VCBMount Error: Failed to export the disk: The requested resource is in use

[MiteeThoR]

I'm having a problem with VCB. I'm using ESX Ranger, which launches several VCB sessions at once, and I get random "Error: Failed to export the disk: The requested resource is in use"

This doesn't happen on all of the backups, just some of them, sometimes, and sometimes when it's the only thing running. The problem is, I don't know WHICH resource is busy - the SAN, the server, the destination, what? no idea. I've attached the logs of a backup, one that failed, and then ran perfectly the 2nd time below:

**********************************

1st run, which failed with resource error

**********************************

C:\Program Files\VMware\VMware Consolidated Backup Framework>vcbMounter -h orlvc

01 -u backupadmin -p veritassux -m san -a ipaddr:orlvmris01 -r e:\ESXFRIDAY\orlv

mris01 -M 1 -F 1

\[2007-04-18 15:20:36.837 'App' 3368 info] Current working directory: C:\Program

Files\VMware\VMware Consolidated Backup Framework

\[2007-04-18 15:20:37.837 'BaseLibs' 5996 warning] \[Vmdb_Unset] Unsetting unknown

path: /vmomi/

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01/ORLVMRIS01.vmx":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01/ORLVMRIS01.nvram":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-2.log":

0%=====================50%=====================100%

*************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-0.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-1.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-3.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-4.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware.log":

0%=====================50%=====================100%

**************************************************

\[2007-04-18 15:21:01.835 'BaseLibs' 1928 warning] \[Vmdb_Unset] Unsetting unknown

path: /vmomi/

Converting "e:\ESXFRIDAY\orlvmris01\scsi0-0-0-ORLVMRIS01.vmdk" (flat file):

0%=====================50%=====================100%

**************************************************

\[2007-04-18 15:25:29.301 'BlockList' 3368 warning] Could not log out (orlvc01:ba

ckupadmin)

\[2007-04-18 15:25:29.301 'vcbMounter' 3368 error] Error: Failed to export the di

sk: The requested resource is in use

\[2007-04-18 15:25:29.301 'vcbMounter' 3368 error] An error occured, cleaning up.

..

******************************************

Then I ran it again and it worked fine....

******************************************

C:\Program Files\VMware\VMware Consolidated Backup Framework>vcbMounter -h orlvc

01 -u backupadmin -p veritassux -m san -a ipaddr:orlvmris01 -r e:\ESXFRIDAY\orlv

mris01 -M 1 -F 1

\[2007-04-18 15:32:41.900 'App' 5892 info] Current working directory: C:\Program

Files\VMware\VMware Consolidated Backup Framework

\[2007-04-18 15:32:42.947 'BaseLibs' 5524 warning] \[Vmdb_Unset] Unsetting unknown

path: /vmomi/

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01/ORLVMRIS01.vmx":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01/ORLVMRIS01.nvram":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-2.log":

0%=====================50%=====================100%

*************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-0.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-1.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-3.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware-4.log":

0%=====================50%=====================100%

**************************************************

Copying "\[ORLESX01-iSCSI03] ORLVMRIS01//vmware.log":

0%=====================50%=====================100%

**************************************************

\[2007-04-18 15:32:58.010 'BaseLibs' 2572 warning] \[Vmdb_Unset] Unsetting unknown

path: /vmomi/

Converting "e:\ESXFRIDAY\orlvmris01\scsi0-0-0-ORLVMRIS01.vmdk" (flat file):

0%=====================50%=====================100%

**************************************************

\[2007-04-18 15:37:57.711 'BlockList' 5892 warning] Could not log out (orlvc01:ba

ckupadmin)

\[2007-04-18 15:37:58.117 'BaseLibs' 4404 warning] \[Vmdb_Unset] Unsetting unknown

path: /vmomi/

Converting "e:\ESXFRIDAY\orlvmris01\scsi0-1-0-ORLVMRIS01_1.vmdk" (flat file):

0%=====================50%=====================100%

**************************************************

\[2007-04-18 15:50:57.398 'BlockList' 5892 warning] Could not log out (orlvc01:ba

ckupadmin)

C:\Program Files\VMware\VMware Consolidated Backup Framework>

[Jwoods]

I was having the problem too. It was totally inconsistent. One day all jobs would work then the next, 3,4,8 jobs would fail. It seemed to be a problem with VCB executing multiple snapshots simultaneously. That's my theory at least from my testing. However, the newest version of VCB v1.0.2 resolved those issues! I'm currently running 6 jobs simultaneously without a hiccup.

Download -

http://www.vmware.com/download/vi/

Release Notes -

http://www.vmware.com/support/vi3/doc/releasenotes_vcb102.html

[Schorschi]

Actually 1.0.2 still has some issues, VMware acknowledge with us yesterday that under the specific conditions we created, by chance in our lab, VCB 1.0.2 starts reporting TOO MANY SCSI RESERVATIONS... but the real error is as noted above... Disk is busy or in use... VMware did not give us any ETA for a fix, but acknowledged that after review the 1.0.2 code, there is an issue. The question is how to fix it, and when said fix will be available.

[GBromage]

If it's a help, I've got a similar issue (take a look here: http://www.vmware.com/community/thread.jspa?threadID=73587&tstart=0 )

VMWare support are looking into it at the moment.

[MiteeThoR]

The funny things is, I upgraded to 1.02, and my error changed from "Resource is Busy" to "Too Many SCSI Reservation Conflicts"

I assumed this was related to my using iSCSI but if it's happening on HBA's as well perhaps there is something else going on?

[Topski]

Hi Schorschi,

I think we ran the same tests. I have exactly the same issues. Running parallel vcbmounts at first renders random problems. Then I put some wait states in my scripts (wait 45 secs, that start the next one) but this results indeed in the Too many SCSI reservations error. Do you have some workaround for this ? This is bad news for my management (and for me).

What type of hardware are you using ? We use all HP systems (EVA8000, HP DL580's and Qlogic HBAs).

Best regards,

Ronald Top

Netherlands

[Peaceman]

The same issue:

\[2007-03-24 11:29:49.406 'vcbMounter' 1696 error] Error: Failed to export the disk: The requested resource is in use

\[2007-02-24 15:29:49.406 'vcbMounter' 1696 error] An error occured, cleaning up...

\[2007-03-24 11:44:58.265 'vcbMounter' 1696 warning] Snapshot deletion failed. Attempting to clean up snapshot database...

failed to prepare vm1 for backup, PrepareForBackup() returned error 1

VM fall after this \!!!

Moreover I did not find any snapshots in Snapshot Manager after all.

It is terrible. This problem happens time to time.

What is going on? Any ideas?

I have only one backup task at a time.

[Topski]

Hello !

maybe of interest to somebody, I have done the following :

I do now a full export run from an ESX host instead of the Windows based proxy in the following way :

create a new dedicated ESX VI3 host with two HBAs

create a LUN with ext3 file system

make the ext3 LUN available in the cos of the dedicated system using fdisk, mkfs and mount the ext3 LUN in /mnt/folder

set the esxcfg-mpath policy in such a way that the ext3 LUN has a fixed policy over HBA1

set the esxcfg-mpath policy in such a way that the vmfs3 LUNs has a fixed policy over HBA2

create a vcbmount script that does a fullvm to the /mnt/folder/

use this script as a prebackup script

use a Linux based backup client (for us Legato)

backup the /mnt/folder directory

This works quite well, I have now in parallel 6 sessions and this makes in possible to run at 100GB/hour at this moment. I will now try to increase the amount of sessions to see if the speed can be optimized.

If you need some info, let me know.

BR,

Ronald