[developer] Re: [openzfs/openzfs] 8727 Native data and metadata encryption for zfs (#489)

(No changes, just fixed conflicts with master) -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/489#issuecomment-364290139 -- openzfs-developer

[developer] Re: [openzfs/openzfs] 8727 Native data and metadata encryption for zfs (#489)

Well, that's good.. Not sure what to do next then, perhaps @GernotS can provide a dump file, and someone who actually knows how to use mdb (I sure don't) can be coaxed into glancing at it? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

Maybe even `ZIO_GANG_BIT`, etc. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/505#issuecomment-364269552 -- openzfs-developer Archives:

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

@grwilson this looks good to me! But... But! I've wondered only __just now__ if we could have best of both approaches by pre-defining the child type bits. E.g. ``` #define ZIO_CHILD_BIT_GANGZIO_CHILD_BIT(ZIO_CHILD_GANG) ``` ... or something like that. Maybe that could help to make the code

[developer] [openzfs/openzfs] 9080 recursive enter of vdev_indirect_rwlock from vdev_indirect_remap() (#542)

Reviewed by: Matthew Ahrens Reviewed by: George Wilson Description A scenario came up where a callback executed by vdev_indirect_remap() on a vdev, calls vdev_indirect_remap() on the same vdev and tries to reacquire vdev_indirect_rwlock that

[developer] [openzfs/openzfs] 9079 race condition in starting and ending condesing thread for indirect vdevs (#541)

Reviewed by: Matt Ahrens Reviewed by: Pavel Zakharov The timeline of the race condition is the following: [1] Thread A is about to finish condesing the first vdev in spa_condense_indirect_thread(), so it calls the

[developer] Re: [openzfs/openzfs] 9077 zloop misses core files because they're no longer written into cwd (#540)

Looking at our internal issue, I think we have coreadm configure to put core files in `/var/crash`. We added this so the core files would end up in the CWD and would get collected by our automation (with the other zloop files/logs/etc). So, I think the answer is yes.. @jwk404 can you confirm?

[developer] Re: [openzfs/openzfs] 9077 zloop misses core files because they're no longer written into cwd (#540)

Does this still work if coreadm is configured to put dumps in a different location? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/540#issuecomment-364252166

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

youzhongyang commented on this pull request. It looks good to me. Thanks. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/505#pullrequestreview-95227067

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

@youzhongyang @avg-I can you take another look to make sure I've address all of your concerns? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/505#issuecomment-364239237

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

Your comments are considered and I will be pushing a new commit with the updated logic. I've tried several variations to the suggestions to find which looks the best and has the least amount of impact. The suggested `ZIO_CHILD_BIT` option looks the best so I've gone with that option. -- You

[developer] [openzfs/openzfs] 9077 zloop misses core files because they're no longer written into cwd (#540)

Reviewed by: Serapheim Dimitropoulos Reviewed by: Prakash Surya To make zloop for resilient to coreadm configuration differences on the various illumos distributions, we should enable per process core dumps at the start of zloop. Upstream bug:

[developer] Re: [openzfs/openzfs] 8115 parallel zfs mount (v2) (#536)

@prakashsurya pushed 1 commit. c8a3e65 Fix cstyle; use tab for indentation, not spaces -- You are receiving this because you are subscribed to this thread. View it on GitHub:

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

so our comments will not be considered? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/505#issuecomment-364225821 -- openzfs-developer

[developer] Re: [openzfs/openzfs] 8857 zio_remove_child() panic due to already destroyed parent zio (#505)

ahrens approved this pull request. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/505#pullrequestreview-95183334 -- openzfs-developer

[developer] Re: [openzfs/openzfs] 8115 parallel zfs mount (v2) (#536)

@prakashsurya pushed 2 commits. 20d3e56 Missed a mutex_{enter,exit} pair, causing a crash 3a83672 Fix lint; add guards for _TASKQUSER in lgrp_user.h -- You are receiving this because you are subscribed to this thread. View it on GitHub:

[developer] Re: [openzfs/openzfs] 8727 Native data and metadata encryption for zfs (#489)

This passed zfstest and zloop using our internal test automation. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/489#issuecomment-364181843 --