BladeSystem c7000

We have ladeSystem c7000 det up by external consultant and he said the following:
Out of your 8 blades you have to dedicate one blade for blade system management. He installed win2008 server on the 8th blade and dedicated it EVA Command and View and some other HP Management software. (see attachments)

I'm curioius why would we waste the whole blade.... we have 7 blades dedicated to VMWare, can we just remove Windows 2008 from this server and dedicated it to VMWare as well?

Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

We normally implement a non Blade System for management purposes of EVA storage (like DL360). I consider it a waste of a precious blade slot and relative connectivity ports of the interconnects (ethernet and fc switches).

On the other hand a server for hosting the management tools (specially the EVA) is necessary so if you don't want to use a separate server then a blade is the only solution.

Remember that you can always connect directly to the Blades Manager via web interface (and to servers with iLo) from any computer in the LAN.
andyalderSaggar maker's framemakerCommented:
HP's requirement for a dedicated server to run Command View EVA does bump the cost up a bit, especially since it runs quite happily with the CV EVA server turned off. Still, as r_panos says you can use one of the rack servers you are migrating to VMware from since it doesn't need much RAM or CPU.

BTW, I hope you haven't got the blades populated as per that Visio, it should be filled from left to right. "Half-height server blades should be populated in the front of the enclosure from top to bottom and from left to right. The first two half-height server blades should be placed in bays 1 and 9; the second two half-height server blades should be placed in bays 2 and 10 and so on, until the enclosure is full."
elo-miloAuthor Commented:
Great, so I let me sum it up:
Currently the whole blade in slot #16 is taken by the server that runs only CV EVA, an for that purpose all the resources of this blade are dedicated - 40G of RAM and 2 quad-core processors. In my opinion it is a crime.
So I'm pretty safe just grabbing a simple 1U server, instaling CV on it and using the existing blade for something else?


The blades are actually set up exactly the same way as shown on the picture (it is a snapshot from existing configuration - onlide blade configuration).  So in fact they should be populated like that: (nubers signify the servs, not the slot numbers)
1 3 5 7 9   11 13 15
2 4 5 8 10 12 14 16
Can you explain why it is crucial to do it that way and not acceptable the way it is configured now?

SolarWinds® IP Control Bundle (IPCB)

Combines SolarWinds IP Address Manager and User Device Tracker to help detect IP conflicts, quickly identify affected systems, and help your team take near instantaneous action. Help improve visibility and enhance reliability with SolarWinds IP Control Bundle.

Yes that's exaclty elo-milo, considering the fact that the management tools are resource consuming when you use them :-). 1 unit DL360 with 2 or even 4 GB (the fabric ram depends of the model) is more then enough.

Andyalder is right. That's the way the blade slots must be populated (by design). The way the Blades have been implemented in your enclosure doesn't even allow you to add Full-Height Blade server

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
andyalderSaggar maker's framemakerCommented:
CV EVA requires single core 1.26GHz CPU and 2GB RAM minimum.

It really isn't too important to populate the blades from left to right if you have a full set of fans but it saves power since the fans at the right hand side where there shouldn't be any blades in a half populated enclosure will go very low speed if it's populated properly, they may even power off completely, I'm not sure on that though.

If you look at "Appendix: Fan and server population guidelines" in you'll see that with 8 blades you should have server Zones 1 and 2 populated which means just 6 fans are required.

I'm just surprised that an external consultant got the population order wrong, again it's not critical since the enclosure won't power on the blades if the cooling isn't there for them but it's best practice. If they missed that I bet they forgot to discuss whether you wanted N+1 or N+N power redundancy as well.
elo-miloAuthor Commented:
Great, thanks guys.

"I bet they forgot to discuss whether you wanted N+1 or N+N power redundancy as well."
--- The more I dig into what they did, the more I see that they brainwashed the guys here, for example they promised 100% uptime in VMware environment and I showed that it was not configured to support it... Anyway, speaking about N+1 or N+N power redundancy - can you describe it in few words or provide a link to read about it? (it might be in the same link you sent - I just started reading it)

andyalderSaggar maker's framemakerCommented:
If you've got two power feeds (from two UPSs or left/right feeds as in a datacentre) then to tolerate a feed failure you want N+N, so if 2 PSUs are needed to run your blades then 4 will be powered on. If you've only got one feed then all you need is to cover for a PSU failure so you use N+1 so if 2 PSUs needed 3 will be powered on. You can set that up from the management GUI.

100% uptime isn't possible on anything, there's a tiny chance a PSU will fail short circuit and pull the entire power bus down. There's also a chance that both controllers in the EVA will crash at the same time, especially if they didn't upgrade the xcs firmware from 09521000, may say they haven't heard any negative reports but that's because they stick their fingers in their ears, we had double controller crashes with that flakey version.
elo-miloAuthor Commented:
Great. Thanks.
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Server Hardware

From novice to tech pro — start learning today.