The problem is not easy to reproduce, and I don't know what causes it, but I'll report back on this bug with newer versions from updates-testing, when they're released.
Hmm, can't remember changing the bug status (?), anyway, I set it back to NEEDINFO.
Getting more of these, but batt-stat is still working: ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0. agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode agpgart: Putting AGP V2 device at 0000:01:05.0 into 4x mode ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.BAT0.Z001] (Node dee66840), AE_OWNER_ID_LIMIT ACPI-0509: *** Error: Method execution failed [\_SB_.PCI0.ISA_.EC__._Q08] (Node dee68c40), AE_OWNER_ID_LIMIT
If I could, I'd change the bug title to "ACPI battery status stops working after a while". It seems that if I just leave my laptop on long enough, it will eventually break. This has happened again, after lots of the above ACPI error messages in the log, including those initially reported, now last in the kernel log. Once I start getting "ACPI-0145: *** Error: Invalid owner_id: 00", I can be pretty sure it will stop working, in a few hours. The good news is I'll probably be able to tell, with a fair amount of certainty, if the bug has gone away in an updated kernel release.
My IBM Thinkpad T41 is now affected. My previous kernel was kernel-2.6.12-1.1398_FC4 which was working fine, but kernel-2.6.12-1.1447_FC4 is now broken in this way.
Hmm something related to this is further hosed. Maybe 6 hours after the battery indicator stopped working, the laptop deadlocked. kernel-2.6.12-1.1398_FC4 had previously been fine for weeks of uptime.
I'm still getting: ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 with kernel-2.6.12-1.1447_FC4. I bet the battery status will kill itself in a matter of hours (though how many, I surely cannot say). (In reply to comment #10) > Hmm something related to this is further hosed. Maybe 6 hours after the battery > indicator stopped working, the laptop deadlocked. kernel-2.6.12-1.1398_FC4 had > previously been fine for weeks of uptime. I haven't experienced lockups after batt-stat is dead. But it won't surprise me if it happens, because by that stage my syslog is littered with ACPI ERROR messages.
Created attachment 118249 [details] Cleanup for acquire/release of owner_id
I've been running a custom built 2.6.13 vanilla for the past day and it has been perfect here. I hope we have a kernel update again sometime soon...
Same problem here with kernel-2.6.12-1.1447_FC4 : Aug 31 08:37:24 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:38:36 portable last message repeated 3 times Aug 31 08:40:24 portable last message repeated 4 times Aug 31 08:40:24 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:41:35 portable su(pam_unix)[4001]: session opened for user root by (uid=500) Aug 31 08:41:37 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:41:37 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:42:17 portable su(pam_unix)[4001]: session closed for user root Aug 31 08:42:29 portable su(pam_unix)[4054]: session opened for user root by (uid=500) Aug 31 08:42:31 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:42:31 portable kernel: ACPI-0145: *** Error: Invalid owner_id: 00 Aug 31 08:42:35 portable su(pam_unix)[4054]: session closed for user root Aug 31 08:43:25 portable kernel: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT Aug 31 08:43:25 portable kernel: ACPI-0509: *** Error: Method execution failed [\_SB_.C044.C057.C0E7.C12B] (Node c14df6e0), AE_OWNER_ID_LIMIT Aug 31 08:43:25 portable kernel: ACPI-0509: *** Error: Method execution failed [\_SB_.C12B] (Node c14dc580), AE_OWNER_ID_LIMIT Aug 31 08:43:25 portable kernel: ACPI-0509: *** Error: Method execution failed [\_SB_.C137._BIF] (Node c14dc4a0), AE_OWNER_ID_LIMIT
Just a "me too". See the following on a new ThinkPad X32: Aug 30 19:54:36 makani kernel: ACPI-0145: *** Error: Invalid owner_id: 00 with 2.6.12-1.1447_FC4
I'm gonna try Alexey Starikovskiy's cleanup patch on 2.6.12-1.1447_FC4 (I use a custom build of this kernel, anyway..). Will report back.
(In reply to comment #16) > I'm gonna try Alexey Starikovskiy's cleanup patch on 2.6.12-1.1447_FC4 (I use a > custom build of this kernel, anyway..). Will report back. 2.6.12-1.1447_FC4 + owner_id-3.patch from Alexey Starikovskiy is so far looking good. No ACPI error messages in syslog. But I might be drawing this conclusion to soon, since I haven't had enough uptime to be sure, yet. Will re-confirm, once it's been running a little longer.
Me too. Thinkpad T43. Started after kernel update to: 2.6.12-1.1447_FC4 #1 Fri Aug 26 20:29:51 EDT 2005 i686 i686 i386 GNU/Linux Previous kernel did not have this problem. KLaptop shows "Plugged in - no battery" [root@lt1 ~]# dmesg Non-volatile memory driver v1.2 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0145: *** Error: Invalid owner_id: 00 ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT
(In reply to comment #17) > 2.6.12-1.1447_FC4 + owner_id-3.patch from Alexey Starikovskiy is so far looking > good. No ACPI error messages in syslog. But I might be drawing this conclusion > to soon, since I haven't had enough uptime to be sure, yet. Will re-confirm, > once it's been running a little longer. Still good, the patch fixes the problem. I'm guessing it (or something similar to it) will be included in the next Fedora kernel release. 2.6.12-1.1447_FC4 + owner_id-3.patch is confirmed working.
I'm also seeing this on my JVC minnote laptop 7230. But additionally the ACPI hotkey HOTK events are not been seen by acpid. The hotkeys worked in 1398. Do you think the owner_id-3.patch will fix this too?
I cannot say for sure, but at least it gets rid of all the error messages. Battery status has worked fine ever since I applied it. This bodes well for ACPI HOTK too, I would think (though I don't use it, myself).
After upgrading to the kernel 2.6.12-1.1398_FC4 i have the same messages. After some time the systems hangs completly!!! I have an ASUS L3800c
I have a similar problem since upgrading to FC4 and 1447. It's an older Thinkpad T30, and believe it or not it's triggered when I push the Light LED button, and shortly after that the laptop locks up, but I can keep the laptop moving slowly by hitting the LED button on/off :) Here's some of the output: Sep 3 10:27:17 maat kernel: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT Sep 3 10:27:17 maat kernel: ACPI-0563: *** Error: AE_OWNER_ID_LIMIT while evaluating method [_L18] for GPE[ 0]
I tried owner_id-3.patch. It looks good, no acpi error messages in /var/log/messages. But I found another problem below. Only me? (1) i unplug my ac_adapter. --> acpid receives a event "battery BATA 00000080 00000001" only. No ac_adapter event. (2) i plug my ac_adapter. --> acpid receives some events including an ac_adapter event. (case1: 2.6.12-1.1447_FC4+owner_id-3.patch) $ sudo /usr/sbin/acpid -d [Sun Sep 4 10:17:01 2005] starting up [Sun Sep 4 10:17:01 2005] DBG: parsing conf file /etc/acpi/events/power [Sun Sep 4 10:17:01 2005] 1 rule loaded [Sun Sep 4 10:17:02 2005] client connected from 3022[500:500] [Sun Sep 4 10:17:02 2005] 1 client rule loaded # unplugging my ac_adapter [Sun Sep 4 10:17:10 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:17:10 2005] DBG: rule from 3022[500:500] matched [Sun Sep 4 10:17:10 2005] notifying client 3022[500:500] [Sun Sep 4 10:17:10 2005] DBG: 1 total rule matched [Sun Sep 4 10:17:10 2005] completed event "battery BATA 00000080 00000001" # plugging my ac_adapter [Sun Sep 4 10:17:41 2005] received event "processor CPU0 00000080 00000000" [Sun Sep 4 10:17:41 2005] DBG: rule from 3022[500:500] matched [Sun Sep 4 10:17:41 2005] notifying client 3022[500:500] [Sun Sep 4 10:17:41 2005] DBG: 1 total rule matched [Sun Sep 4 10:17:41 2005] completed event "processor CPU0 00000080 00000000" [Sun Sep 4 10:17:41 2005] received event "processor CPU0 00000081 00000000" [Sun Sep 4 10:17:41 2005] DBG: rule from 3022[500:500] matched [Sun Sep 4 10:17:41 2005] notifying client 3022[500:500] [Sun Sep 4 10:17:41 2005] DBG: 1 total rule matched [Sun Sep 4 10:17:41 2005] completed event "processor CPU0 00000081 00000000" [Sun Sep 4 10:17:41 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:17:41 2005] DBG: rule from 3022[500:500] matched [Sun Sep 4 10:17:41 2005] notifying client 3022[500:500] [Sun Sep 4 10:17:41 2005] DBG: 1 total rule matched [Sun Sep 4 10:17:41 2005] completed event "battery BATA 00000080 00000001" [Sun Sep 4 10:17:41 2005] received event "ac_adapter AC 00000080 00000001" [Sun Sep 4 10:17:41 2005] DBG: rule from 3022[500:500] matched [Sun Sep 4 10:17:41 2005] notifying client 3022[500:500] [Sun Sep 4 10:17:41 2005] DBG: 1 total rule matched [Sun Sep 4 10:17:41 2005] completed event "ac_adapter AC 00000080 00000001" (case2: 2.6.12-1.1398_FC4) $ sudo /usr/sbin/acpid -d [Sun Sep 4 10:28:08 2005] starting up [Sun Sep 4 10:28:08 2005] DBG: parsing conf file /etc/acpi/events/power [Sun Sep 4 10:28:08 2005] 1 rule loaded [Sun Sep 4 10:28:09 2005] client connected from 3027[500:500] [Sun Sep 4 10:28:09 2005] 1 client rule loaded # unplugging my ac_adapter [Sun Sep 4 10:28:26 2005] received event "processor CPU0 00000080 00000000" [Sun Sep 4 10:28:26 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:26 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:26 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:26 2005] completed event "processor CPU0 00000080 00000000" [Sun Sep 4 10:28:26 2005] received event "processor CPU0 00000081 00000000" [Sun Sep 4 10:28:26 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:26 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:26 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:26 2005] completed event "processor CPU0 00000081 00000000" [Sun Sep 4 10:28:26 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:26 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:26 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:26 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:26 2005] completed event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:26 2005] received event "ac_adapter AC 00000080 00000000" [Sun Sep 4 10:28:26 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:26 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:26 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:26 2005] completed event "ac_adapter AC 00000080 00000000" [Sun Sep 4 10:28:27 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:27 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:27 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:27 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:27 2005] completed event "battery BATA 00000080 00000001" # plugging my ac_adapter [Sun Sep 4 10:28:40 2005] received event "processor CPU0 00000080 00000000" [Sun Sep 4 10:28:40 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:40 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:40 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:40 2005] completed event "processor CPU0 00000080 00000000" [Sun Sep 4 10:28:40 2005] received event "processor CPU0 00000081 00000000" [Sun Sep 4 10:28:40 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:40 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:40 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:40 2005] completed event "processor CPU0 00000081 00000000" [Sun Sep 4 10:28:40 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:40 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:40 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:40 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:40 2005] completed event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:40 2005] received event "ac_adapter AC 00000080 00000001" [Sun Sep 4 10:28:40 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:40 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:40 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:40 2005] completed event "ac_adapter AC 00000080 00000001" [Sun Sep 4 10:28:42 2005] received event "battery BATA 00000080 00000001" [Sun Sep 4 10:28:42 2005] DBG: rule from 3027[500:500] matched [Sun Sep 4 10:28:42 2005] notifying client 3027[500:500] [Sun Sep 4 10:28:42 2005] DBG: 1 total rule matched [Sun Sep 4 10:28:42 2005] completed event "battery BATA 00000080 00000001"
(In reply to comment #24) > I tried owner_id-3.patch. It looks good, no acpi error messages in > /var/log/messages. But I found another problem below. Only me? > > (1) i unplug my ac_adapter. > --> acpid receives a event "battery BATA 00000080 00000001" only. No ac_adapter > event. > (2) i plug my ac_adapter. > --> acpid receives some events including an ac_adapter event. I've had problems with AC status change on my laptop since the first time I installed Linux on it. acpid clients (like the Gnome batt-stat applet) don't detect state changes because no event is received on AC disconnect. It's a Compaq, and after some investigation into the problem, I found out it is most likely the ACPI hardware (quirky ACPI implementation). However, it did work in pre-installed Windows XP, but that was because of special Compaq driver _crap_, I guess. I've given up on the problem (like I've given up getting ACPI suspend to work =).
Same thing here on a Compaq Evo N800v laptop. ACPI works fine under kernel-2.6.12-1.1398_FC4, the problems started appearing with kernel-2.6.12-1.1447_FC4.
Me too (as comment #26) running on a Compaq Evo N800w.
Patch from owner_id patch from Alexey @ Intel seems to have fixed the basic problem on my T30. Though there is a problem with the default kernel install path it generates with modutils because it uses an @ symbol and requires me to softlink lib/modules otherwise it complains while booting.
Owner_id patch from Alexey @ Intel seems to have fixed the basic problem on my T30. Though there is a problem with the default kernel install path it generates with modutils because it uses an @ symbol and requires me to softlink lib/modules otherwise it complains while booting.
I'm also getting this problem, Thinkpad T42 with an FC3 kernel: kernel-2.6.12-1.1376_FC3 Similar set of symptoms: * After a while of uptime, the battery applet stops responding (in Gkrellm) * /var/log/messages starts to have acpi errors (see below) * Twice so far the system has deadlocked; I hadn't connected the two events before now Here is the current batch of error output, every five seconds: Sep 6 01:25:40 erato kernel: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT Sep 6 01:25:40 erato kernel: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT Sep 6 01:25:40 erato kernel: ACPI-0509: *** Error: Method execution failed [\_SB_.PCI0.LPC_.EC__.BAT0._BIF] (Node dfe4c920), AE_OWNER_ID_LIMIT Sep 6 01:25:40 erato kernel: ACPI-0105: *** Error: Could not allocate new owner_id (32 max), AE_OWNER_ID_LIMIT Sep 6 01:25:40 erato kernel: ACPI-0509: *** Error: Method execution failed [\_SB_.PCI0.LPC_.EC__.BAT0._BST] (Node dfe4c900), AE_OWNER_ID_LIMIT
http://people.redhat.com/davej/kernels/Fedora/ Everyone please test the newer FC4 kernels from here. I have been running 2.6.12-1.1451_FC4 for the past day and it has been good for me.
A new kernel 2.6.12-1.1450_FC4 (perhaps very similar to the one Warren suggested) is also available in updates-testing repository: http://download.fedora.redhat.com/pub/fedora/linux/core/updates/testing/4/ and reading the changelog it looks promising: * Sat Sep 03 2005 Dave Jones <davej> - Various post 2.6.13 ACPI updates. (20050902) I'll test it asap.
I've tried davej/...1451_FC4.i686 on my JVC mininote I'm still not seeing any acpi events, there's nothing appearing in the acpid log after startup. Hotkeys don't work, as well as missing ac power events etc But the invalid owner_id messages have gone away.
2.6.12-1.1450_FC4 seems OK wrt. ACPI error messages and battery status.
Just another incident reporting... this began happening on my Compaq nc6000 after I upgraded to 2.6.12-1.1447 and dbus-0.33-3 via yum. After reading this bugzilla, I doubt dbus has anything to do with the issue. However, unlike some of these other incident reports, I see the problem very quickly (less than an hour, closer to 20 minutes).
Began using 2.6.12-1.1450_FC4 yesterday. No reoccurrence of the problem as yet.
*** Bug 167013 has been marked as a duplicate of this bug. ***
I tried davej/...1451_FC4.i686 on my panasonic cf-w4. The invalid owner_id messages had gone away, but my another problem #24 was NOT. So, I tried davej/....1454_FC4.i686 with "CONFIG_ACPI_DEBUG=y"(rebuilded). The error messages below was logged in /var/log/messages, when I had disconnected(connected) my ac_adapter. osl-0913 [42] os_wait_semaphore : Failed to acquire semaphore[c1476d80|1|0], AE_TIME osl-0913 [52] os_wait_semaphore : Failed to acquire semaphore[c1472280|1|0], AE_TIME osl-0913 [65] os_wait_semaphore : Failed to acquire semaphore[c1476d80|1|0], AE_TIME osl-0913 [40] os_wait_semaphore : Failed to acquire semaphore[c1476d80|1|0], AE_TIME osl-0913 [40] os_wait_semaphore : Failed to acquire semaphore[c1476d80|1|0], AE_TIME ...(same messages was logged many times.)
Same symptoms on hp nx7010 running 2.6.12-1.1447_FC4. Will try 1454 now.
Same problem on an OEM laptop. 2.6.12-1.1454_FC4 seems to fix it, but it does not yet fix the missing /proc/acpi/button
With davej's kernel-2.6.12-1.1454_FC4 on JVC mininote - the batstat app is unable to access the stats consistently, and after about 15 minutes gets stuck reading 0%. - "cat /proc/acpi/batery/BAT0/state" is inconsistent, sometimes reading 0, and very slow to complete, 1+seconds. - acpid still not getting any events -- lid, hotkey etc
I installed the kernel-2.6.12-1.1376_FC3 package from davej's people.r.c page. So far, so good. 28 hours of uptime, no complaints from ACPI to syslog, battery status applet is working fine (in gkrellm), and no deadlock.