539717 - update is paused, can't unpause

Status: RESOLVED FIXED

(Toolkit :: Application Update, defect)

Description

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

I'm still running

Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2b6pre) Gecko/20091228 Namoroka/3.6b6pre

likely because when I go to Check For Updates, it says

( ) Paused downloading Namoroka 3.6b6pre

Clicking the |> button does nothing (it seems disabled), and restarting the browser doesn't help.

What else can I inspect to figure out why update is stuck for me?  Will avoid restarting the browser in case it jars something, though I've rebooted this machine many a time since Dec 28th.

Comment 1

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

Try adding a boolean pref with the name app.update.log and a value of true. The error console might provide some insight

Comment 2

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

AUS:UI gUpdates:onLoad - setting current page to startpage downloading
AUS:SVC readStatusFile - status: null, path: C:\Users\shaver\AppData\Local\Mozilla\Firefox\Minefield\updates\0\update.status
AUS:SVC Downloader:_selectPatch - found existing patch with state: null
AUS:SVC Downloader:_selectPatch - failed to apply complete patch!
AUS:SVC Downloader:downloadUpdate - no patch to download
AUS:SVC readStatusFile - status: null, path: C:\Users\shaver\AppData\Local\Mozilla\Firefox\Minefield\updates\0\update.status

Comment 3

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

I was able to reproduce by deleting the active-update.xml and updates.xml. Shaver, by chance did you do this?

Comment 4

[Mike Shaver (:shaver -- probably not reading bugmail closely)]

Not intentionally, at least.  I can check if they're there on that laptop when I get home, if I remember.

Comment 5

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

They'll likely get recreated but attaching them to this bug might shed some light. Also, attaching the update.status file might help.

Comment 6

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

The good news is that I believe I have one of my installs in this state which I can use to compare against your files and diagnose this.

Comment 7

[David Westberg]

I have the same problem. It occured while performing and update from 3.6.2 to 3.6.3.
Is there a way to restart the download ?

Comment 8

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

David, if you don't mind could you attach (using the 'Add an attachment' link above) the active-update.xml, updates.xml, and if present the update.status files which can be located with a default Windows install as follows:

for Windows Vista and above type the following in the Start Menu search box and press return.
%LOCALAPPDATA%
Then navigate to \Mozilla\Firefox\Mozilla Firefox\updates\

for Windows prior to Windows Vista type the following in the Start Menu -> run box and press return.
%USERPROFILE%
Then navigate to \Local Settings\Application Data\Mozilla\Firefox\Mozilla Firefox\updates\

The update.status file will be located in the 0 sub-directory.

After you have done this you can delete the active-update.xml file to get things working again.

Thanks

Comment 9

[David Westberg]

Attachment: Active update xml file

Comment 10

[David Westberg]

Attachment: updates xml

Comment 11

[David Westberg]

Attachment: update status file

Comment 12

[David Westberg]

Uploaded as attachment the requested information.

Comment 14

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

I have an idea on how to fix this but I need to think it through a tad more.

Comment 15

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

Attachment: patch - revert to pre 3.6 behavior

Comment 16

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

Attachment: patch - revert to pre 3.6 behavior with test

Forgot to add the test to the patch

Comment 17

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

note: I filed bug 560316 to improve the previous fallback behavior that this patch restores.

Comment 19

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

Pushed to mozilla-central
http://hg.mozilla.org/mozilla-central/rev/c2085ab04bfe

Comment 21

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

Comment on attachment 440006 [details] [diff] [review]
patch - revert to pre 3.6 behavior with test

Drivers, this fixes a regression where the update gets into a bad state where updates can't be downloaded by reverting to the behavior in Firefox 3.5 / Gecko 1.9.1. The patch is extremely small / safe and contains a test.

Comment 22

[Mike Beltzner [:beltzner, not reading bugmail]]

Bad regression, needed, a=beltzner for 1.9.2.4, please land on mozilla-1.9.2 and the GECKO1924_20100413_RELBRANCH

Comment 23

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

Pushed to mozilla-1.9.2 default
http://hg.mozilla.org/releases/mozilla-1.9.2/rev/a326f318bfbe

Pushed to mozilla-1.9.2 GECKO1924_20100413_RELBRANCH
http://hg.mozilla.org/releases/mozilla-1.9.2/rev/1002fd078a87

Comment 24

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

Pushed followup test only fix

mozilla-1.9.2 default
http://hg.mozilla.org/releases/mozilla-1.9.2/rev/03812f38fa82

Pushed to mozilla-1.9.2 GECKO1924_20100413_RELBRANCH
http://hg.mozilla.org/releases/mozilla-1.9.2/rev/be910b14c879

Comment 25

[Al Billings [:abillings - ex-MoCo]]

The checked in test for this is passing now. Is there a manual repro case for verifying this fix in 1.9.2 or will the unit test suffice in this instance?

Comment 26

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

The test should suffice and tests everything we can test by simulating the failure condition. If there were steps to reproduce it might provide some value manually verifying this fix but as it stands the test is as good as any manual verification.

Comment 27

[Al Billings [:abillings - ex-MoCo]]

I'm marking this as verified for 1.9.2 now then.

Comment 28

[Marcia Knous [:marcia]]

This is causing test failures - please see http://tinderbox.mozilla.org/showlog.cgi?log=Firefox3.6/1271883732.1271885410.4074.gz for details. Looking for what needs to be done, should we reopen or mark as a blocker, or do we need to change the test? Or is it small problem we can deal with later. Murali made made add this comment.

Comment 29

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

Marcia, that log is from 2010/04/21 14:02:12

I fixed the test failure and stated so in comment #24