I'm not so sure we want to ASSERT in any case. It seems to me that regardless
of how the pool may be damaged importing it should never result in a panic.
That needs to be a safe operation. This patch was just a tiny tiny baby step
in that direction. How about we just use `zfs_panic_recover()` in the ENOENT
case for this and the `ddt_object_info()` call a few lines down.
---
Reply to this email directly or view it on GitHub:
https://github.com/openzfs/openzfs/pull/19#issuecomment-151322986
_______________________________________________
developer mailing list
[email protected]
http://lists.open-zfs.org/mailman/listinfo/developer
