549140 - updates search looping

Status: RESOLVED INCOMPLETE

(Toolkit :: Application Update, defect)

Description

[George R. Goffe]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.8) Gecko/20100216 Fedora/3.5.8-1.fc12 Firefox/3.5.8
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.8) Gecko/20100216 Fedora/3.5.8-1.fc12 Firefox/3.5.8

Drag down Help and release on "check for updates" produces a popup that says "checking for updates" that NEVER ENDS.

Reproducible: Sometimes

Steps to Reproduce:
1."click" on Help 
2.release on "check for updates"
3.
Actual Results:  
Drag down Help and release on "check for updates" produces a popup that says "checking for updates" that NEVER ENDS.

Expected Results:  
either find updates or not.

This is the firefox distributed with Fedora Core 12.

Comment 1

[George R. Goffe]

Attachment: strace -xvfp <pid> from just before creating this problem

FF was running when I started this strace -xvfp <pid>. I "clicked" on Help and released on "Check for updates..." and let it run a little so you can see what happens.

Comment 2

[George R. Goffe]

Additional output from firefox:

Registering '@mozilla.org/module-loader/python;1' (libpyloader.so)
Registering '@mozilla.org/network/protocol/about;1?what=python' (pyabout.py)
/usr/lib/xulrunner-1.9.1/python/xpcom/__init__.py:54: DeprecationWarning: BaseException.message has been deprecated as of Python 2.6
  self.message = message
*** AUS:SVC UpdateService:canUpdate - testing /usr/lib/firefox-3.5.8/update.test
*** AUS:SVC UpdateService:canUpdate - testing /usr/lib/firefox-3.5.8/updates/0/update.test
*** AUS:SVC UpdateService:canUpdate - able to update
*** AUS:SVC General:readStringFromFile - file doesn't exist: /usr/lib/firefox-3.5.8/updates/0/update.status
*** AUS:SVC General:readStatusFile - status: null, path: /usr/lib/firefox-3.5.8/updates/0/update.status
*** AUS:SVC UpdateService:_postUpdateProcessing - no status, no update
*** AUS:SVC UpdateManager:_loadXMLFileIntoArray: XML file does not exist
ALSA lib pulse.c:229:(pulse_connect) PulseAudio: Unable to connect: Connection refused

Comment 3

[Robert Strong (they/them - no direct email)]

Could you set the following boolean pref to true in about:config
app.update.log.all

If you are running Firefox 3.6.x now you will need to set app.update.log to true instead.

Select Help -> Check for Updates and open the error console. Copy the entries that start with AUS: into this bug. Thanks

Comment 4

[George R. Goffe]

Robert,

I'm running "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.3a5pre) Gecko/20100421 Minefield/3.7a5pre" now. Is it still ok to set this boolean?

Regards,

George...

Comment 5

[Robert Strong (they/them - no direct email)]

For that version use the same pref name as is used for 3.6.x which is app.update.log

You'll also find it is easier if you first open the error console and click the Clear button and then select Help -> Check for Updates. Copy the entries that start with AUS: into this bug. Cheers

Comment 6

[Robert Strong (they/them - no direct email)]

Forgot to add that you should exist and restart after adding the pref. Are you not getting updates when using this version as well?

Comment 7

[George R. Goffe]

Robert,

Got it.

I have already updated today so I'll have to post tomorrow. Unless you know of
a way to force an update. :-)

I did see these msgs though. I'm running the noscript plugin which might add
clutter but then this problem was happening before I started using it. Do you
want me to turn it off before I try the next update?

aus2.mozilla.org : potentially vulnerable to CVE-2009-3555 
bugzilla.mozilla.org : potentially vulnerable to CVE-2009-3555

George...

Comment 8

[Robert Strong (they/them - no direct email)]

If you are able to update then are you still seeing this bug?

Comment 9

[Robert Strong (they/them - no direct email)]

For example, according to comment #0 it is not telling you whether there is or isn't an update. If that is what was happening and you are still experiencing this bug then you should be able to check for updates right now and it won't tell you that there isn't an update whether you have updated today or not.

Comment 10

[George R. Goffe]

Robert,

The failure has gone random now.

Comment #0 gets a popup that says "searching for updates". It sometimes NEVER ends.

The next time it goes into "forever" mode I'll have a trap and will post my results.

George...

Comment 11

[Wayne Mery (:wsmwk)]

(In reply to George R. Goffe from comment #10)
> Robert,
> 
> The failure has gone random now.

George, do you still see this problem?
If not, can you close the bug please as WORKSFORME.

Comment 12

[George R. Goffe]

Wayne/Robert,

I don't recall this situation happening recently. I did have a few days where starting FF Nightly after an update failed to start/crashed. I suspect that this was due to an update that failed with an out of space condition. I'll try running with app.update.log.all set to the opposite of the default and watch it.

Of course if  you want to close this ticket, that would be fine. I can always open a new one when/if the problem re-appears.

Thanks!

George...

Comment 13

[George R. Goffe]

Oops, app.update.log.all does not exist any more, app.update.log does though. I have set this to true now.

Comment 14

[Phoenix]

Resolved per whiteboard

Comment 15

[George R. Goffe]

Phoenix,

Resolved by whiteboard? What is that? I was not aware of this whiteboard. I would have thought that since I opened this ticket, that I would have been made aware of what was going on.

What does incomplete mean? Did you ask me for something that I didn't provide?

I have not experienced this problem for a couple of weeks now.

George...

Comment 16

[Phoenix]

(In reply to George R. Goffe from comment #15)
> Resolved by whiteboard? What is that?
That's mean, that for this bug was set close date in case if reporter's problem gone, but he doesn't close the bug. Incomplete is default resolution for whiteboard-closed bug, where both not reason of appearing the problem, nor reason of "fix" is known.
In case if problem appear again, you can either reopen this bug or fill new