FreeBSD ccdconfig error

[Alex Cavnar, aka alc6379]

I was running a set of concatenated SCSI disks using ccd under FreeBSD 4.8 until recently the power went out during a thunder storm.

Now, when I attempt to run ccdconfig -C to configure my concatenated disk array, I get this error:

ccdconfig: ioctl (CCDIOCSET): /dev/ccd0c: Invalid argument


I've check all of the disklabels on my disks, and it appears they're all set up properly. I have 4 SCSI disks:

/dev/da0
/dev/da1
/dev/da2
/dev/da3

here's the contents of my /etc/ccd.conf file:


#device ileave flags devices
ccd0 1 none /dev/da0e /dev/da1e /dev/da2e /dev/da3e


I'm using my disks in dedicated mode, but I'm thinking of changing that to see if it does any good. I was expecting this array to screw up sometime, so I didn't have any data that I cared about. Has anyone run a ccd setup before? Maybe some of you have some insight?

[oubipaws]

"The symptom of this is:

# ccdconfig -C
ccdconfig: ioctl (CCDIOCSET): /dev/ccd0c: Inappropriate file type or format
This usually happens when you are trying to concatenate the c partitions, which default to type unused. The ccd driver requires the underlying partition type to be FS_BSDFFS. Edit the disklabel of the disks you are trying to concatenate and change the types of partitions to 4.2BSD."

Did you change the types of partitions to 4.2BSD? If so, I don't understand why it would not be working.

[Alex Cavnar, aka alc6379]

That's a good point, but I forgot to mention that I wasn't using the c partition of a slice. Just as many HOWTO's have recommended, I'm using my disks in dedicated mode-- da0e,da1e,da2e, and da3e. They are all of type 4.2BSD.

If I had gotten the inappropriate file format error, then I would definitely have known that I wasn't using 4.2BSD labelled partitions. Also, I have CCD compiled into my kernel, so I know that I have support for it. Odd thing is, it worked, right up until I had the power failure.

I'm pretty sure my hardware is good, and I know that my partitions are properly labelled, as everything configuration-wise seems the same as before the power outage. :confused:

[xulfralos]

I've never used ccd, so all I can offer is general device...

Have you checked the manpage for enabling debugging (if possible)?
Have you run 'fsck' on all your disks?
What does /dev/ccd0c correlate to, disk /dev/da2?

I believe there's other tools in ports for checking drives more thoroughly.

[Alex Cavnar, aka alc6379]

Originally posted by xulfralos
I've never used ccd, so all I can offer is general device...

Have you checked the manpage for enabling debugging (if possible)?
Have you run 'fsck' on all your disks?
What does /dev/ccd0c correlate to, disk /dev/da2?

I believe there's other tools in ports for checking drives more thoroughly.

1. Nope, that's a good idea, though. I'll check if it can be more verbose than "Invalid Argument"

2. It wouldn't do any good to run fsck, AFAIK. When you make a ccd array, the filesystem is spread accross all of the disks in the array. It's basically like a striped disk set.

3. /dev/ccd0c correlates to the striped set of concatenated disks: /dev/da0e, /dev/da1e, /dev/da2e, and /dev/da3e.

[xulfralos]

If that's the case, it sounds like there's some corruption going on somewhere.

[xulfralos]

Whatever happened with this?

[Alex Cavnar, aka alc6379]

Originally posted by xulfralos
Whatever happened with this?

I got too lazy to fool with it. :o

I think you're right, though-- there may have been some corruption somewhere. I mean, it was just working, and then one day it quit. It may have been drive failure, but who knows?

In the meantime, I've pulled the SCSI drives from the system. I'm going to set them up in another system to see if I can replicate the problem. But, like many other projects, that'll be another day...