CAM status: Command timeout

I

irukandji

I am having an interesting issue, two weeks back, on my ZFS drive I started to get the errors like this after I turn on the system and is turned on for a while:

Code: 
ahcich1: Timeout on slot 31 port 0
ahcich1: is 00000000 cs 00000000 ss 80000001 rs 80000001 tfd 40 serr 00000000 cmd 0000c017
(ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 20 e5 4e c1 40 88 00 00 00 00 00
(ada1:ahcich1:0:0:0): CAM status: Command timeout
(ada1:ahcich1:0:0:0): Retrying command, 3 more tries remain

I have first changed the SATA cable, the problem remained. Turned disk off and waited for the spare.

Now I have plugged in the spare drive and I am trying to copy the data (at least as much as I can) to the new hdd. The problem is that it hangs in kernel for too long and the copying will take forever.

Is there any way to lower the timeout to a second or two, so I can copy data from broken disk in some reasonable time?
 
In some cases I had the similar errors on 100% good HDDs.
That HDDs had perfect SMART and passed complete surface test.
In some cases this issue related to SATA cables,
but I had the similar cases where disks only from specified vendor (WD) did not work correctly in the specified computer/server with FreeBSD.
Also I had a situation when the same errors happened after upgrading FreeBSD, but the same hardware (disks, mainboard, sata cables) worked well before upgrade.

So you need to know exactly about a condition of your disk. It allows you to choose a correct recovery strategy.

First of all, try to check SMART on the affected disk.
You can do it on FreeBSD with sysutils/smartmontools.
Also you can turn off your FreeBSD, temporary detach the disk and to check the SMART on the another PC.

Just in case:

229745 – ahcich: CAM status: Command timeout

bugs.freebsd.org bugs.freebsd.org
 
The issue is that contacting SMART is failing all the time.

I have tested the cable and port with another disk and they are both ok, but I am buying disks in pairs just for this reason. Will mount controller pcb from second disk to failing one and backup the data.
 
You must log in or register to reply here.
Back
Top