Project

General

Profile

Bug #3199

PFS label not found panic

Added by tse 2 months ago. Updated about 2 months ago.

Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
Start date:
08/03/2019
Due date:
% Done:

0%

Estimated time:

Description

Are there any suggestions for what I should do at this point?

Boot stops at
hammer2_mount: PFS label not found

Then enters the debugger, saying: stopped at hammer2_chain_lastdrop

Can I reboot into rescue disk and replace "ROOT" with a recent snapshot
label?

The ahci0.5 (da1) disk_rw timeout is a seperate issue, and one I was giving
up on solving. Linux could read that internal disk, but I never used it
anyway

A second issue though is on upgrading to 5.6, dfly would no-longer shutdown
gracefully - it would hang after printing the reboot + pf service stop
message to the console

Ironically I was going through a compile-install-reboot bisect to find
where the hang started, but probably force-rebooted to quickly after the
shutdown hangs, resulting in this panic

-- At least, that's what I imagine the cause of the panic to be ;)

Any suggestions or pointers for fixing any of those issues would be
appreciated,

Thanks


Files

20190803_214041.jpg (1.62 MB) 20190803_214041.jpg tse, 08/03/2019 12:12 PM

History

#1

Updated by tse about 2 months ago

The hang on shutdown is caused by doing `doas reboot`. Any other command does a clean shutdown -- i.e. `sudo reboot`, `doas poweroff`, etc

So there is something wrong with `doas reboot`?

I still don't know what caused the hard disk to be inaccessible. Applying the DA_Q_NO_SYNC_CACHE, etc quirks in scsi_da.c does not fix it. But... switching bios to use `ide` rather than `sata` mode does allow dfly to recognize the disk

#2

Updated by tse about 2 months ago

Sorry, using `ide` does expose two new disks in /dev/serno, but attempting to format the first gives

/home/tse# newfs_hammer2 /dev/serno/50026B725502A482
Volume /dev/serno/50026B725502A482 size 119.24GB
write: Input/output error

Which shows the right size for the problematic disk. And attempting the second 'disk' causes the computer to panic shortly afterwards

/home/tse# newfs_hammer2 /dev/serno/KWFF3M72041
Volume /dev/serno/KWFF3M72041 size 3.50GB
write: Input/output error

Aug 21 12:23:42 beloved kernel: intr 19 on cpu3 at 40001/40000 hz, livelocked limit engaged!
Aug 21 12:23:42 beloved kernel: ad10: FAILURE - READ_DMA timed out LBA=250069677
Aug 21 12:23:42 beloved kernel: ad10: timeout waiting to issue command
Aug 21 12:23:42 beloved kernel: ad10: error issuing READ_DMA command
Aug 21 12:23:42 beloved kernel: intr 19 on cpu3 at 17610/20000 hz, livelock removed
Aug 21 12:23:42 beloved kernel: ad10: timeout waiting to issue command
Aug 21 12:23:42 beloved kernel: ad10: error issuing READ_DMA command
Aug 21 12:23:42 beloved kernel: ad10: timeout waiting to issue command
Aug 21 12:23:42 beloved kernel: ad10: error issuing READ_DMA command
Aug 21 12:23:42 beloved kernel: ad10: timeout waiting to issue command
Aug 21 12:23:42 beloved kernel: ad10: error issuing READ_DMA command
Aug 21 12:23:42 beloved kernel: ad10: 122104MB <KINGSTON SMSM151S3128GD S9FM01.8> at ata5-master UDMA33
Aug 21 12:23:42 beloved kernel: acpi_wmi0: <ACPI-WMI mapping> on acpi0
Aug 21 12:23:42 beloved kernel: acpi_wmi1: <ACPI-WMI mapping> on acpi0
Aug 21 12:23:42 beloved kernel: atkbdc0: <Keyboard controller (i8042)> port 0x64,0x60 irq 1 on acpi0
Aug 21 12:23:42 beloved kernel: atkbd0: <AT Keyboard> irq 1 on atkbdc0
Aug 21 12:23:42 beloved kernel: kbd0 at atkbd0
Aug 21 12:23:42 beloved kernel: ad10: timeout waiting to issue command
Aug 21 12:23:42 beloved kernel: ad10: error issuing READ_DMA command
Aug 21 12:23:42 beloved kernel: ad10: reading primary partition table: error accessing offset 000000000000 for 512

Also available in: Atom PDF