Marking as "ReleaseBlock-Stable" to investigate this further. Please note M57 (57.0.2987.98) went out to small percentage of users last week on Thursday (03/09).

I guess if this starts happening again he could use the "--no-sandbox --enable-logging" flags and paste the logs here.

The --disable-gpu command-line flag would probably make Chrome work again in this case, though it'll be slower than normal.

I've also seen this before when views decides not to show its tree. If we can get a repro, I can verify that.

Ref #4, this last happened with  http://crbug.com/625076  during Canary.

more info:
It's a GTX 770 / i7 4770K system running GPU driver 376.53
current Chrome version is 57.0.2987.98 (64-bit), but //crashes is empty, I think reinstall wiped
the old version and crash records. I will definitely get screencaps if it happens again

Your bug is labelled as Stable ReleaseBlock, pls make sure to land the fix and get it merged into the release branch ASAP. Thank you!

Know that this issue shouldn't block the release?  Remove the ReleaseBlock-Stable label or move to M58.

Level setting here: You should never mark a blocker based on a single, uncorroborated, and unreproducible report. If we get a noticeable volume of similar reports or a reproduction case, then a blocker *may* be appropriate. But for now I'm resetting the flags to something more accurate.

This is almost certainly not a problem with the installer. Possibly something with views (as Rob mentioned), GPU, or some odd internals thing. We'll need more info to narrow it down.

Added TE-NeedsTriageHelp to remove from TE-Bucket until there is a proper steps to reproduce it.

Pretty sure it's not a GPU bug if a bunch of reinstall and a lunch break can fix the issue.

This bug doesn't seem actionable. Is there any reason to keep it open?

Agreed.