Potential crash triggered by malformed/incorrect PDF |
||||||||||
Issue descriptionWe got a report from a beta user that they are noticing certain types of PDFs are crashing their chrome browser on chromeos. These crashes are from samus/beta. 4f72b04800000000 e25fd85400000000 691a318800000000 a3adbe9400000000 I'm tagging this as stable blocker until we get clarity on the scope of the issue. This customer is in UK and information sharing may be slightly laggy.
,
Mar 11 2016
,
Mar 11 2016
Found these errors in the logs repeat multiple times before the crash... ./log/ui/ui.20160224-070551-[1:5:0100/000000:FATAL:memory.cc(19)] Out of memory. size=4198400 ./log/ui/ui.20160224-070551:[10624:10624:0224/141955:ERROR:gles2_cmd_decoder.cc(13979)] [.CommandBufferContext.Compositor-0xf899b579840]GL ERROR :GL_INVALID_OPERATION : glCreateAndConsumeTextureCHROMIUM: invalid mailbox name ./log/ui/ui.20160224-070551:[10624:10624:0224/141955:ERROR:gles2_cmd_decoder.cc(7639)] [.CommandBufferContext.Compositor-0xf899b579840]RENDER WARNING: texture bound to texture unit 0 is not renderable. It maybe non-power-of-2 and have incompatible texture filtering. ./log/ui/ui.20160218-072319-[3167:3167:0219/071003:ERROR:gles2_cmd_decoder.cc(13414)] [.CommandBufferContext.CompositorWorker-0x1d2570944000]GL ERROR :GL_INVALID_VALUE : glCopySubTextureCHROMIUM: source texture has no level 0 ./log/ui/ui.20160218-072319-[3167:3167:0219/071004:ERROR:gles2_cmd_decoder.cc(13414)] [.CommandBufferContext.RenderWorker-0x82005aeac60]GL ERROR :GL_INVALID_VALUE : glCopySubTextureCHROMIUM: source texture has no level 0 ./log/ui/ui.20160218-072319:[2008:3160:0219/071006:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.CHXsyi failed: Too many open files ./log/ui/ui.20160218-072319:[2008:3160:0219/071006:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.kAyl5g failed: Too many open files ./log/ui/ui.20160218-072319:[2008:3160:0219/071006:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.q1YeCf failed: Too many open files
,
Mar 11 2016
,
Mar 12 2016
,
Mar 14 2016
Crashes in c#0 do not seem related to this issue .. Can you get confirmation if this problem is repro on latest build? File a feedback if still seen There was a critical fix in Samus related to FD Albert/Lei, can you help triage/evaluate?
,
Mar 14 2016
Can we get one of the PDFs that triggers the crash? Some PDF triggering a crash does NOT justify P0. Dropping pri.
,
Mar 14 2016
,
Mar 14 2016
+royans@ - as requested in c#7, please provide the PDF that crash your system. - provide the repo steps? from your video seems like you are remote desktop to a PC and create the pdf with MS powerpoint, save locatly and open in cros ? The 3 crash ids you provided point to https://b.corp.google.com/u/0/issues/27591108 but seems unrelated.
,
Mar 14 2016
,
Mar 14 2016
File sent by the customer is here (inside archive.zip) https://drive.google.com/corp/drive/u/0/folders/0Bx12NVyIouI0Q1BwOTg2X2NaWm8 Agreed on all the comments. We can remove stable blocker and continue with the update. The customer who reported this has been requested to block AU on their side and provide more information for us to follow up on this bug. If we don't hear back from the customer in next one week, and if there are no other reports, lets close this bug as well. Sorry for the false alarm... but I wanted to make sure we over communicate for such last minute beta bugs.
,
Mar 14 2016
+restricted view The pdf is in the zip file under c#1
,
Mar 14 2016
David tested on 49.0.2623.95/7934.60.0 samus: Could not repro the PDF issue. I have a samus with 7647.84.0/48.0.2564.116 and it did not crash with the PDF either.
,
Mar 22 2016
removing block label and reducing priority of this due to lack of more info from customer.
,
Oct 13 2016
Checking in to see if it's still happening. I cannot reproduce.
,
Oct 13 2016
We couldn't reproduce at the time, so if you can't reproduce now either, let's just close this |
||||||||||
►
Sign in to add a comment |
||||||||||
Comment 1 by royans@chromium.org
, Mar 10 2016