Resume fails on kernel 2.6.15-21 and higher

Bug #49276 reported by Matthew Patterson
4
Affects Status Importance Assigned to Milestone
Ubuntu
Fix Released
Undecided
Unassigned

Bug Description

I have a dell 630m running a fully updated dapper install (Jun 10, 2006) with bios A03. When I first installed I ran the 2.6.15-18 and then the 2.6.15-20 kernel. These both provided fully functional sleep and resume (I have not tried hibernate). Starting with kernel 2.6.15-21 and with no change through 2.6.15-23 the resume portion no longer functions. When resuming I receive a blank screen. If I press ctrl-alt-f7 I can bring up a blank screen with a pointer. If I go to a text console I get a corrupted screen. I can adjust the screen brightness using the hotkeys, but nothing else besides console switching and brightness seems to work, and I can also kill X with ctrl-alt-backspace but then nothing else happens. Even pressing the power button to initiate a shutdown has no effect.

Things I have tried with no result:
Removing the sdhci and mmcore modules before sleep
Removing the network (ipw2200) module before sleep
Stopping mysqld before sleep

Here is a normal standby/resume cycle on 2.6.15-20 in the syslog. You can see where the suspend ends and the resume begins because of the 1 minute jump:

Jun 10 07:51:24 localhost gnome-power-manager: Suspending computer because the DBUS method Suspend() was invoked
Jun 10 07:51:24 localhost NetworkManager: <information>^IGoing to sleep.
Jun 10 07:51:24 localhost dhclient: receive_packet failed on eth1: Network is down
Jun 10 07:51:24 localhost NetworkManager: <information>^Imatch
Jun 10 07:51:24 localhost last message repeated 6 times
Jun 10 07:51:24 localhost dhclient: DHCPRELEASE on eth1 to 192.168.2.1 port 67
Jun 10 07:51:24 localhost dhclient: send_packet: Network is unreachable
Jun 10 07:51:24 localhost dhclient: send_packet: please consult README file regarding broadcast address.
Jun 10 07:51:24 localhost postfix/master[5004]: reload configuration /etc/postfix
Jun 10 07:51:25 localhost postfix/master[5004]: reload configuration /etc/postfix
Jun 10 07:51:25 localhost mysqld[4860]: 060610 7:51:25 [Note] /usr/sbin/mysqld: Normal shutdown
Jun 10 07:51:25 localhost mysqld[4860]:
Jun 10 07:51:25 localhost mysqld[4860]: 060610 7:51:25 InnoDB: Starting shutdown...
Jun 10 07:51:27 localhost mysqld[4860]: 060610 7:51:27 InnoDB: Shutdown completed; log sequence number 0 58300
Jun 10 07:51:27 localhost mysqld[4860]: 060610 7:51:27 [Note] /usr/sbin/mysqld: Shutdown complete
Jun 10 07:51:27 localhost mysqld[4860]:
Jun 10 07:51:27 localhost mysqld_safe[5836]: ended
Jun 10 07:51:27 localhost NetworkManager: <debug info>^I[1149951087.723877] nm_hal_device_removed (): Device removed (hal udi is '/org/freedesktop/Hal/devices/net_00_14_22_99_86_34').
Jun 10 07:51:27 localhost kernel: [4295171.931000] ACPI: PCI interrupt for device 0000:02:00.0 disabled
Jun 10 07:51:27 localhost NetworkManager: <debug info>^I[1149951087.753123] nm_hal_device_removed (): Device removed (hal udi is '/org/freedesktop/Hal/devices/net_00_16_6f_2c_20_69').
Jun 10 07:51:27 localhost NetworkManager: <information>^IDeactivating device eth1.
Jun 10 07:51:27 localhost kernel: [4295171.969000] ACPI: PCI interrupt for device 0000:02:03.0 disabled
Jun 10 07:51:30 localhost NetworkManager: <WARNING>^I nm_device_802_11_wireless_get_mode (): error getting card mode on eth1: No such device
Jun 10 07:52:24 localhost NetworkManager: <information>^IWaking up from sleep.
Jun 10 07:52:24 localhost kernel: [4295174.958000] Stopping tasks: ===================================================================================================|
Jun 10 07:52:24 localhost kernel: [4295175.698000] ACPI: PCI interrupt for device 0000:00:1f.2 disabled
Jun 10 07:52:24 localhost kernel: [4295175.709000] ACPI: PCI interrupt for device 0000:00:1d.7 disabled
Jun 10 07:52:24 localhost kernel: [4295175.720000] ACPI: PCI interrupt for device 0000:00:1d.3 disabled
Jun 10 07:52:24 localhost kernel: [4295175.720000] ACPI: PCI interrupt for device 0000:00:1d.2 disabled
Jun 10 07:52:24 localhost kernel: [4295175.720000] ACPI: PCI interrupt for device 0000:00:1d.1 disabled
Jun 10 07:52:24 localhost kernel: [4295175.720000] ACPI: PCI interrupt for device 0000:00:1d.0 disabled

Here is a suspend/resume attempt using 2.6.15-23 (reboot goes to -20):

Jun 10 07:38:41 localhost gnome-power-manager: Suspending computer because the DBUS method Suspend() was invoked
Jun 10 07:38:41 localhost NetworkManager: <information>^IGoing to sleep.
Jun 10 07:38:41 localhost NetworkManager: <information>^Imatch
Jun 10 07:38:41 localhost last message repeated 9 times
Jun 10 07:38:41 localhost dhclient: DHCPRELEASE on eth1 to 192.168.2.1 port 67
Jun 10 07:38:41 localhost dhclient: send_packet: Network is unreachable
Jun 10 07:38:41 localhost dhclient: send_packet: please consult README file regarding broadcast address.
Jun 10 07:38:42 localhost postfix/master[5014]: reload configuration /etc/postfix
Jun 10 07:38:43 localhost postfix/master[5014]: reload configuration /etc/postfix
Jun 10 07:38:43 localhost mysqld[4852]: 060610 7:38:43 [Note] /usr/sbin/mysqld: Normal shutdown
Jun 10 07:38:43 localhost mysqld[4852]:
Jun 10 07:38:43 localhost mysqld[4852]: 060610 7:38:43 InnoDB: Starting shutdown...
Jun 10 07:43:30 localhost syslogd 1.4.1#17ubuntu7: restart.
Jun 10 07:43:30 localhost kernel: Inspecting /boot/System.map-2.6.15-20-386
Jun 10 07:43:30 localhost kernel: Loaded 22974 symbols from /boot/System.map-2.6.15-20-386.
Jun 10 07:43:30 localhost kernel: Symbols match kernel version 2.6.15.
Jun 10 07:43:30 localhost kernel: No module symbols loaded - kernel modules not enabled.
Jun 10 07:43:30 localhost kernel: [4294667.296000] Linux version 2.6.15-20-386 (buildd@vernadsky) (gcc version 4.0.3 (Ubuntu 4.0.3-1ubuntu3)) #1 PREEMPT Tue Apr 4 17:48:51 UTC 2006

It looks rather suspicious that the last entry is the mysqld stopping, that is why I tried killing it before suspending, but that had no effect.

It appears that on the newer kernels the system goes to sleep before it is ready.

Hopefully someone can help, Thanks,
Matt

Revision history for this message
Vassilis Pandis (pandisv) wrote :

Has this been fixed? Resume and suspend were broken in an update but fixed in a subsequent update. The final release kernel should be OK.

Revision history for this message
Matthew Patterson (matt-v8zman) wrote : Re: [Bug 49276] Re: Resume fails on kernel 2.6.15-21 and higher

Yes, I have been using the latest kernels since 24 or 25 without a problem.

Thanks for all the work!

Matt

Vassilis Pandis wrote:
> Has this been fixed? Resume and suspend were broken in an update but
> fixed in a subsequent update. The final release kernel should be OK.
>
>

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.