8.0.2 -> 8.0.3 works but it takes 30s until the bar appears, looks like fixing this introduced an error.
Oct 25 10:03:33.000 [notice] Catching signal TERM, exiting cleanly.
JavaScript error: chrome://torbutton/content/tor-circuit-display.js, line 436: TypeError: myController is null
Fontconfig warning: "/home/tbb/tor-browser_en-US8.0.2/Browser/TorBrowser/Data/fontconfig/fonts.conf", line 85: unknown element "blank"
1540461904853 addons.webextension. WARN Loading extension 'null': Reading manifest: Error processing background.persistent: Event pages are not currently supported. This will run as a persistent background page.
1540461905523 addons.xpi WARN Please specify whether you want browser_style or not in your options_ui options.
1540461906559 addons.xpi WARN Please specify whether you want browser_style or not in your options_ui options.
Fontconfig warning: "/home/tbb/tor-browser_en-US8.0.2/Browser/TorBrowser/Data/fontconfig/fonts.conf", line 85: unknown element "blank"
Oct 25 10:05:09.477 [notice] Tor 0.3.4.8 (git-da95b91355248ad8) running on Linux with Libevent 2.1.8-stable, OpenSSL 1.0.2p, Zlib 1.2.11, Liblzma N/A, and Libzstd N/A.
Do you need more info? (happy to test more:) Filed #28198 (moved) for the Fontconfig lines.
What do you mean with "looks like fixing this introduced an error"?
It was my impression that the alpha shuts down faster, but i might be wrong. Was thinking of #28113 (moved) but it looks unrelated.
The previous alpha updates worked for you?
Yes, it worked until now.
The issue is that 8.5a3 which is what is important here just fixes two moz sec bugs in the js engine on linux. so, i don't see how that can lead to your problems.
It seems the browser is killing itself because the shutdown takes too long which messes with the updates.
Because tor takes 30 seconds to exit? Maybe this should go or the timeout needs to be raised.
Does it make a difference if you apply partial updates or full updates?
The updates above were all incremental. Fished this from the log: Testing the upgrade from 8.5a1 it downloads 82mb, so incremental updates are only supported for the direct successor it seems. It worked well and also restarted as it should.
So the answer to your question is yes, it does seem to make a difference.
However when i retested the update from 8.5a1 -> 8.5a4 just now it just exited without restarting and without error. It just exits when it should restart instead. On the next start the update goes well, so this is minor. The last line in the terminal was
Oct 25 11:24:52.000 [notice] Catching signal TERM, exiting cleanly.
Do you find a last-update.log file in your Tor Browser when the update fails? That might give us some clue about what is going on. It might be worth enabling app.update.log as well and check in the browser console that everything is fine before the browser is supposed to get restarted.
Trac: Cc: N/Ato mcs, brade Status: new to needs_information
All updates are ok, restarting fails. When TB crashes the update is not invoked and no last-update.log or app.update.log are created. Every last-update.log file i found ends with
succeeded
calling QuitProgressUI
The different errors on exit do not only happen on updates, for example below with 8.0.3.
My computer is slow, maybe it can be fixed by raising some timeout?
Also folder name seems to have an effect (#28199 (moved)).
Updating 8.5a3
AUS:SVC Downloader:onStopRequest - setting state to: pendingAUS:SVC getCanStageUpdates - staging updates is disabled by preference app.update.staging.enabledAUS:SVC showUpdateDownloaded - Notifying observers that an update was downloaded. topic: update-downloaded, status: pendingDiscarding onCreatedNavigationTarget for 27: received source tab data without any created tab data available[10-25 21:26:31] Torbutton DBUG: Got timer update, but no cookie change.Discarding onCreatedNavigationTarget for 27: received source tab data without any created tab data available[10-25 21:27:31] Torbutton DBUG: Got timer update, but no cookie change.[10-25 21:27:46] Torbutton INFO: controlPort >> 650 STREAM 51 CLOSED 6 <scrubbed>:443 REASON=DONE[10-25 21:27:53] Torbutton INFO: controlPort >> 650 STREAM 8 CLOSED 6 <scrubbed>:443 REASON=DONE[10-25 21:28:31] Torbutton DBUG: Got timer update, but no cookie change.AUS:SVC getCanApplyUpdates - testing write access /media/user/tbb/tor/manual/tor-browser_en-US/Browser/TorBrowser/UpdateInfo/update.testAUS:SVC getCanApplyUpdates - able to apply updatesAUS:SVC getCanStageUpdates - staging updates is disabled by preference app.update.staging.enabledAUS:SVC readStatusFile - status: pending, path: /media/user/tbb/tor/manual/tor-browser_en-US/Browser/TorBrowser/UpdateInfo/updates/0/update.status[10-25 21:29:31] Torbutton DBUG: Got timer update, but no cookie change.<Restart>Oct 25 21:32:14.000 [notice] Owning controller connection has closed -- exiting now.Oct 25 21:32:14.000 [notice] Catching signal TERM, exiting cleanly.Crash Annotation GraphicsCriticalError: |[C0][GFX1-]: Receive IPC close with reason=AbnormalShutdown (t=473.767) [GFX1-]: Receive IPC close with reason=AbnormalShutdown[Child 15153, Chrome_ChildThread] WARNING: pipe error (3): Connection reset by peer: file /var/tmp/build/firefox-0c004047822a/ipc/chromium/src/chrome/common/ipc_channel_posix.cc, line 353
Exitin 8.0.3 without (visible) update
FATAL ERROR: AsyncShutdown timeout in profile-before-change Conditions: [{"name":"Crash Reporter: blocking on minidumpgeneration.","state":"(none)","filename":"/var/tmp/build/firefox-858720263bed/ipc/glue/CrashReporterHost.cpp","lineNumber":189,"stack":"Minidump generation"}] At least one completion condition failed to complete within a reasonable amount of time. Causing a crash to ensure that we do not leave the user with an unresponsive process draining resources.
WARNING: No crash reporter available
[Parent 6862, Main Thread] ###!!! ABORT: file /var/tmp/build/firefox-858720263bed/ipc/glue/CrashReporterHost.cpp, line 189