VMware Cloud Community
juanjico
Contributor
Contributor

ESXi 5.5 and FreeBSD 10: mtp0 timeout

HI!

I'm using FreeBSD 10 32 bits with ESXi 5.5, and the virtual machine reboots itself every day with this error in dmesg:

mpt0: request 0xc6c66c60:4363 timed out for ccb 0xc7b39800 (req->ccb 0xc7b39800)

mpt0: attempting to abort req 0xc6c66c60:4363 function 0

mpt0: abort of req 0xc6c66c60:4363 completed

mpt0: attempting to abort req 0xc6c66c60:4363 function 0

mpt0: completing timedout/aborted req 0xc6c66c60:4363

(da0:mpt0:0:0:0): WRITE(6). CDB: 0a 00 ff db 01 00

(da0:mpt0:0:0:0): CAM status: Command timeout

(da0:mpt0:0:0:0): Retrying command

mpt0: abort of req 0xc6c66c60:0 completed

mpt0: request 0xc6c6b720:4573 timed out for ccb 0xc7b39800 (req->ccb 0xc7b39800)

mpt0: attempting to abort req 0xc6c6b720:4573 function 0

mpt0: completing timedout/aborted req 0xc6c6b720:4573

(da0:mpt0:0:0:0): WRITE(6). CDB: 0a 00 ff db 01 00

(da0:mpt0:0:0:0): CAM status: Command timeout

(da0:mpt0:0:0:0): Retrying command

mpt0: abort of req 0xc6c6b720:0 completed

mpt0: request 0xc6c6e690:4706 timed out for ccb 0xc7b39800 (req->ccb 0xc7b39800)

mpt0: attempting to abort req 0xc6c6e690:4706 function 0

mpt0: completing timedout/aborted req 0xc6c6e690:4706

(da0:mpt0:0:0:0): WRITE(6). CDB: 0a 00 ff db 01 00

(da0:mpt0:0:0:0): CAM status: Command timeout

(da0:mpt0:0:0:0): Retrying command

mpt0: abort of req 0xc6c6e690:0 completed

mpt0: request 0xc6c65e5c:4837 timed out for ccb 0xc7b39800 (req->ccb 0xc7b39800)

mpt0: attempting to abort req 0xc6c65e5c:4837 function 0

mpt0: completing timedout/aborted req 0xc6c65e5c:4837

(da0:mpt0:0:0:0): WRITE(6). CDB: 0a 00 ff db 01 00

(da0:mpt0:0:0:0): CAM status: Command timeout

(da0:mpt0:0:0:0): Retrying command

mpt0: abort of req 0xc6c65e5c:0 completed

mpt0: request 0xc6c68474:4943 timed out for ccb 0xc7b39800 (req->ccb 0xc7b39800)

mpt0: attempting to abort req 0xc6c68474:4943 function 0

mpt0: completing timedout/aborted req 0xc6c68474:4943

(da0:mpt0:0:0:0): WRITE(6). CDB: 0a 00 ff db 01 00

(da0:mpt0:0:0:0): CAM status: Command timeout

(da0:mpt0:0:0:0): Error 5, Retries exhausted

mpt0: abort of req 0xc6c68474:0 completed

g_vfs_done():da0p2[WRITE(offset=33452544, length=512)]error = 5

panic: cannot reassign paging buffer

cpuid = 1

KDB: stack backtrace:

#0 0xc0af6f12 at kdb_backtrace+0x52

#1 0xc0abce21 at panic+0x121

#2 0xc0b5d179 at reassignbuf+0x239

#3 0xc0b44406 at bdirty+0x36

#4 0xc0b43252 at brelse+0x112

#5 0xc0b46990 at bufdone+0x60

#6 0xc0a22fba at g_vfs_done+0x27a

#7 0xc0b464a3 at biodone+0xf3

#8 0xc0a1f844 at g_io_schedule_up+0x1d4

#9 0xc0a1fddd at g_up_procbody+0x6d

#10 0xc0a8bff3 at fork_exit+0xa3

#11 0xc0f80704 at fork_trampoline+0x8

Uptime: 22h11m18s

The machine is working perfect, until this error appears.

I've been searching with no luck.

Any idea ?

Tags (3)
Reply
0 Kudos
3 Replies
hadricus
Enthusiast
Enthusiast

Hi juanjico,

Is the problem occurring at the same time every day? Are there any tasks or large workloads hitting your backend storage at the same time?

Does the VM have any snapshots?

You can try changing your disk timeout.

kern.cam.da.default_timeout = 300

Reply
0 Kudos
juanjico
Contributor
Contributor

Thanks hadricus for your reply.

It's not occurring at the same time every day, but only happen every day except weekends. So, yes, it's occurring on heavy workloads. This machine is a "simple" webserver using PHP and MySQL.

I have more VM in the same host and all is working perfect, but this VM is the only using FBSD 10.

Previously, the host was using ESXi 4.1, so I've upgraded to 5.5 to see if this could be the problem. But it's occuring even in 5.5.

I tried to change the disk controller type, with no luck. It's still failing.

The VM does not have any snapshots.

I changed the timeout to see if this helps.

Thanks.

Reply
0 Kudos
juanjico
Contributor
Contributor

Has crashed again. Timeout is not the problem.

Any idea ?

Reply
0 Kudos