Lots of storage.MailBox.SpmMailMonitor

My vdsm log files are huge: -rw-r--r-- 1 vdsm kvm 1.8G Nov 22 11:32 vdsm.log And this is juste half an hour of logs: $ head -1 vdsm.log 2018-11-22 11:01:12,132+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 2 checksum failed, not clearing mailbox, clearing new mail (data='...lots of data', expected='\xa4\x06\x08\x00') (mailbox:612) I just upgraded vdsm: $ rpm -qi vdsm Name : vdsm Version : 4.20.43

On Thu, Nov 22, 2018, 13:43 Fabrice Bacchella <fabrice.bacchella@icloud.com wrote:
My vdsm log files are huge:
-rw-r--r-- 1 vdsm kvm 1.8G Nov 22 11:32 vdsm.log
And this is juste half an hour of logs:
$ head -1 vdsm.log 2018-11-22 11:01:12,132+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 2 checksum failed, not clearing mailbox, clearing new mail (data='...lots of data', expected='\xa4\x06\x08\x00') (mailbox:612)
Are you sure this is the log line? the error from line 612 should be: mailbox %s checksum failed, not clearing mailbox, clearing new mail (data=%r, checksum=%r, expected=%r) Please open a bug and attach the interesting part of the log - from starting the SPM until the first error was seen, and some errors after that. It would be useful if you cam share the compressed log somehow. We also need the contents of the inbox and outbox: For iSCSI/FC domain, the logical volumes /dev/domain-uuid/{inbox,outbox} For NFS/Gluster, the files: /rhev/data-center/mnt/server:_path/domain-uuid/dom_md/{inbox,outbox} You can copy them with dd, compress, and attach to the bug.
I just upgraded vdsm: $ rpm -qi vdsm Name : vdsm Version : 4.20.43
This started after the upgrade? As first aid fix you can clear the inbox and outbox files like this: 1. Stop vdsm on the SPM host systemctl stop vdsmd 2. Clear the mailboxes For iSCSI/FC: blkdiscard -z /dev/domain-uuid/{inbox,outbox} For NFS/Gluster: dd if=/dev/zero of=/rhev/data-center/mnt/server:_path/domain-uuid/dom_md/inbox bs=1M count=16 oflag=direct conv=fsync dd if=/dev/zero of=/rhev/data-center/mnt/server:_path/domain-uuid/dom_md/outbox bs=1M count=16 oflag=direct conv=fsync 3. Start vdsm on the SPM host systemctl start vdsmd Nir

On Thu, Nov 22, 2018 at 9:53 PM Nir Soffer <nsoffer@redhat.com> wrote:
On Thu, Nov 22, 2018, 13:43 Fabrice Bacchella < fabrice.bacchella@icloud.com wrote:
My vdsm log files are huge:
-rw-r--r-- 1 vdsm kvm 1.8G Nov 22 11:32 vdsm.log
And this is juste half an hour of logs:
$ head -1 vdsm.log 2018-11-22 11:01:12,132+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 2 checksum failed, not clearing mailbox, clearing new mail (data='...lots of data', expected='\xa4\x06\x08\x00') (mailbox:612)
Are you sure this is the log line? the error from line 612 should be:
mailbox %s checksum failed, not clearing mailbox, clearing new mail (data=%r, checksum=%r, expected=%r)
Please open a bug and attach the interesting part of the log - from starting the SPM until the first error was seen, and some errors after that.
It would be useful if you cam share the compressed log somehow.
We also need the contents of the inbox and outbox:
For iSCSI/FC domain, the logical volumes
/dev/domain-uuid/{inbox,outbox}
For NFS/Gluster, the files:
/rhev/data-center/mnt/server:_path/domain-uuid/dom_md/{inbox,outbox}
You can copy them with dd, compress, and attach to the bug.
I just upgraded vdsm: $ rpm -qi vdsm Name : vdsm Version : 4.20.43
This started after the upgrade?
As first aid fix you can clear the inbox and outbox files like this:
Or this less intrusive fix - restart vdsm on the host with host id 2. To find the host with host id 2, look for this log on all hosts: [vdsm.api] START connectStoragePool(..., hostID=2, ... This will clear the mailbox with the bad checksum. Once cleared the SPM will stop complaining about the bad checksum. Nir

On Thu, Nov 22, 2018 at 10:17 PM Nir Soffer <nsoffer@redhat.com> wrote:
On Thu, Nov 22, 2018 at 9:53 PM Nir Soffer <nsoffer@redhat.com> wrote:
On Thu, Nov 22, 2018, 13:43 Fabrice Bacchella < fabrice.bacchella@icloud.com wrote:
...
As first aid fix you can clear the inbox and outbox files like this:
Or this less intrusive fix - restart vdsm on the host with host id 2.
To find the host with host id 2, look for this log on all hosts:
[vdsm.api] START connectStoragePool(..., hostID=2, ...
This will clear the mailbox with the bad checksum. Once cleared the SPM will stop complaining about the bad checksum.
Finally, when you find host 2, we need vdsm log from this host - the log may help to understand why there was a bad checksum, and why the bad checksum was not fixed when the next message was sent to the mailbox.

Hello Nir, I recently upgrade oVirt engine to 4.4.9 from 4.3.10 (hosts will follow ASAP). I found out in the vdsm.log same strange messages: 2022-01-06 10:35:41,333+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 65 checksum failed, not clearing mailbox, clearing new mail (data='\xff\xff\xff\xff\...lot of data...\x00\x00\x00', checksum=<function checksum at 0x7f15892967d0>, expected='\xbfG\x00\x00') (mailbox:603) 2022-01-06 10:35:41,334+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 66 checksum failed, not clearing mailbox, clearing new mail (data='\x00\x00\x00\x00\...lot of data...\xff\xff', checksum=<function checksum at 0x7f15892967d0>, expected='\x04\xf0\x0b\x00') (mailbox:603) We have 7 iSCSI and 1 NFS (old export domain). lvscan | grep inbox ACTIVE '/dev/8ee251ed-a50b-4235-a279-0829d7e8e9a0/inbox' [128.00 MiB] inherit ACTIVE '/dev/dfd0134d-2d63-432f-af9f-b60aa6e1fefb/inbox' [128.00 MiB] inherit ACTIVE '/dev/0633a601-d73a-4750-8ff6-c893fe064469/inbox' [128.00 MiB] inherit ACTIVE '/dev/47814a07-b6bc-4d1f-b01d-1919c07878a6/inbox' [128.00 MiB] inherit ACTIVE '/dev/1c61030e-91a5-4e17-af37-92e1dada7c19/inbox' [128.00 MiB] inherit ACTIVE '/dev/a74c32e3-ddd5-4f06-9d9c-3ba7aa153d98/inbox' [128.00 MiB] inherit ACTIVE '/dev/333a7e7e-0da9-4db8-b486-fea1f1ee8171/inbox' [128.00 MiB] inherit I am not fully sure which inbox/outbox I should try to clear manually. Can you try help me with this please?

Le 22 nov. 2018 à 20:53, Nir Soffer <nsoffer@redhat.com> a écrit :
On Thu, Nov 22, 2018, 13:43 Fabrice Bacchella <fabrice.bacchella@icloud.com <mailto:fabrice.bacchella@icloud.com> wrote: My vdsm log files are huge:
-rw-r--r-- 1 vdsm kvm 1.8G Nov 22 11:32 vdsm.log
And this is juste half an hour of logs:
$ head -1 vdsm.log 2018-11-22 11:01:12,132+0100 ERROR (mailbox-spm) [storage.MailBox.SpmMailMonitor] mailbox 2 checksum failed, not clearing mailbox, clearing new mail (data='...lots of data', expected='\xa4\x06\x08\x00') (mailbox:612)
blkdiscard -z /dev/domain-uuid/{inbox,outbox}
This command solved my problem for now. They was too much logs to send. So I purged all my logs and will open a bug if this problem come again.
participants (3)
-
Fabrice Bacchella
-
Nir Soffer
-
Petr Kyselák