Live Migration Can Benefit From Jumbo Frames

Posted on by
5

Does live migration benefit from Jumbo frames? This question always comes back so I’d just blog it hear again even if I have mentioned it as part of other blog posts. Yes it does! How do I know. Because I’ve tested and used it with Windows Server 2008 R2, 2012 & 2012 R2. Why? because I have a couple of mantra’s:

  • Assumption are the mother of all fuckups
  • Assume makes an ASS out of U and ME
  • Trust but verify

What can I say. I have been doing 10Gbps since for Live Migration with Hyper-V. And let me tell you my experiences with an otherwise completely optimized server (mainly BIOS performance settings): It will help you with up to 20% more bandwidth use.

And thanks to Windows Server 2012 R2 supporting SMB for live migration we can very nicely visualize this with 2*10Gbps NICS, not teamed, used by live migration leveraging SMB Multichannel. On one of the 10Gbps we enable Jumbo Frames on the other one we do not. We than live migrate a large memory VM back and forth. Now you tell me which one is which.

image

Now enable Jumbo frames on both 10Gbps NICs and again we live migrate the large memory VM back and forth. More bandwidth used, faster live migration.

image

I can’t make it any more clear. No jumbo frames will not kill your performance unless you have it messed up end to end. Don’t worry if you have a cheaper switch where you can only enable it switch wide instead op port per port. The switch is a pass through. So unless you set messed up sizes on sender/receiving host that the switch in between can’t handle, it will work even without jumbo frames and without heaven falling down on your head Smile. Configure it correctly, test it, and you’ll see.

DELL Server DRAC Card Soft Reset With Racadmin

Posted on by
4

Sometimes a DRAC goes BOINK

Sometimes a DRAC (Dell Remote Access Card) can give you issues. Sometimes it’s some lingering process or another hiccup that causes this. You can try a reboot but that doesn’t always fix the issue. You can go into the BIOS and cancel any running System Services. A “confused” DRAC card can also be fixed by shutting down the server and cutting power for 5 to 10 minutes. That’s good to know as a last resort but not very feasible a lot of times, bar a maintenance window when you’re on premise.

You can also try to do a local or a remote reset of the DRAC card via OpenManage  (OMSA), racadmin. See RACADM Command Line Interface for DRAC for more information on how and when to use this tool. The racadmin can be used for a lot of remote configuration and administration and one of those is a “soft reset” or basically a powercycle, aka reboot, of the drac card itself. Don’t worry your server stays up Smile.

Local: racadmin racreset soft

Remote: racadm -r <ip address> -u <username> -p <password> racreset soft

Real life example

I was doing routine maintenance on 4 Hyper-V clusters and as part of that DUPs (Dell update packages) were being deployed to upgrade some firmware. This can be automated nicely via Cluster Aware Updating and the logging option will help you pin point the issue. See https://blog.workinghardinit.work/2013/01/09/logging-cluster-aware-updating-hotfix-plug-in-installations-to-a-file-share/ for more information on this.

Just like we found that the DRAC upgrade was not succeeding on two nodes.

One it was due to the DUP not being able to access the Virtual USB Device

Software application name: iDRAC6
   Package version: 1.95
   Installed version: 1.92

Executing update…

Device does not impact TPM measurements.

Device: iDRAC6, Application: iDRAC6
  Failed to access Virtual USB Device

==================> Update Result <==================

Update was not applied

================================================

Exit code = 1 (Failure)

and the other was because there was some other lingering DRAC process.

 iDRAC is currently unable to process this request because of another task.
  Please attempt one or more of the following steps to cancel the pending iDRAC task:
  1) Wait 30 minutes and retry your request.
  2) Reboot the system; Press F10; select ‘Exit and Reboot’ from Unified Server Configurator, and retry your request.
  3) Reboot the system; Press Ctrl-E; select ‘System Services’. Then change ‘Cancel System Services’ to YES, which will close the pending task;
      Then press Enter at the warning message. Press ESC twice and select ‘Save Changes and Exit’ and retry your request.

==================> Update Result<==================

Update was not applied

================================================
Exit code = 1 (Failure)

They give some nice suggestions but the racreset is another nice one to have I your toolkit. It’s fast and effective.

Run racadmin racreset soft

image

Wait for a couple of minutes and then run the DUP or the items in SUU that failed. With some luck this will succeed now.

image

Attending The Global MVP Summit 2013 (November 18-21)

Posted on by
1

We have Windows 8.1 running on our desktops & laptops and meanwhile Windows Server 2012 R2 is crunching numbers in our (virtualized) data centers. So it’s time to grab one of those magnificent British Airways Boeing 747 aircraft seats once again and make my way to SEATAC. No rest for the wicked. BAS7400066

Yup, but for now I’ve parked myself in LHR whilst waiting for my flight. Soon I’ll be in the air again for the long haul to the USA. I’m off to Washington State, Seattle to be exact, and from there to Bellevue/Redmond. You might have guessed where I am going already, indeed to the Microsoft campus. I’m attending the Global MVP Summit 2013, November Edition. image

Apart from that magnificent educational & networking opportunity I will spend a lot of the “free” time discussing technology, visions & strategies with my peers and Microsoft employees. I’d like to thank the latter for their patience with me when bugging them with questions Smile. To my buddies, acquaintances & connections, I’ll see you soon. We have a lot to learn & discuss. That’s one of the reasons I’m off a bit earlier. It helps with the jet lag but it also gives me time to meet up with friends and acquaintances I’ve made in the Puget Sound area and talk shop. This helps to keep in touch with what’s happening over the world and to understand where their priorities are, what’s keeping them occupied. While I’m a firm believer in remote and teleworking there is value in getting your boots on the ground every now and then. It prevents tunnel vision and helps avoiding  teleology in our views while enhancing early detection of small trend changes to whole sale tectonic shifts. This is not to be confused with thinking you have a crystal ball or anything.

To my readers & community members I’d like to extend the invitation to pass along feedback to Microsoft. They do listen. So leave a comment, send me a mail (contact via Blog) or ping me on Twitter.

In case you don’t know, everything discussed at the MVP Summit is under NDA, even for MVPs of another expertise. So basically bar some tweets to find our where other MVPs I’ll be going dark.

How To Save A Company From Death By Meetings

Posted on by
5

I had a very interesting discussion with a fellow virtualization expert & strategic advisor at E2EVC Rome 2013. We discussed many issues and the topic came up that we see way too many potentially strong organizations sub come to “death by meetings”.  Instead of fixing this they are treating the symptoms and are declaring the symptoms to be “illegal”. It’s almost pandemic. You really need to opt out of this this madness for the sake of the company and getting work done.

I think we need to take this further. Bar reverting to the tactics of a UK manager who locked all meeting rooms and took the keys away whilst telling his staff to stop meeting and start working, we should implement these rules:

  1. There is an absolute maximum of 5 hours meeting per 8 hour workday. Work less than full time? Adjust accordingly. This is non negotiable for anyone. You must decline any meeting that violates this rule. You must not strive for this maximum.
  2. You must decline any meeting that has no agenda.
  3. No meeting can have more than 5 attendees unless a very valid reason and need is motivated in the agenda. If not, you have to decline the meeting.
  4. Meetings have to be planned in advance & cannot be made permanent. That is reserved for councils or meetings of the board.
  5. Everyone, from the lowest pay grade to the top manager hast to abide by these rules
  6. Teleconferencing is a perfectly valid way of meeting and is included in these 5 hours.

meetingsdemotivator[1]

People, really, stop meeting for 8 hours a day. It’s scary so many companies can’t get a grip on this. Your thoughts?