August 14th, 2008 @ 8:41 am
I’m excited about this month’s House Pay-Off Spectacular. The green and pink are actually touching now! I’m praying and looking forward to being able to go all green very, very soon! Hopefully in the next month. At this point, I’m just waiting on some funds in CD’s to become available.
This is awesome!
Total Squares: 562
Paid-For Squares: 323 336
Savings to be committed to mortgage: 232 219
Squares Remaining: 7 0

August 14th, 2008 @ 8:34 am
If you are not aware yet, a major bug has been revealed in ESX 3.5 and ESXi 3.5 Update 2. Apparently, the beta was coded to expire on August 12, 2008 and this code failed to be removed from the actual release. Details are available in the VMware Knowledge Base and This Topic on their forums. You might also checkout This Post on Matthew Marlow’s blog for more information.
On the morning of the 12th, I was greeted with several errors like this one in the logs for our ESX cluster:

VMware finally released the patch really late Tuesday night, which, unfortunately kept me up most of the night getting our cluster patched. It involved setting back the clock on all of the hosts so Vmotion would work, manually migrating VM’s off of a host, going into maintenance mode, applying the patch via the command line, then migrating the VM’s back.
VMware is one of my favorite companies and it know for delivering rock-sold, enterprise-class products, so it really disappoints me that they would let something like this slip through the cracks. Imagine how many hundreds of thousands (Maybe millions?) of VM’s this affected. They do seem to be committed to fixing their mistake and making things right. You can check out the Letter From Their CEO for more info.
August 6th, 2008 @ 11:51 am
I had a conversation yesterday with Bakbone about their NetVault product. As we’ve moved heavily into virtualization (90% of our infrastructure is virtualized at this point), backup and DR has become a growing challenge. Ideally, we need to be able to back up entire virtual machines directly from the SAN, with the ability to restore and entire VM, or individual files within a VM. In addition, properly protecting Active Directory, SQL Server, and Exchange are high priorities. The ability to do message level restore in Exchange is also somewhat important.
Our aging Backup Exec installation seems to become more and more cumbersome and problematic, and seems to have the common problem of one product trying to do way too much and not doing any one thing exceptionally well. I think it’s time to move into a more enterprise-class product - something more closely tuned to our needs. NetVault initially seems like a potentially good fit. If anyone has any experiences with NetVault or has any other recommendations, I’d love to hear from you.
August 3rd, 2008 @ 2:42 pm
Last December, we implemented 7TB of Equallogic storage as the backbone of our VMware Virtual Infrastructure implementation, as well as to serve as primary storage for our file, Exchange, and SQL servers. Little did we know that 6 months later we’d be near capacity and shopping for more storage.
Thanks to James at EIS, we now have another 16TB of raw storage online. Combined with our existing array and considering RAID overhead, we now have just under 15TB of usable iSCSI storage. I’m excited to have this done!
I absolutely love our Equallogic SAN! In less than 30 minutes, the new storage array was configured and added into the cluster. The volumes were automatically distributed and network traffic load balanced among the arrays. The only complaint I have is that their rack rail system could use some improvement. Getting the array installed in the rack is the most time consuming part of the entire implementation.
Check it out:
Array was sitting in my office when I arrived on Friday:

Unpacked and ready to be installed. It has 16 drives with a capacity of 1TB each:

Racked next to it’s little brother. This is a total of 30 disks and 23TB of raw storage capacity:

Total storage capacity of 14.82TB:

The performance and raw throughput of the Equallogic gear amazes me. Orion was reporting 534mbps of traffic between the arrays during setup:

Hopefully we will have plenty of space for a while now. Although, as we move toward implementing Final Cut server and centralized storage for digital video, that may change.