Details
-
Bug
-
Resolution: Cannot Reproduce
-
Critical
-
None
-
None
Description
I can make this happen reproducibly after following the reproduction steps of
CORE-18516.
The ros session will crash then reproducibly during OS shutdown via the startbutton.
We observed that already maybe 1.5years ago, also after we tried building ros on ros.
Today 2 of my VMs died the same way, after having attempted to build ros on ros and then attempting to shutdown via startmenu.
And the DarkFire remembered slightly, that he might has seen something similar in the past when he tried to shutdown after trying to copy very big files within the session.
A common denominator of those test-cases could be "maxing out the memory to a point that the MM attempted to free some memory again".
It will not happen if you just install ros, and then do the shutdown. You ust stress ros before like described.
0.4.15-dev-5188-gabaf0d1_winLogonCrashDuringShutdown_follwedbyBSOD.log
Yes the reproduction steps are very time-consuming, but @hbelusca encouraged me to create the ticket. I am sure, that I will be able to retest. But I wouldn't want to do that too often!
Attachments
Issue Links
- relates to
-
CORE-18516 Regression, building ros on ros RosBE2.1.6 'ninja bootcd -j1' does fail since some 0.4.15-dev commit at step237 during compiling boot/environ with random errors
- Open
-
CORE-19332 first-chance-exception in ntdll tiggered during shutdown, shutdown hangs then (after 'successful' compilation of ros on ros)
- Untriaged