Chris Siebenmann
Our busiest #OmniOS NFS fileserver can (& has) lock up without any trigger we can try to mitigate. It seems clearly a deep kernel issue.
Another fileserver lockup tonight has me now rather gloomy about our future with #OmniOS. We have a bad problem and no clear solutions.
-
-
- View other replies
-
Attempts to get crash dumps have failed (takes too long). Merely find the bug(s) in the kernel will be hard enough, never mind a fix.
-
Finding and fixing deep kernel bugs may well be beyond the bounds of normal
@OmniTI support, and it's not clear we can afford it anyways. -
And it's not as if our
#OmniOS lockup is reproducible or leaves any particular evidence behind; we've failed at both. It just hits randomly. -
Running a production fileserver that we know locks up randomly isn't viable. But there's no clear mitigation, especially an inexpensive one.
-
One endgame is that we'll be forced to abandon
#OmniOS for either FreeBSD or in a really bad case a non-ZFS solution. - View other replies
-
-
@ecdysone It happens on only one of three identical fileservers (different pools/loads/etc) and not in testing. That's part of the problem. - View other replies
- Show more
-
-
@thatcks :( ZFS related?
eric saxby
Guillaume Ceccarelli
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.