VMware Cloud Community
Joaquim
Contributor
Contributor
Jump to solution

Adding storage to esx. Problems with VCB proxy

Hi,

I just added 4 FC LUNs and have initialized them on ESX 3.02 without any problem.

After, I went to the VCB proxy, which had the automount disable, and I think I made a mistake because I put them online (did not initialized) and after that Vmware can't access these vmfs drives.

Trying to fix this I did the automount disable and scrub, but I am not sure what to do next. Should I restart VCB proxy?

According to some posts here, I ran "fdisk -l all /dev/sd?" command and it shows that those 4 drives didn't have the fb partition type.

Based in another post, I ran fdisk against each drive but I didn't get the option to put the partition type.

Can someone help me on this?

Thanks

Reply
0 Kudos
1 Solution

Accepted Solutions
Lightbulb
Virtuoso
Virtuoso
Jump to solution

My DSL is flaking out. But looking over your steps yes. I may be knocked off any second good luck. I can not provide reliable help as I keep going offline.

View solution in original post

Reply
0 Kudos
14 Replies
Lightbulb
Virtuoso
Virtuoso
Jump to solution

Easy answer is that since you have just provisioned your LUNS to ESX, I would assume nothing is on them yet

If this is the case you might be better off starting from scratch. Delete the datastores, destroy and recreate the LUNS on storage and then present new LUNS to ESX again and format as VMFS.

Then carefully setup you VCB again.

This is what I would do. If you are not in production and the boss is not leaning on you; get it right from the get go. You may be able to sort the issues that you currently have but then a month down the line something comes to bite you in the backside when there are a dozen production VMs running on the system.

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

Yeah, I would say you are right.

Can you please tell me what did I do wrong? Was it to put the disks online in the VCB proxy? What should I do then?

Thanks

Reply
0 Kudos
Lightbulb
Virtuoso
Virtuoso
Jump to solution

I would assume that it was during the presentation of the LUNS to the VCB proxy. I would guess that disabling automount did not take or whatever and the 2003 server initialized the drives.

You were fine up to the point that you setup VCB. I would recommend once you have everything set on the ESX side that you carefully proceed with VCB setup. If it was a fat finger type of mistake the issue should not happen again.

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

Ok!

I will try it with my little finger, just in case Smiley Wink)

thanks

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

I am not able to remove any of the 4 datastores from virtual center.

Is there any other way to do it?

Should I shutdown the proxy server?

Reply
0 Kudos
Lightbulb
Virtuoso
Virtuoso
Jump to solution

Yes it is probably locking the LUNS. I am currently immersed in Child care duties so let me know how it goes.

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

I shutdow the proxy server and it didn't work. I cannot remove the datastore.

Now, I am a little bit worried...

Reply
0 Kudos
Lightbulb
Virtuoso
Virtuoso
Jump to solution

Shutdown proxy. Disconnect fiber. Destroy LUNS on san. Reboot ESX hosts. These are not live systems so no worries yet Smiley Happy

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

No, no. These are production systems.

The storage I added of course was not yet used, but I have several vm machines in these esx cluster on other datastore.

Reply
0 Kudos
Lightbulb
Virtuoso
Virtuoso
Jump to solution

Ok well do All I mentioned but reboot the ESX hosts. Instead onec the unused LUNS are gone rescan storage.

Reply
0 Kudos
Joaquim
Contributor
Contributor
Jump to solution

let's see if understood it right:

1- Shutdown the proxy server

2- Destroy the LUN in the SAN (by the way it is a IBM TS 4700)

3- Rescan the storage in the esx server

4- Define the LUN in the SAN

5- Rescan the storage in the esx again

6- Define the vmfs/datastores

7- Start up proxy server

8- Do automount disable and scrub on the proxy server

9- Put the drives online on the proxy server

Is that it?

Questions:

A- What happens with the 4 datastores now defined and unusabled? Shouldn't I do something to them or they will disapear as soon as I do the rescan (3)?

B- The procedures for the proxy server (7,8,9) are they ok?

Reply
0 Kudos
Lightbulb
Virtuoso
Virtuoso
Jump to solution

My DSL is flaking out. But looking over your steps yes. I may be knocked off any second good luck. I can not provide reliable help as I keep going offline.

Reply
0 Kudos
Texiwill
Leadership
Leadership
Jump to solution

Hello,

If all you did was mount them to VCB and HAVE yet to shutdown/reboot the ESX host that had those LUNs, and you have fixed the VCB proxy issue.

If fdisk -l returns no filesystems of type fb, then you can restore your partitions using the instructions within http://communities.vmware.com/servlet/JiveServlet/download/2130-1-3516/How%2Bto%2Brecover%2Byour%2BV...

The above will restore your partitions only if the VMFS was left untouched after the VCB issue. Note if this is for an RDM, then there could be other partitions to restore.


Best regards,
Edward L. Haletky
VMware Communities User Moderator
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
Blue Gears and SearchVMware Pro Blogs -- Top Virtualization Security Links -- Virtualization Security Round Table Podcast

--
Edward L. Haletky
vExpert XIV: 2009-2023,
VMTN Community Moderator
vSphere Upgrade Saga: https://www.astroarch.com/blogs
GitHub Repo: https://github.com/Texiwill
Joaquim
Contributor
Contributor
Jump to solution

Boy! What a stressed momemt!

Everything is now ok! I have my datastores with the fb type and the proxy server has this very nice message saying that the disks are "Healthy(Unknow Partition)".

I created a vm on one of the new datastores and made a VCB full backup and everything ran nice and smooth! Uff!

Just for the history and knowledge of newcommers:

There was no need to do all the steps above, just shutdown the proxy server, rescan the disks on esx (the previous datastores defined should disapear), add the new vmfs-datastores, go to proxy server and do the diskpart automount disable (beaware with your fat fingers...) and that is it!

Thanks Tom for your care and support.

Thanks Edward for the doc. It will be useful for other situations (knocking off....)

Reply
0 Kudos