Comment # 58 on bug 1188896 from LTC BugProxy

------- Comment From geraldsc@de.ibm.com 2022-02-15 08:36 EDT-------
(In reply to comment #53)
> so we have the "panic_on_taint=5" in the kernel cmdline in the OBS builds at
> the moment, but the VMs do not seem to crash or stop. Did I misinterpret the
> comments in that regard ?

No, the expectation would be that adding this kernel parameter will lead to a
kernel panic as soon as you hit the "Bad page state" warning.

I would expect that you did not (yet) hit it, do the systems have any other
problems, or just run fine?

Sometimes adding debug options subtly change timing, which does not allow to
reproduce a problem, but I do not see how adding panic_on_taint kernel
parameter would do that.

Maybe just wait a little longer :-)

Have there also been changes other than just adding the kernel parameter? Maybe
a kernel update? Then in theory the bug might also have been silently fixed,
but I think that is very unlikely and just wishful thinking...