Date: Fri, 1 Apr 2022 14:01:20 -0400 From: Rick Summerhill <rrsum@summerhill.org> To: questions@freebsd.org Subject: Re: zpools disappearing on reboot with 13.1Beta3 (and now RC1) Message-ID: <e0dadd98-8447-7d77-f6a0-75d26d744f5a@summerhill.org> In-Reply-To: <a353e5ed-2506-273f-d198-ffca99ad3657@summerhill.org> References: <383af7d8-294f-bd43-9a4e-660b56261d3e@summerhill.org> <20220329210600.b4e405ae3a5006eceaaf44cb@sohara.org> <a353e5ed-2506-273f-d198-ffca99ad3657@summerhill.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 3/29/22 6:20 PM, R Richard Summerhill wrote: > On 3/29/22 16:06, Steve O'Hara-Smith wrote: >> On Tue, 29 Mar 2022 15:37:19 -0400 >> Rick Summerhill <rrsum@summerhill.org> wrote: >> >>> I'm running a server under 13.1Beta3 and it has 2 zpools, one a mirror >>> and the other a raidz1. When I reboot the server, both pools disappear >>> but come back when imported. Do I have something set wrong on the >>> server? >> Does your /etc/rc.conf contain >> >> zfs_enable="YES" > > > Yes, it does. Let me give you a bit more info. I rebuilt 13.1Beta3 > with just the mirror pool. Rebooted several times to test everything > and it came back correctly each time. Then I added one of these 5x1 > port multiplier arrays to an esata port (this is one of my test boxes) > and rebooted. All the disks were there (they are all labeled > appropriately), and the mirror zpool (which uses the labels) was not > there. Importing it, however, brought it back. Then I created the > raidz1 pool on the multiplier array. Both pools showed up with a good > status. On rebooting, the disks are there and the pools are not, but on > importing, both come back. > > Not that the multiplier array has been working fine under 13.0Release > and ZFS. > > I upgraded to 13.1RC1 and the condition persists. Also, dmesg shows ZFS filesystem version: 5 ZFS storage pool version: features support (5000) pid 49 (zpool), jid 0, uid 0: exited on signal 6 and the disks for the pools appear after that in dmesg. I also tried adding the following lines to loader.conf kern.cam.boot_delay="5000" # Delay (in ms) of root mount for CAM bus kern.cam.scsi_delay="2000" # Delay (in ms) before probing SCSI That did not solve the problem, however. Any thoughts? Rick -- Rick Summerhill Retired, Chief Technology Officer, Internet2 10233 Timberhill Rd Manchester, MI 48158 USA Home: 734-428-1422 Web: http://www.rick.summerhill.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?e0dadd98-8447-7d77-f6a0-75d26d744f5a>