VMware Cloud Community
mbender71
Contributor
Contributor

VCBMounter Problem : FULLVM vs FILE

Hi again everyone,

I decided to start another thread/question since I believe the problem has

taken a different course. Originally I had a LUN ID mapping error, but since

fixing that I am able to use vcbmounter to perform a full VM copy via the

'-t FULLVM' option. The problem is that I am NOT able to do the file level

copy via '-t FILE'. Here is the error message I get when attempting the

FILE-LEVEL mounting :

\[2007-04-19 14:32:04.531 'vcbMounter' 3584 info] Authenticating user WebAdmin

\[2007-04-19 14:32:04.562 'vcbMounter' 3584 info] Logged in!

\[2007-04-19 14:32:04.578 'vcbMounter' 3584 info] Performing SearchIndex find.

\[2007-04-19 14:32:04.625 'vcbMounter' 3584 info] Attempting index search.

\[2007-04-19 14:32:04.640 'vcbMounter' 3584 info] Attempting cache lookup.

\[2007-04-19 14:32:04.656 'vcbMounter' 3584 info] Attempting traversal search.

\[2007-04-19 14:32:04.812 'vcbMounter' 3584 info] Found VM using traversal search.

\[2007-04-19 14:32:04.828 'vcbMounter' 3584 info] Got VM MoRef

\[2007-04-19 14:32:05.109 'vcbMounter' 3584 info] Got access method

\[2007-04-19 14:32:05.125 'vcbMounter' 3584 info] Got coordinator object

\[2007-04-19 14:32:05.140 'vcbMounter' 3584 info] Attempting data access.

\[2007-04-19 14:32:05.171 'vcbMounter' 3584 info] No snapshot info for this VM, nothing to do.

\[2007-04-19 14:32:05.187 'vcbMounter' 3584 info] Creating snapshot

\[2007-04-19 14:32:08.140 'vcbMounter' 3584 info] Snapshot created, ID: snapshot-235

\[2007-04-19 14:32:08.156 'vcbMounter' 3584 info] Mount operation created snapshot.

\[2007-04-19 14:32:08.234 'vcbMounter' 3584 verbose] Found a device: vim.vm.device.VirtualLsiLogicController

\[2007-04-19 14:32:08.250 'vcbMounter' 3584 verbose] Found a device: vim.vm.device.VirtualDisk

\[2007-04-19 14:32:08.312 'vcbMounter' 3584 info] Mount operation obtained backup info.

\[2007-04-19 14:32:09.812 'vcbMounter' 3584 error] Error: Error while opening disk blklst://snapshot-235\[GuestOS Storage]

TestTSD/TestTSD-OS.vmdk@orlvmcon.navair.navy.mil?xxxx/xxxx: A virtual disk could not be opened. -- Cannot open disk fil

e: Error : Disk library failed (One LUN could not be opened). Unable to open 'blklst://snapshot-235\[GuestOS Storage] Tes

tTSD/TestTSD-OS.vmdk@orlvmcon.navair.navy.mil'. Failed to configure disk scsi0:0.

\[2007-04-19 14:32:09.890 'vcbMounter' 3584 error] An error occured, cleaning up...

\[2007-04-19 14:32:13.343 'App' 3584 verbose] Got vmdb error -14 (Pipe connection has been broken) when invoking \[logout]

on \[vim.SessionManager:SessionManager]

\[2007-04-19 14:32:13.375 'vcbMounter' 3584 info] Error on logout (ignored): vmodl.fault.HostNotReachable

\[2007-04-19 14:32:13.390 'vcbMounter' 3584 verbose] Shutting down VMDB service...

\[2007-04-19 14:32:13.406 'vcbMounter' 3584 verbose] Unregistering callback...

\[2007-04-19 14:32:13.421 'vcbMounter' 3584 verbose] ...done

It is interesting to note that in the first error line there is an absence of the

username and password after the filename (?xxxxxx/xxxxxxx):

\[2007-04-19 14:32:09.812 'vcbMounter' 3584 error] Error: Error while opening disk blklst://snapshot-235\[GuestOS Storage]

TestTSD/TestTSD-OS.vmdk@orlvmcon.navair.navy.mil?xxxx/xxxx: A virtual disk could not be opened. -- Cannot open disk fil

e: Error : Disk library failed (One LUN could not be opened). Unable to open 'blklst://snapshot-235\[GuestOS Storage] Tes

tTSD/TestTSD-OS.vmdk@orlvmcon.navair.navy.mil'. Failed to configure disk scsi0:0.

Here is a short excerpt of the logging I get with a successful run of the

mounter using the FULLVM option :

Copying "\[GuestOS Storage] TestTSD//vmware.log":

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

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

\[2007-04-19 14:20:31.734 'vcbMounter' 624 info] Exporting disks

\[2007-04-19 14:20:31.734 'vcbMounter' 624 info] Filenames: blklst://snapshot-234\[GuestOS Storage] TestTSD/TestTSD-OS.vmd

k@orlvmcon.navair.navy.mil?username/password --> d:\san\TestTSD\scsi0-0-0-TestTSD-OS.vmdk

\[2007-04-19 14:20:31.796 'App' 624 trivia] Evaluating 1 paths.

\[2007-04-19 14:20:31.796 'App' 624 trivia] Trying to open path
?\scsi#disk&ven_dell&prod_perc_5#i&rev_1.00#6&e45e330&0

&010000#\{53f56307-b6bf-11d0-94f2-00a0c91efb8b}.

\[2007-04-19 14:20:31.796 'App' 624 info] Now using Path
?\scsi#disk&ven_dell&prod_perc_5#i&rev_1.00#6&e45e330&0&010000

#\{53f56307-b6bf-11d0-94f2-00a0c91efb8b}.

\[2007-04-19 14:20:31.796 'App' 624 trivia] Reading 32256 bytes from offset 0.

\[2007-04-19 14:20:31.796 'App' 624 trivia] Reading 32256 bytes from offset 1388699648.

\[2007-04-19 14:20:31.812 'App' 624 trivia] No partitions could be found on this device.

\[2007-04-19 14:20:31.812 'App' 624 trivia] Evaluating 1 paths.

\[2007-04-19 14:20:31.812 'App' 624 trivia] Trying to open path
?\scsi#disk&ven_dgc&prod_raid_5&rev_0220#6&2b16ca3a&0&0

00000#\{53f56307-b6bf-11d0-94f2-00a0c91efb8b}.

\[2007-04-19 14:20:31.828 'App' 624 info] Now using Path
?\scsi#disk&ven_dgc&prod_raid_5&rev_0220#6&2b16ca3a&0&000000#{

53f56307-b6bf-11d0-94f2-00a0c91efb8b}.

\[2007-04-19 14:20:31.828 'App' 624 trivia] Reading 32256 bytes from offset 0.

\[2007-04-19 14:20:31.828 'App' 624 trivia] Found 1 partition(s) on this device.

\[2007-04-19 14:20:31.828 'App' 624 info] Partition #1 is a VMFS partition.

\[2007-04-19 14:20:31.828 'App' 624 trivia] Reading 1574400 bytes from offset 1114112.

\[2007-04-19 14:20:31.890 'App' 624 trivia] Found an ESX LVM v3 signature on partition 1.

\[2007-04-19 14:20:31.890 'App' 624 info] Found logical volume 4548d582-2289feeb-5718-00137264, PE index 0.

\[2007-04-19 14:20:31.890 'App' 624 trivia] Evaluating 1 paths.

The username/password (I have editted it to be generic) shows up prorperly

in the logging/request instead of the xxxxx/xxxxx as displayed in the failed

FILE-level mounting. Could this be a clue? Does FULLVM use a different

method of connecting to the LUNS (via FC) than FILE? I realize one is a copy

and the other is a true mounter but I would assume the code to access the

SAN is the same in both. Many thanks in advance for the help!

Sincerely,

Mike B

Orlando, FL

PS> previous thread topic was "VCB & ESX LUN Configuration Problem"

0 Kudos
3 Replies
Linus76
Contributor
Contributor

If I run a file level from my Backup software's save commands, using all the built in scripts of the VCB Infrastructure, but I am unable to manually execute the same thing from a vcbmounter perspective.

Can anyone shed light on what command the Integreation scripts runs, specifically, for vcbmounter?

0 Kudos
The_Ether
Enthusiast
Enthusiast

I have the same problem, but only on a VMFS volume that has an original partition plus an extent.

Works fine on any other datastore and is not specific to the VM. VM's moved on fail, VM's moved off work.

0 Kudos
The_Ether
Enthusiast
Enthusiast

Both LUNs need to be presented on the same Controller/Target ID:

\- Move both LUNs to the same Controller/Target ID

\- In Computer Management right click on Storage | Disk Management and select Rescan Disks

\- Restart the VMware Virtual Mount Manager Extended service on the VCB Proxy

http://theether.net/kb/100032

0 Kudos