Category: Uncategorized

VMAX3 Deleting volume stuck in RDF mode.

I recently had some volumes set in a storage group also set in RDF.   They were in the RDF1-TDEV mode. I removed them from the storage group without first breaking the RDF.  I then found out the hard way that it was impossible to delete that volume until the RDF was broken.  I need the volume in the TDEV mode so it can be deleted.  To do this I had to go into command line and use solutions enabler.

First I had to find the volume I wanted to delete and the RDF group it was in.

symrdf -sid 123 list -rdfg -all

That gave me a list of volumes and what volumes it was being synced up to.

I then created a text file with the two volumes in it.

00010 00020

To break this pair I ran this command

symrdf -f text.txt suspend -sid 123 -rdfg # -cons_exempt

Then I can finally delete the pair.

symrdf -f text.txt deletepair -sid 123 -rdfg #

Now that is done I can finally delete the volume.

A lot of work just to remove 1 volume.

0

Clearing Microsoft Cluster Persistent Reservations

Recently I had the active node on a Microsoft cluster go down hard.  There were 3 applications running in this cluster and 2 came up.  The third would not come up so with a little Googling I found this article from SQLShip.

Microsoft clusters put a reservation on the disk to prevent other nodes from being able to write to them. If that reservation is not cleared and is brought offline then no one will be able to use it.

Go to command prompt with elevated privileges and type in this command.  Replace the # with the disk number that is giving you issues.  This should clear the reservation and allow you to bring everything up.

cluster node servername /clearpr:#

 

0

Setting up vCenter Update Manager

Recently I was trying to setup vCenter Update Manager for vSphere 6.  I setup the ODBC connection for 32-bit using the ODBC Driver 11 for SQL server.  Setting up this connection was successful, but when I tried to run updates in my vCenter environment I would get an error stating that it could not connect to the database.  This was an odd error because I knew the ODBC connection was successful.  I contacted support and all they told me to do was build up a new update server.  I tried that and it did not work.  So after some deep googling I found a website which stated that you had to setup the update manager service to have a log on as.  (Sorry this was done a few weeks ago, and I do not know what website it was)

So when setting up update manager make sure you configure the server to log on as a certain account.  I did not notice that in any of the documentation.  Maybe I some how missed it.

0

Delete a SRDF pair on VMAX3

Recently I have had some issues with SRDF on the VMAX3 that I manage.  I was having issues trying to resolve it using the Unisphere GUI, which can be very limited in its capabilities, so I had to switch to command line.  First you create a text file and put in it something like “00001 00002” (No quotation marks in file).  Listing the two devices that need there pair broken.  Then you run this command.

symrdf -sid 123 -rdfg 1 -file rdf.txt deletepair -force

 

0

vBrownBag and preperation for the VCIX

Currently I am working towards the VMware VCIX6-DCV certification.  To prepare for this I am going through the vBrownBag videos on the VCP6-DCV certification.  It is a good deep dive that covers a lot of things you may have missed or forgotten over time.  It has helped me improve things in day job that I was not doing properly.   If you are working towards an VMWare certification or just trying to learn more. I really recommended going through the videos that they have put together.  They have also covered teh VCP6-NV and VCP6-CMA.  Hopefully soon they will cover the VCP6-DTM.

0

Using PowerCLI to change the Syslog server for your ESX host.

I recently discovered SexiLog to use with VMware ESX host.  I needed a quick and easy way to change where the ESX host sent there logs to.  Doing some google searches I was able to put together a quick PowerCLI script.  In about 5 minutes I was able to change the syslog for all of my ESX host.

$ESXLOG = “1.2.3.4”
Get-VMHost | Foreach {
Write-Host “Adding $ESXLOG as Syslog server for $($_.Name)”
$SetSyslog = Set-VMHostSysLogServer -SysLogServer $ESXLOG -SysLogServerPort 514 -VMHost $_
Write-Host “Reloading Syslog on $($_.Name)”
$Reload = (Get-ESXCLI -VMHost $_).System.Syslog.reload()
Write-Host “Setting firewall to allow Syslog out of $($_)”
$FW = $_ | Get-VMHostFirewallException | Where {$_.Name -eq ‘syslog’} | Set-VMHostFirewallException -Enabled:$true
}

0

Using Solution Enabler to Delete a Storage Group

Recently I have been working on cleaning up storage on a EMC VMAX3.  You can do a lot using the GUI for most of the day to day maintenance, but using the CLI is a lot more efficient when it comes to removing storage groups.

symsg -sid 1234 delete storgegroup_SG -force

Using that command is a simpler and quicker way to remove storage groups than to use the Unisphere GUI.

0

Network, Meditech, and a VMAX3 failed backup

Recently the backup server rebooted itself in the middle of Networker running Meditech Magic backups.

Then all the Meditech backups would fail after 2 minutes.  This was caused by locked volumes with the VMAX3.  Backups with Meditech use BCV (Business Continuity Volumes) which are synced with TDEV volumes (Thin Devices).  These two volumes stay synced up at all times.  Then when a backup starts this connection is broken, and the backup is pulled from the BCV.  When the backup finishes the BCV reconnects to the production TDEV and syncs again.

Here are the commands I used with Solutions Enabler to unlock the volumes.

symdg -sid 1234 list (This will give you a list of all the LUNs on your server in the Device Group

symdev -lock list (This will give you a list of all LUN that are in locked mode)

symdev -lost # release (This will release the lock on the volumes)

Now you should be able to run the Meditech Magic backup.

0

Using Ravello Systems for your lab.

Creating a home lab can get very expensive. Turning to the cloud can be a cheaper and easier way to go.  I decided to setup my lab using Autolab on Ravello.  Running my Ravello lab cost about $2 an hour.  That can add up to a lot of savings by not using all the hardware, power and cooling that I would need at my house.

Using Autolab will spin up your lab in a matter of hours instead of days if you were doing it yourself.  Basically your provide ISO’s and it will run the scripts to spin everything up.  There is even a template already in Ravello that makes spinning up your lab even faster.

In the next few weeks I will show how I setup everything with Ravello.

0

Blog at WordPress.com.

Up ↑