Thursday, July 28, 2016

VMworld 2016: The End of an Era

I sadly report that I will not be attending VMworld 2016 this year.  This will be the first conference I've missed - "13" really is an unlucky number!  I was not selected per the criteria my organization used to determine who goes and who stays.  I respect their decision, but that doesn't change the fact that I will be watching the general sessions remotely on a computer screen.  Yeah, that won't quite be the same - no assisting customers, meeting friends, attending sessions or keeping up with what's going on with partners in the Solutions Exchange this year for me.

My passion for the conference won't change - I still believe it to be the best IT conference and highly recommend attending.

So the VMworld Alumni Elite group will be at least one less this year.  What a prestigious group!   I know from meeting with these guys year after year that they have a real passion for VMworld, VMware and its products.  Guys - I'll catch up with you next time!

Its ironic - during my interview I was told that the surest way to not attend VMworld was to become a VMware employee.  My first year VMware decided to send my entire organization!  My second year they decided to send most of the organization, yours truly included.  This is my third year and it appears as though my luck has run out.

I did try multiple alternative options but none of those panned-out.  So all good things must come to an end?  C'est la vie - its the end of an era... and the start of a new era - carpe diem!

For those that are going - have a great conference!  For the reset of us, make sure to tune in to the general sessions which will be live-streamed again this year.  And don't forget about the breakout sessions that will be made available shortly after the conference.

Thursday, February 11, 2016

Enabling the Digital Enterprise: VMware Announcements

VMware made several major product announcements this week and I'm super-excite about some of the improvements and new features coming our way.

Our EUC team is really firing on all cylinders and continues to update and integrate products.  In this case AirWatch, Horizon and Identity Manager are brought together with Workspace One.  Did you know we coined the term "Workspace" (well okay, a company we acquired years ago did).  I used to be one of those guys that connected to my View desktop and worked from there.  Now, more often than not I browse our internal Workspace One portal and launch whatever app I need to get work done.  How did I work without this before?

Also worth noting are several SDDC product updates.  The big news here is VSAN 6.2 - what an awesome release!  VSAN was ready to host your tier 1 applications with the release of 6.0 - now with 6.2's dedupe, compression and RAID5/6 features, the nay-sayers won't be able say it's not ready for the enterprise (well, they can but they would be wrong).

Here is a "Did you know" PSA: 
Did you know that if you use vSphere Replication to replicate VMs hosted on VSAN storage, you can set the RPO as low as 5 minutes?
Now you know.

Finally. take a look at the vCloud Suite pricing and packaging changes.  We've really change the Suites to refocus on what customer's really want - the tools needed to run a true SDDC datacenter (wait, isn't that redundant?).
Replay’s of this week's events are available here: http://www.vmware.com/digitalenterprise

vROps Alert: One or more ports on the Distributed Port Group are experiencing network contention due to dropped packets

Here is a vROps problem experienced by one of my customers:
Alert: One or more ports are experiencing network connection
(One or more ports on the Distributed Port Group are experiencing network contention due to dropped packets)


However, this seems to only apply to vROps 6.0.x and the customer is running the 6.2.x release.  An SR was opened and the support tech recommended the fix per KB2052917: vCenter Server 5.1/5.5/6.0 performance charts report dropped network packets

Okay great, so a patch needs to be applied to the ESXi 5.5 host.  The customer tried to applied the patch via VUM but it was marked obsoleted.  "Obsoleted" is an Update Manager compliance state.  As explained in the Installing and Administering VMware vSphere Update Manager documentation:


This compliance state applies mainly to patches. The target object has a newer version of the patch. For example, if a patch has multiple versions, after you apply the latest version to the host, the earlier versions of the patch are in Obsoleted By Host compliance state.


(https://pubs.vmware.com/vsphere-50/index.jsp?topic=%2Fcom.vmware.vsphere.update_manager.doc_50%2FGUID-EBAA4F4A-57E0-45ED-8730-4B851FC846A9.html)

Hmmm.... so the host has a newer version of the patch, so now what?

After some discussion we landed on the fact that the vCenter instance managing this host was likely the culprit as it has a patch level several versions behind the host.  I always recommend customers keep their vCenter and ESXi hosts at the same Update patch level.  Installing patches on hosts in-between major Updates is okay.  In this case, they were on different Update versions.

The vCenter instance was patched to the latest update and voila, the error disappeared from vROps.



Monday, February 1, 2016

Horizon View Restart/Reboot Sequence

View admins - please make sure you restart View infrastructure servers in the proper order for fastest uptime!  I recently had a customer that rebooted vCenter and View servers in an effort to fix an apparent connection problem.  When they did not appear to be coming online fast enough, the admin rebooted them again.

Now normally rebooting these servers can cause ADLDS synchronization to take up to fifteen minutes or so, but rebooting them again, before the initial sync completes, can cause them to take thirty minutes or more - and then leave inconsistencies within the pools.

For more information on the proper boot sequence:
Restart order of the View environment to clear ADLDS (ADAM) synchronization in View 4.5, 4.6, 5.0, and 5.1

And to clear out any inconsistencies:
Manually deleting linked clones or stale virtual desktop entries from the View Composer database in VMware View Manager and VMware Horizon View


Thursday, August 27, 2015

VMworld 2014 Recap

Well, VMworld 2015 is next week so I guess I'd better get this one published!  I actually found this sitting in my drafts, I was waiting for a picture to come through which took so long I forgot to go back and publish it after receiving it!  So here it is in all it's glory...

This was my first VMworld as an employee.  My new role gave me a unique perspective to the conference.  Having attended every previous conference as a VMware customer has given me knowledge and experience I can pass on to my customers so they can get the most out of the conference.

Consequently, I walked through the Solutions Exchange but didn't spend as much time there as I normally do.  I attended a whopping one break-out session, and all of the general sessions.  Why?  Because I spent most of my time with my customers - assisting them in some form or shape.  I also spent a lot of time at TAM Customer Central.  What is TCC you ask?  It's a top secret location we reserve for VMware TAM customers where we provide special break-out sessions not available to VMworld attendees, receptions and meeting rooms among other things.

The theme this year was NO LIMITS.  Here is some cool desktop wallpaper for you:

And here is the Moscone all dressed up:

We were asked to provide pics of us with our customers for use as a collage at VMworld to be displayed at various times such as just before and after the general sessions.  Here's the pic I sent in with "Mr. Corvette" Jarod and his red sports car that made us mini-celebrities for the week:

Another one of those TAM customer benefits is the Hands-on Labs VIP tour.  This gives select customers a behind-the-scenes look at HOL operations and how the whole thing is managed.  Check out these stats:


Its interesting that the NSX lab remained the most popular lab during the entire conference - indicates a lot of interest by our customers.  And, consequently I think, this product continues to gain momentum in the marketplace.

And pics of how we used Log Insight and vCOps to monitor the environment via custom dashboards:




Even though I attended as a VMware employee this year, I was still invited to the CTO party (I had customers in attendance, so I'm sure that helped):

Why yes Johnny, everything is better with bacon:




And a pic from the Alumni Elite pre-party party - the pic I was waiting for as it was taken by someone else's camera - which wouldn't be complete without a shot with the big guy (me=literally, Mr. Gelsinger=figuratively):

And finally, lookie at what I found:

I found a "Meet the Experts" area hidden (not really) on the third floor of Moscone West.  If you have any technical or strategic product questions, these are the guys to ask and anyone can sign-up.  I highly recommend this resource if we offer it again in 2015.

Here's to seeing you there next year (he says with bacon in one hand, bourbon in the other)!

Thursday, August 6, 2015

Moving a DFS-Replicated Share to a New Drive

Here is the process I've followed successfully several times:

Add drives/VMDKs to file server VM

In the Windows OS:

  •    I used 500GB VMDKs and created a Spanned Volume
  •    Use GPT partition style to allow volumes greater than 2TB (future-proof the volume)
  •    Use Dynamic type disk to allow for Spanned Volumes
  •    I named the volume "Spanned" so I could easily recognize the drive

Copy Data to the new volume

Create Robocopy batch file on C drive:


robocopy d:\home n:\home /E /DCOPY:T /COPYALL /R:1 /W:1 /NP /LOG:c:\scripts\HomeCopy.log
The source path should include the directory on the root drive (probably drive D) because of inherited permissions
Run this once per day to minimize the amount of files required to be copied for the final sync
Perform the final sync

Stop DFS Replication service

Stop DFS Namespace service

Delete the share

Run last robocopy batch file

Make sure to close all sessions (except yours) and open files.

Delete the old pagefile

wmic computersystem set AutomaticManagedPagefile=False
wmic pagefile (lists pagefiles)
wmic pagefileset where name="D:\\pagefile.sys" delete



Switch Drive Letters

Final Robocopy job must be complete!
Review the log file to ensure all files have been copied in the final pass.
Note that Robocopy will probably error on "desktop.ini" files - these are unimortant and will re-created by the OS so you can safely ignore these errors on these files.

Delete all data on the replicated share

DFSR will notice the drive has changed and will try to do a full sync.  If you don't perform this step the replicated share destination will likely fill up and run out of space.

Start DFS Replication service
Start DFS Namespace service
OR REBOOT SERVER

Create Pagefile on new drive

Create new instance:

wmic.exe pagefileset create name="D:\pagefile.sys"

Set Size:

wmic.exe pagefileset where name="D:\\pagefile.sys" set InitialSize=8191,MaximumSize=8191

Dont' use "MB" at end of size (ulike MS KB); Set for 8191MB (Good for 8GB system)
shutdown -r -f -t 0

For memory dump files and other settings, see:
http://technet.microsoft.com/en-us/library/ee424384(WS.10).aspx

Post Move Testing
Test namespace
Test replication
Test VSS

Troubleshooting

Check the log files at C:\Windows\Debug

If you see the message "content not set" in the log run the following command to set primary replication membership:
DFSRADMIN Membership Set /RGName:<replication group name> /RFName:<replicated folder name> /MemName:<member you want to be primary> /IsPrimary:True

See:  http://technet.microsoft.com/en-us/library/cc725893(WS.10).aspx - What happens during the initial replication

Check the DFS Management MMC

Re-add namespace server if its missing.
Also try disabling/re-enabling target folder.

Friday, August 22, 2014

Back in Black!

I hit the sack.  It's been too long I'm glad to be back...

Yes, it has been one year since I've blogged on VMware and virtualization.  Well, in my defense, a lot has happened keeping me from spending quality time with my beloved blog.

My biggest excuse = I joined VMware!  I am a Technical Account Manager for customers in the Dayton/Cincinnati area.  Shameless plug = if you would like to know more about what a TAM can do for you, please reach out to me by leaving a comment and I'll get back to you.

I was warned that the first year is like drinking from the fire hose and I found that to be true, but in a good way.  So yes, I did neglect my blog to the point of having to delete drafts that are no longer relevant.

But the good new is:  I'm back!  I've personally re-committed to posting articles on a more-frequent/regular basis (although not on a specific schedule, no change there).

Note that the opinions in my blog are mine alone and do not necessarily reflect that of my employer.  Also note the subject matter won't change either - I won't post "newsie" articles as there are too many other blogs that do a good job of that.

What I will publish are articles describing technical problems I'm seeing in the field along with their resolution.  I may also post articles of a more strategic nature on occasion.

My experience with customers in the field gives me a unique perspective (and fresh content!) on many IT topics.  Three of my accounts are in healthcare so I may also publish articles discussing challenges specific to that segment.

Finally, next week is VMworld 2014!  I'm doing things a little differently this year - instead of one VMworld recap article (which I may still publish), I'm going to try to post bits through-out the week so stay tuned!