HA Userworld Swap Disk

Joined
Nov 22, 2011
Messages
4
Hello to all

My question is regarding the swap disk used by Esxi to hold the HA config for each host.

We have 4 ESXi4.1 hosts (blade enclosure C7000 - Fibre to SAN) and a HA User-world Swap Disk on SAN which is 5gb in size. There is only 438mb left on this drive. I understand that each host will require its scratch config to point to a vmfs folder on this drive and this takes up 1gb of space for each host, so my question is can i create a second V disk on the SAN, present it to the Esxi hosts, and use the Storage > Properties > Increase on the current User-world drive to span the new V disk to the existing User-world V disk?

Im concerned about data loss if doing this. Basically I would like some advice or best practise on increasing the size of the HA User-world Swap Disk and if this can be done live as we have over 40 VM's and I am not able to shutdown all the VM's and reboot the hosts in one go.

Any advice appreciated

Many thanks

Regards

John
 
ESXi 4.x does not use the scratch location to store the config for HA, it is used for temporary logs, dump data, and system swap. Not only that, but a persistent scratch isn't required(though its not a bad idea if you have the space to spare). Unless I'm misunderstanding you, I just want to make that clear before answering your question.

To answer your question; Creating a new lun on your SAN, presenting it to the ESXi hosts, and then extending the datastore by spanning it across the new lun should cause no downtime(on the ESXi side anyways). Just make sure you present the lun to every ESXi host and rescan each host after spanning the datastore. Also note that this is a one way change, you will not be able to undo this change(not easily or without downtime anyways).

Now, is it the best way or best practice? Personally I would try and see if your SAN supports growing the lun first, and then rescan and grow the datastore that way. I can't think of other ways that wouldn't incur at least some downtime(one host at a time), but maybe others will be able to chime in.

Read these links;
http://kb.vmware.com/kb/2000643 - How to use extents in general
http://kb.vmware.com/kb/1033696 - A little about how to create and use a persistent scratch
 
Last edited:
Thanks for your reply Chilly

From my reading I could see that the hosts may not need a persistent location for the scratch config for this version of ESXi but there was nothing definitive i could find. I would like to have this anyways as we have storage available.

To be clearer-

Esxi4.1 embedded (no local disks)
HP EVA 6400 FC SAN
1 x 5gb LUN presented to all 4 ESXi hosts (438mb remaining)

I could not find a way to extend the LUN within Command View for the SAN

I have read through the links you posted and with a bit of navigation managed to come across this procedure for anyone else who may need it:

Procedure
1 Log in to the vSphere Client and select a host from the Inventory panel.
2 Click the Configuration tab and click Storage.
3 From the Data stores view, select the data store to increase and click Properties.
4 Click Increase.
5 Select a device from the list of storage devices and click Next.
If you want to add a new extent, select the device for which the Expandable column reads No. If you want to expand an existing extent, select the device for which the Expandable column reads Yes.
6 Select a configuration option from the bottom panel.
Depending on the current layout of the disk and on your previous selections, the options you see might vary.

Option Description
Use free space to add new extent - Adds the free space on this disk as a new data store extent.
Use free space to expand existing Extent - Grows an existing extent to a required capacity.
Use free space - Deploys an extent in the remaining free space of the disk. This option is available only when adding an extent.
Use all available partitions - Dedicates the entire disk to a single data store extent. This option is available
only when adding an extent and when the disk you are formatting is not blank. The disk is reformatted, and the data stores and any data that it contains are erased.

7 Set the capacity for the extent. By default, the entire free space on the storage device is available.
8 Click Next.
9 Review the proposed layout and the new configuration of your data store, and click Finish.

What to do next
After you grow an extent in a shared VMFS data store, refresh the data store on each host that can access this data store, so that the vSphere Client can display the correct data store capacity for all hosts.

Thanks for your reply and your help Chilly, most appreciated!
 
Don't use extents and don't worry about this scratch space.

This. For the love of God and all that is sacred and holy, This.

Extents are so completely and utterly broken it's not even funny, and will never, ever, EVER, be fixed. Depreciated.

Command view - you have an EVA? It has a way to grow a lun. Grow it and use volume grow instead. PLEASE.
 
NetJunkie said:
Don't use extents and don't worry about this scratch space.
This. For the love of God and all that is sacred and holy, This.

Extents are so completely and utterly broken it's not even funny, and will never, ever, EVER, be fixed. Depreciated.

Command view - you have an EVA? It has a way to grow a lun. Grow it and use volume grow instead. PLEASE.
As I thought, I've never used or bothered with extents myself. The idea of spanning a filesystem over multiple luns just has never sat well with me.

I mentioned it before, and lopoetve stresses it as well, if you can grow the lun, then do that.
 
Back
Top