TruCluster
cancel
Showing results for 
Search instead for 
Did you mean: 

"cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

Reda El-Masry
Occasional Contributor

"cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

We are facing a serious problem where a cluster failed without any reason. When we tried to reboot the cluster it fails with

WARNING: cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain
vm_swap_init: warning /dev/disk/dsk8b swap device not found
vm_swap_init: swap is re-set to lazy (over commitment) mode


Please your URGENT request is highly apreciated.

Thanks.
5 REPLIES
Hein van den Heuvel
Honored Contributor

Re: "cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"


For URGENT requests be sure to contact your support center. These forum are 'best effort' no response time or response quality commited.

Anyway... you have:

(err 19) for domain root1_domain
:
grep 19 /usr/include/errno.h
:
#define ENODEV 19 /* No such device */

and

/dev/disk/dsk8b swap device not found


I would appear that the root disk: dsk8b is not available.

>> We are facing a serious problem where a cluster failed without any reason.

IMHO, the boot disk not being avaible is a good reason for cluster failure.

However... this system appears to be booting! Therefor there is a boot disk, but it's name does not seem to correspond with the expected dsk8
Poke around with commands like 'hwmgr -show scsi'

What changed? Storage?
Did it reboot? Volontueerly?
Why was it rebooted?
When at the console prompt... does the SHOW DEV output look reasonable?

Is the other member up?
Is the right member being associateed with the system?

Good luck!

Hein.
Reda El-Masry
Occasional Contributor

Re: "cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

Heuvel,
Thanks for your reply.

I have already triggered the support but it is almost 24 hours without any progress so I decided to try you people.

We have modified nothing. The service was working properly until it unexpectedly failed yesterday. At that time, the problem was only with the first member but when we failed to relocate the service to the second member, we tried to reboot the whole cluster!!! Since then, we are unable to reboot any member. They ALL report the same message!!!

We are still able to boot the first member as it has the original disk that was used for the cluster installation.

Just to give you a quick view of the cluster. It consists of two ES45 machines + quorum. Each machine is hosting internally its member boot disk. In addition, member one is hosting internally the disk that was originally used to install Tru64

Question:
I had a similar problem before and it was due to sysconfigtab: cluster_seqdisk_major & cluster_seqdisk_minor?

Your help is really appreciated.

Thanks.

Reda El-Masry
Occasional Contributor

Re: "cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

Also, all HW are working properly. I have already checked show dev and everything is OK
Han Pilmeyer
Esteemed Contributor

Re: "cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

It does look like the device is missing. If you look at the devices when you boot the "installation disk", do the device still exist and can you access them? Make sure that the WWID of the device hasn't changed from what it was when you installed the system. The "installation disk" probably still has the root1_domain. Can you mount the boot disk for member 1 (root1_domain) from the installation disk?
Reda El-Masry
Occasional Contributor

Re: "cfs_devlist_update: could not enumerate advfs devts (err 19) for domain root1_domain"

Thanks.

I have found the problem. Someone at the customer site has deleted '/'

I had to restore everything and it is OK now.

Thanks everyone.