stevenstapleton
Contributor
Contributor

Migration Failed

Error:    Cannot migrate serverxx from , LOCAL_15K_datastore to serverxx, shared datastore.

No Snapshots, No lock. Plenty of SPace on the shared storage.

attempted normal migrate of datastore only, with server on and with server shutdown.

Used advance options to move larger 262GB drive and vol Info.

reattempted the smaller 42GB drive keeps failing.

vCenter 6.0

Attempted with server shutdown and cd drive that sometimes cause this type of issue removed..

Third attempt was to go to browse datastore and copy vmdk file to another folder in same storage or to second storage both failed.

Server is up and running and DB is up and running, but really want to get the c:\ drive moved to the shared storage.

esxi host password unknown, or I would try winscp to copy the two remaining files (42GB vmdk and 2MB -ctk.vmdk) server is back running till I figure out a plan.

all other files are moved to shared storage.

Thanks

18 Replies
dekoshal
Hot Shot
Hot Shot

Try to make a clone of the VM and select the shared datastore as traget storage.

0 Kudos
stevenstapleton
Contributor
Contributor

Clones fail, around 60-70% complete, same place the migrate fails.

Backups fail, disk will not copy or move.

0 Kudos
jhague
VMware Employee
VMware Employee

Have you looked at the vmware.log file in the VMs home folder on the datastore? Any useful info/errors in there?

One option is you could potentially clone it as a workaround.

John Hague http://linkedin.com/in/john-hague | twitter @jhague10 VCIX-DCV | VCP-DCV 3/4/5/6 | VCP6-NV | VCP7-CMA | VCAP7-CMA Design
0 Kudos
dekoshal
Hot Shot
Hot Shot

Try V2V migration using VMware Converter.

0 Kudos
stevenstapleton
Contributor
Contributor

Nothing I can use.

MigrateESXVmkEventHandler: migration not in progress, not handling message 6.

0 Kudos
stevenstapleton
Contributor
Contributor

2017-06-20T16:05:44.960Z| vmx| I120: Migrate: Overriding message callbacks with migration handlers.

0 Kudos
jhague
VMware Employee
VMware Employee

Hmm nothing else? Are you able to post up the log or an extract? Anything useful in vmkernel.log?

What are you using for backup? Are there any orphaned snapshots? You can verify from the name of the vmdk.

Are you able to restart the host?

The only other thing I can think of is increasing the timeout but I suspect it may not help...

Using Storage vMotion to migrate a virtual machine with many disks timeout (1010045) | VMware KB

John Hague http://linkedin.com/in/john-hague | twitter @jhague10 VCIX-DCV | VCP-DCV 3/4/5/6 | VCP6-NV | VCP7-CMA | VCAP7-CMA Design
0 Kudos
stevenstapleton
Contributor
Contributor

It is possible that there were some orphaned snaps

there was nothing to consolidate.

Sometime in the distant past (in a galaxy far) one server crashed and was rebuilt

all the VMware files had been and were on local datastores  (bad)

then a long time passes, and I am hired.

Now time to fix this, so I setup a shared storage, fix another dead drive on the server

and successfully move 6 other servers to shared storage, but not this one.

did find a _1-000003.vmdk and a _1-00004.vmdk  from back in 2015 17.4MB but Provisioned size 262GB

after several attempts did remove them from the folder (now stored separately)

The server is now running without them in the folder

so since I couldn't move the entire thing

used advance options and started with the second drive.

Success

So then moved the server info - again success

then attempted drive c:\ . - Failed.

Yes Server is shutdown

So server log/ nvrma/ /vswp/ in shared storage - check

Drive 2 (with DB) in shared storage - check

Server running - check

Drive 1 - still in old location, without suspect files

Time on This Job - Three months.

0 Kudos
dekoshal
Hot Shot
Hot Shot

If the snapshot disks are not attach to the VM, it is not a matter of concern. I would suggest you to try VMware standalone converter to perform a V2V

migration of the VM where traget VM datastore will be your shared datastore.

Best Regards,

Deepak Koshal

CNE|CLA|CWMA|VCP4|VCP5|CCAH

0 Kudos
virtualDD
Enthusiast
Enthusiast

There might be a lock on the disk somehow. Next time you shutdown this vm I'd try the vmfsfilelockinfo: Finding the lock owners of a VMDK or file on a VMFS datastore in VMware ESXi 5.5 P05 and later (2110...

That way you'll find out if there is something else accessing the disk and making it impossible to move.

0 Kudos
dekoshal
Hot Shot
Hot Shot

screenlock.jpg

In the Image we can see that the highlighted number is MAC of vmnic0 of esxi host and its holding the lock on w10-base-01a-flat.vmdk.

vmkfstools -D  command can also be used to identify MAC address of esxi host holding the lock on the VM file.

Out put of the file is suppose to show the MAC address of the esxi host where the VM is currently running if it show any other host while VM is running or

If the output of the command show MAC address when the VM is powered off  its a issue and you might need to flip the NIC or reboot the esxi host to

release the lock on the VM file/files.

Best Regards,

Deepak Koshal

CNE|CLA|CWMA|VCP4|VCP5|CCAH

0 Kudos
stevenstapleton
Contributor
Contributor

My server Nic MAC information was already moved to new storage, and is running fine. The NIC is not locked to the host.

Think I mentioned that the Server Vol Info /vswp/ nram/ and disk 2 moved successfully

There are only two remaining files, but that is just because the server had to be started and is running.

Using the disk that remains in the old folder

attempted to install VMware convertor, and getting error on that, but that is another topic.

0 Kudos
stevenstapleton
Contributor
Contributor

Yes. if only I had the root and password of the esxi host from the admin who left a long time before I was hired.

So far we have not located the password or guessed it.,

so I can not SSH or SFTP into the host

0 Kudos
stevenstapleton
Contributor
Contributor

Placing the host in maintenance mode will not help either since it cant move the vdmk off the local storage it will just become unavailable.

0 Kudos
stevenstapleton
Contributor
Contributor

regedit fix also posted in Forums worked to fix convertor error so now I have it installed, waiting time to shutdown "Test" Production server.

yeah, test doesn't always mean it is not production. And since it affects payroll.... will wait.

0 Kudos
stevenstapleton
Contributor
Contributor

If this doesn't work V2V then yes move all virtual to other host and reboot this esxi host.

With the remaining virtual machine in question shutdown of course.

see if that doesn't shake it loose.

dekoshal
Hot Shot
Hot Shot

Good Luck.

0 Kudos
stevenstapleton
Contributor
Contributor

Password to host found, exported files sent to support.

Thanks everyone for Ideas posted, and when I find the solution I will post it.

We now return to "How the Data Center get Glitches" where we hear Maria say, it was that way when I found it.

0 Kudos