VM-Aware

I have almost completed an installation of DataCore’s SANmelody storage virtualisation product and wanted to find out about how the wider SQL Server (in my case) community feel about storage virtualisation.

It just so happened that one of the articles I ended up reading was written by the always excellent Brent Ozar. It discusses the topic using real life use cases and quite a lot of detail.

For the customer I have implemented DataCore, they don’t quite need to worry about some of the issues that Brent points out, but some of the points are certainly worth considering.

Nick Triantos has posted another brilliant post over at his Storage Nuts & Bolts blog about VMware Site Recovery Manager and the adapter that NetApp have created to plug into SRM.

The article talks about what is supported in the SRA (Site Recovery Adapter) including:

• SnapMirror integration with SRM
• Support for Sync, Async & Semi Sync
• Dissimilar H/W configuration support
• FC & iSCSI Support
• FlexClone support for DR testing

He also ponders on what VMware might want to consider for future releases including:

• Multiple live sites to a single DR site
• Failback as well as Failover
• NFS Support

If you can it is well worth bookmarking his site, or even better using the RSS feed to get the new posts immediately.

I recently came across a problem whereby a Rescan SAN was taking an absolute age, like 15 minutes from the VI Client. Now I love the VI Client, but sometimes it just doesn’t give you enough information when it is doing things. To be fair most of the time the verbose information is simply not needed, but when things go wrong it is a god send.

In comes the Service Console command line, more specifically the esxcfg-rescan and esxcfg-mpath commands. (more…)

I don’t know if you are like me and feel initially a little bit nervous about messages that contain phrases like “there may be some interruption to I/O”.  This is the type of message that you get when you make any changes in the Zoning Admin area of a Brocade switch when you’ve made changes to an effective config.

I recently added a lot of new zones to a customer fabric as they were implementing DataCore.  With this brought lots of new WorldWide Names and some of which needed to be kept completely separate from the ESX hosts.  I went ahead and created all my zones and saved the config using the Save Config Only option.  Nothing happens at this point as it isn’t actually updating the live config just saving the changes.

If you want to apply the changes to make them live then you’ll need to use the Enable Config option and with that brings the dialog to choose the config to apply.  Choose the correct config and then you are warned of the possible interruption to I/O.  This is where I was slightly perturbed.

I can confirm that VMware ESX 3.0.2 & 3.5 does handle any interruption (which appeared to be either minimal or non-existent) seamlessly.

That said, if you can get your customer to agree to do it out of hours this is obviously preferable, however the customer I was at is a 24/7 operation.  One other thing I can’t vouch for as yet is how well Windows nodes handle the config update so progress with care here.

Following on from an earlier post NetApp have released an updated best practices guide for the Open Systems SnapVault.

A 24 page beauty that will get you heading in the right direction with this new development.

Download the pdf here.

I’m sure that most people are either aware of these documents already or would have stumbled across them at some point.  Here though is the link to all the SRM related documentation on the VMware website.

Of particular interest to me was the storage vendor HCL.