Bug 12366 - aoe: unknown cmd 240 in syslog
Summary: aoe: unknown cmd 240 in syslog
Status: VERIFIED CODE_FIX
Alias: None
Product: Networking
Classification: Unclassified
Component: IPV4 (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: Ed Cashin
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-01-05 10:09 UTC by karaluh
Modified: 2009-01-18 06:11 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.24
Subsystem:
Regression: ---
Bisected commit-id:


Attachments
ignore vendor extention type command responses (742 bytes, patch)
2009-01-13 07:38 UTC, Ed Cashin
Details | Diff

Description karaluh 2009-01-05 10:09:03 UTC
Distribution: Kubuntu 8.04
Hardware Environment: Welland ME-747K-SI

Software Environment:
cat /etc/fstab
[...]
/dev/etherd/e0.0 /mnt/backup/ ext3 relatime 0 0
[...]

Problem Description:
I see lots of aoe: unknown cmd 240 errors in syslog, the device seems to be working fine.
Comment 1 Ed Cashin 2009-01-12 14:37:09 UTC
This sounds like it could be one of the udev helpers doing a SCSI-specific ioctl to a non-scsi device.  I think that /lib/udev/scsi_id was doing SG_IO to aoe devices?  Thanks for the report.  I will look into it.
Comment 2 Ed Cashin 2009-01-13 07:34:00 UTC
Could you please try the patch aoe-2.6.24-vendor.diff attached below?  I see now that you are using a Welland box as your AoE target.  Those devices send unsolicited AoE responses with an AoE command type that is in the range set aside for vendor extensions.  It is appropriate for the aoe driver to ignore such responses.
Comment 3 Ed Cashin 2009-01-13 07:38:57 UTC
Created attachment 19774 [details]
ignore vendor extention type command responses

Ignore AoE responses with a command type that is in the range set aside for vendor extensions.
Comment 4 karaluh 2009-01-14 13:05:08 UTC
Looks like it's fixed, I've had aoe complains once on every six seconds, now they are gone, thank you for fixing it. Now I have to figure out why the aoe device isn't mounted at startup.
Comment 5 Ed Cashin 2009-01-14 14:18:41 UTC
Thank you for the report.  I'll send the patch to the LKML (and put up a quilt tree for linux-next to pull from).

My reading of the bugzilla FAQ suggests it's time to change this bug's status to "VERIFIED" and that I can't do that.

Regarding mounting at startup, there is a FAQ entry in the EtherDrive HOWTO that you might find helpful:

http://support.coraid.com/support/linux/EtherDrive-2.6-HOWTO-5.html#ss5.14

Note You need to log in before you can comment on or make changes to this bug.