@GernotS Thanks.  I'm still not sure what's going on with the sa.c dump, and 
have a few more questions for you:

Can you describe how you reproduce the sa.c dump?  I could gather some of the 
sequence below but if you can confirm that and fill in any other details that 
would be helpful.

It looks like you created `/zones/data` from a zfs receive that is from a raw 
encrypted send stream?  At the time of the panic you're doing `ls --color -lhB 
--color /zones/data/` and it's doing a stat() syscall on `/zones/data/fritz`.  
Is the `ls` essentially all you did after booting the machine?

Could you try doing `ls -l /zones/data/fritz` (preferably just after doing the 
"zfs receive" to create zones/data), and see if that also panics?  This will 
help me tell if the problem is a race condition with the reading vs with the 
data on-disk.

Could you try getting a "full" dump by running `dumpadm -c all` before 
triggering the panic?  This will capture the encrypted data in the ARC as well 
(it will be bigger, but it looks like your machine only has 8GB RAM so 
hopefully it's still palatable).

-- 
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-377078558
------------------------------------------
openzfs: openzfs-developer
Permalink: 
https://openzfs.topicbox.com/groups/developer/discussions/T91797982fdd5b7d9-Mb5e6e80e2912cc97d9899543
Delivery options: https://openzfs.topicbox.com/groups

Reply via email to