Description of problem: After full update today, Chromium tabs (and the uBlock Origin extension) crash within seconds. Version-Release number of selected component (if applicable): glibc-2.31-1, chromium-browser-79.0.3945.130-1 How reproducible: Open chromium-browser with uBlock Origin installed and/or visit Javascript-heavy sites like Gmail. Steps to Reproduce: (1. Open Chromium) (2. Install uBlock Origin and activate it) (3. wait a few seconds for the "extension crashed" popup.) 4. Restart Chromium from a shell and visit mail.google.com/mail 5. Wait a few seconds until uBlock (if installed) and tab have crashed 6. see error message on Standard Output: ../../sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall 0230 To fix: Upgrade Chromium from upstream (see discussion on linked Redhat bugzilla).
Thank you for the report, the pointer and the diagnostic. Our latest version 79.0.3945.130 is actually ready as an M7 security update (bug 26103). I think we should let that go. The circumstance of this bug are obscure. Assigning to Christiaan as the registered active maintainer for the chromium-browser-stable SRPM. @Ralf : When the newer Chromium browser arrives in Cauldron, can *you* please test the specific fault you report here; and report the result.
CC: (none) => lewyssmithAssignee: bugsquad => cjwSource RPM: glibc-2.31-1.mga8 chromium-browser-79.0.3945.130-1.mga8 => chromium-browser-stable-79.0.3945.130
Confirmed: with chromium M79 on glibc 2.31, each tab crashes after a couple of seconds, except in one case all the ads crashed but the page did not. With M80, no crashes so far. Building chromium-browser-stable 80.0.3987.87 for cauldron on the build system will take a couple of hours (and may fail).
Status: NEW => ASSIGNED
(In reply to Christiaan Welvaart from comment #2) > Confirmed: with chromium M79 on glibc 2.31, ... > in one case all the ads crashed but the page did not. Sounds ideal! A selling point?
Installed 80.0.3987.87 and have been using it for the past half hour. Running stable so far, so I think we can mark this issue resolved.
Status: ASSIGNED => RESOLVEDResolution: (none) => FIXED