Selecting 17,415 files in files app crashes the chromebook |
||||||||
Issue descriptionIMPORTANT: Your crash has already been automatically reported to our crash system. Please file this bug only if you can provide more information about it. Chrome Version: 67.0.3383.0 Operating System: Linux 4.4.125-13322-ge7f21383b904 URL (if applicable) where crash occurred: Can you reproduce this crash? Yes. 100% of the time. What steps will reproduce this crash? 1. Install "wicket good unarchiver" [to open a tar file] 2. go into the "files" app 3. open file "m2_baseline_system_icons.tar" in files app (https://drive.google.com/open?id=1pSRmjrUdqjfCw5IaJl6uzrX1RA3KbXoq) 4. select all 17,415 files in the newly open tar-file overview with CTRL+A Furthermore subsequent crashes were not recorded under chrome://crashes. Only the first instance! ****DO NOT CHANGE BELOW THIS LINE**** Crash ID: crash/b069f8d81111925d
,
Apr 11 2018
Then I suppose this type of crash is not logged at all. Or is there another location where I can look for crash logs?
,
Apr 15 2018
,
Apr 17 2018
Tested with 67.0.3375.0 on Cave - UI becomes unresponsive for several seconds and recovers. Tested with 67.0.3366.3 on Eve - Works fine. Going to give it a shot with the latest canary on Cave (waiting for update now) If that doesn't work I'll reimage my eve to the specific version (trying to avoid that as I have other stuff being tested on that device right now) Is there anything here that requires the bug to remain restricted? I don't see anything.
,
Apr 17 2018
67.0.3390.0 on Cave is the same as 3375 -- Unresponsive then recovers. +vapier@ for wicked good unarchiver. At a minimum, selecting a large number of files in an archive shouldn't stall the system like this. I'll keep hunting for a repro of the crash
,
Apr 17 2018
Allowing triage team to prioritize and assign owner.
,
Apr 18 2018
I suspect it is a browser hang than a crash. UI thread seems really busy with the select-all operation. If chrome fails to respond to session_manager's liveness check ping, it would get killed. Could we file a feedback? We can look at syslog part of the report and looking for "Browser hang detected!". That line should be present when chrome is killed due to failed liveness check ping.
,
Apr 19 2018
Please reproduce again and send feedback with Alt+Shift+I so we get the syslog in the reports.
,
Apr 19 2018
i don't think WGU is really a factor here. the Files App/API/structure should take care of staying interactive regardless of the requests.
,
Apr 19 2018
I can't repro on 67.0.3383.0 on Eve either. The UI stalls for a few seconds but recovers. I think xiyuan@ is right; some operation is blocking the UI and in some circumstances leads to the browser getting killed by hang detection. pashkov@ can you repro the crash and then file a feelback report with alt+shift+i? That should allow us to see more detailed logs that will let us confirm or deny the suspicion. I'm also going to unrestrict this since I don't see anything confidential here.
,
May 10 2018
pashkov@ - can you repro and file a feedback report so we can further investigate? Thanks :)
,
May 10 2018
Sorry, I cannot reproduce it either anymore.
,
May 17 2018
|
||||||||
►
Sign in to add a comment |
||||||||
Comment 1 by derat@chromium.org
, Apr 11 2018Labels: OS-Chrome