gpilotd runs away with CPU after completing hotsync

Bug #29043 reported by Rylie Pavlik
12
Affects Status Importance Assigned to Milestone
gnome-pilot (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

I have a Palm Vx connected via a serial dock. I have added the gnome-pilot applet to my panel. After any hotsync (and the gnome-pilot window closes), my system-monitor panel applet (and the Gnome System Monitor confirms) indicates 100% CPU usage. The system monitor reports that gpilotd is using around 80% of cpu power. This is an Athlon 64 3200+ running the AMD64 arch of Ubuntu Breezy.

Right clicking the panel applet for the palm sync and choosing Restart stops the runaway process.

Revision history for this message
Sitsofe Wheeler (sitsofe) wrote :

Ryan:

Does this still happen in a dapper install? If so can you note the version of gnome-pilot using dpkg -l gnome-pilot ?

Changed in gnome-pilot:
status: Unconfirmed → Needs Info
Revision history for this message
Daniel Holbach (dholbach) wrote :

A new version (2.0.14) was uploaded to Edgy. It'd be nice if you had a chance to test this and report back, if the problem still persists.

Changed in gnome-pilot:
assignee: nobody → desktop-bugs
Revision history for this message
rlaska (rlaska) wrote :

I can confirm this bug, and I'm running Edgy with the latest updates.
Output of top:
Tasks: 98 total, 1 running, 97 sleeping, 0 stopped, 0 zombie
Cpu(s): 31.0%us, 69.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 1556036k total, 817400k used, 738636k free, 69340k buffers
Swap: 0k total, 0k used, 0k free, 448256k cached

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 5054 ram 25 0 71544 13m 8556 S 98.2 0.9 17:10.26 gpilotd

----
I'm syncing with a Palm Zire 31. The sync ran fine, but now the CPU is maxed out with gpilotd. Also, the gpilot applet is unresponsive.

Revision history for this message
rlaska (rlaska) wrote :

More info:
It *seems* so far that if i let my Palm sleep after it syncs, then the issue described above happens. But if I turn it off after syncing, it doesn't happen.

That should help narrow it down... something about the Palm sleep isn't shutting the USB connection down cleanly (or *something* like that).

Revision history for this message
rlaska (rlaska) wrote :

Forget my last comment, gpilotd still pegged out the CPU after a few minutes, even when I put my palm to sleep. I'll try putting it to sleep and pulling the USB cable next time.
Curiously, I noticed this in dmesg:
[17187459.704000] visor ttyUSB0: Handspring Visor / Palm OS converter now discon
nected from ttyUSB0
[17187459.712000] visor ttyUSB1: Handspring Visor / Palm OS converter now discon
nected from ttyUSB1

Looks like my Zire 31 is getting two ttyUSB[0-9] ports somehow

Revision history for this message
Lauren Matheson (inan-coldsnap) wrote :

I have had the same problem with gnome-pilot 2.0.14-0ubuntu and my Palm Tungsten|E2

Revision history for this message
Rylie Pavlik (abiryan) wrote :

The two USB ports for the Zire are normal - you want to use the second, I think. I no longer have this problem on Edgy.

Revision history for this message
Rylie Pavlik (abiryan) wrote :

(and I initially reported this and am still using a serial sync dock)

Revision history for this message
Sebastien Bacher (seb128) wrote :

closing the bug then

Changed in gnome-pilot:
status: Needs Info → Fix Released
Revision history for this message
Fedor Isakov (fisakov) wrote :

I think this bug needs to be reviewed. I can confirm the same problem as the original reporter: gpilotd consuming 100% CPU after a successful sync. Actually, there are two outcomes after a sync: either gpilotd starts eating up the CPU, or the "Pilot applet" stops responding to requests (gnome-pilot settings dialog are not displayed). In both situations the solutions is to run 'killall gpilotd'. In the latter case, this follows by several gnome-pilot settings windows being opened.

All this is happening on Feisty with Palm TX connected via USB. I'm using 'usb:' as the palm device, not /dev/pilot.

Changed in gnome-pilot:
status: Fix Released → Confirmed
Revision history for this message
ClaudioRJ (cmirandarj) wrote :

I confirm the same bug using Ubuntu 7.04 (32 bits) in a AMD64, with a Tungsten T5 connected via USB, and using /dev/pilot.

After sync finishes, and as soon as the Palm turns off, the gpilotd app begins to use CPU time at 90%, more or less. So, in System Monitor, I stop thist application, then kill it. So it turns again to 'Executable' status and CPU turns normal again.

Revision history for this message
Fedor Isakov (fisakov) wrote :

Updating on the bug status from fresh Gutsy install.

I noticed the gpilotd doesn't consume 100% CPU after the hotsync finishes, nevertheless one still has to 'killall gpilotd' in order to unlock gpilot-applet, which otherwise stays completely frozen.

Revision history for this message
jnewm (jessenewmark) wrote :

Similar for me on Gutsy. After I run a successful JPilot sync, "gpilotd" starts using massive CPU. I'm also using "usb:" in my JPilot preferences.

Revision history for this message
Rocky Road (m-baert) wrote :

Though I had no problem synchronizing my old palm M500 (on Feisty, then Gutsy), I can't do it with my new TX (the M500 is HS), on Gutsy.
Often, the palm is restarted instead of synchronizing.
gpilotd is eating 95 to 100% CPU.
When I kill gpilotd, many similar applet windows popup (2 to 6).

Revision history for this message
Rocky Road (m-baert) wrote :

errata: with "HS", I meant "out of order"

Revision history for this message
Pedro Villavicencio (pedro) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Thanks in advance.

Changed in gnome-pilot:
status: Confirmed → Incomplete
Revision history for this message
rlaska (rlaska) wrote : Re: [Bug 29043] Re: gpilotd runs away with CPU after completing hotsync

Hi Pedro,

I'm not actively using Ubuntu or a Palm at this time, so I'm afraid I can no
longer comment on it.

Thank you,

Ram Laska

On Thu, Aug 21, 2008 at 2:42 PM, Pedro Villavicencio <email address hidden>wrote:

> Thank you for taking the time to report this bug and helping to make
> Ubuntu better. You reported this bug a while ago and there hasn't been
> any activity in it recently. We were wondering is this still an issue
> for you? Thanks in advance.
>
> ** Changed in: gnome-pilot (Ubuntu)
> Status: Confirmed => Incomplete
>
> --
> gpilotd runs away with CPU after completing hotsync
> https://bugs.launchpad.net/bugs/29043
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Revision history for this message
Rocky Road (m-baert) wrote :

Hi Pedro,

Apologies. I forgot to add a comment here about how I (nearly) solved the problem for me.
I had to play with udev configuration, following these instructions:

  http://www.pilot-link.org/README.usb

In some cases, the gpilotd problem (eating CPU) arises again.
The workaround is to kill the gpilotd process.
gnome-pilot-applet, if running, would popup and suggest to restart gpilotd.
Accept, replug your Palm device, and that's it.

HTH
Thanks to bug maintainers and developers.

RockyRoad

Changed in gnome-pilot:
assignee: desktop-bugs → nobody
Revision history for this message
Gary V Deutschmann Sr (classichauslimited) wrote :

Hi Gang

Same problem here January 30 2009 with Palm Tungsten E2!

System Monitor shows gpilotd SLEEPING but eating 93% of CPU

Htop shows 98% cpu usage /usr/bin gpilotd

I hit this the last time I tried using the Palm on Ubuntu Hardy to update my calendar.
Probably another reason why I still keep a Doze Machine around..........

Over 2 years with no resolution in sight eh!

TTUL
Gary

Revision history for this message
Matt Davey (mcdavey) wrote :

gnome-pilot 2.0.17 has been released (though I doubt there's an ubuntu package yet). It would be interesting to hear if it has any effect on the problem.

This is probably a duplicate of upstream bug 450901 (http://bugzilla.gnome.org/show_bug.cgi?id=450901) and it could probably be fixed by a code change to ignore re-connections immediately after a sync.

Revision history for this message
rusivi2 (rusivi2-deactivatedaccount) wrote :

Thank you for posting this bug.

Does this issue occur in Lucid?

Revision history for this message
Phillip Susi (psusi) wrote :

This package has been removed from Ubuntu. Closing all related bugs.

Changed in gnome-pilot (Ubuntu):
status: Incomplete → Invalid
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.