Re: FreeBSD Errata Notice FreeBSD-EN-22:21.zfs

I spoke to Richard today, and asked him your question on your behalf. He said the following:

"Most of the time, the issue does not go to disk. When it does they are not well understood yet. He would probably know if it did since things could go horribly wrong, but if he wants to try to be sure, he could try using zdb -bcc ... to try to verify certain things about his pool, but again, the results of undefined behavior in the btree code is not well understood. It is a very complex thing to analyze and I have not had time to do that. There are a couple of bug reports in GitHub that I suspect are related, but cannot prove currently."


Jonathan Vasquez
PGP: 34DA 858C 1447 509E C77A D49F FB85 90B7 C4CA 5279
Sent with ProtonMail Secure Email



------- Original Message -------
On Tuesday, November 1st, 2022 at 19:45, Thomas Zander <riggs@freebsd.org> wrote:


> On Tue, 1 Nov 2022 at 23:21, FreeBSD Errata Notices
> errata-notices@freebsd.org wrote:
> 
> > Systems with debug kernels may sometimes detect this issue after a kernel
> > memory corruption has happened. When they do, they will trigger a kernel
> > panic to protect the system from further damage. The following is printed
> > to dmesg at the time of the panic:
> 
> 
> On production systems, is there any way to check for corruptions? Will
> a "zpool scrub" find problems?