Link to home
Start Free TrialLog in
Avatar of jnordeng
jnordeng

asked on

So, one of our 4 production DDC's was upgraded, but DB not updated, why can't I access Studio on any DDC's.

We have 4 DDC's in our production environment.  We are running XenApp 7.15.5.000 in prep for an upcoming CU7 upgrade, 1 of the DDC's was upgraded, but the database automatic upgrade nor the database generated scripts were run.  However; Citrix Studio in none of the DDC's will display the contents of the farm, they all prompt to run automatic updates or create the manual scripts to update the DB backend.

Thankfully users are not impacted to use Citrix and can launch published apps, however; this means we can't manage the farm.

I would have understood if we updated the database, but was simply the node that was done.

Is there a way to get the other 3 DDC's displaying Studio contents correctly?  Or is there a way to degrade the 1 DDC without impacting the farm and again, getting Studio contents displaying correctly?

Thanks in advance.  Though we followed the doc, that's why we stopped the upgrade at that point and didn't run the DB scripts.  I was able to view Studio contents until the session timed out on the DDC I was on which was not the one upgraded.

Thanks in advance.
Learning curve for sure, thought this part would have been independent.
Avatar of Carl Webster
Carl Webster
Flag of United States of America image

Are you following these instructions?

Upgrade a deployment​​​
Avatar of jnordeng
jnordeng

ASKER

Well nope I guess not the one I was following.  The VDA's still need to be upgraded and there says you can do 1/2 the DDC's. so only one was upgraded, so why isn't Studio working on the rest?
Did you see where you upgrade half the DDCs AND run the DB upgrade?

Did you already upgrade your license server, StoreFront, and Director?

As long as users are not impacted, follow the steps.
License server yes, Director in Test was presenting errors so no, this was kind of an accident really, that's why one was done...  StoreFront is not complete either.  The article you provided is much clearer on what needs to happen when than what I saw.
If we were to run the DB scripts in the environment in this state will it break it worse?  Anyway to fix it?  Users are not impacted but we can't manage anything.
Those are the steps I always follow. Did you not test the upgrade sequence in your lab first?

Did you back up the databases? If so, open a Citrix support case and have them walk you through a rollback to your original state.
Ok, yes, i have verified the backups, the test environment is much smaller, not a problem, this was just a whoops and paying for it.  Will open a ticket and use this doc for future reference, laid out better than what I saw.

Thanks
I still don't understand how the database is impacted at this point though since the DB scripts were not run.
If you look at the steps Citrix outlined, you see the DB upgrade is done, you run the Site check to make sure there are no issues, and immediately upgrade the remaining DDCs.

Did you run the Site check BEFORE you started the upgrade process?
Please don't take my responses as condescending,  I am trying to help and teach and explain.
I appreciate your help, I did run the preliminary checks within the installer, they all came out ok. So it's just a matter, do I have the DBA's run the scripts is this the only way or roll back.  i did open a ticket, but now I see in the article you sent  that I likely have to do that.  For example, let’s say an upgrade includes new versions of the Controller and Studio. You upgrade the Controller but you do not run the installer on the machine where Studio is installed. Studio will not let you continue to manage the Site until you upgrade Studio.

Studio is installed on all the VDA's but they point to the Delivery Controller's Studio session.  As on the Delivery Controller's I have Studio and Director as well.  
I'm ok with the DBA's doing a restore, but the DB wasn't actually upgraded why I'm so confused.  I also see that Remove any hotfixes applied after the backup before performing the restore. so to me that would imply removing CU7 on the one Delivery Controller.  Maybe I'm wrong there.
Just wanted to note here Citrix Supports recommendation for anyone else that may have experienced this.

  • Make sure you have a good backup of the database prior to the issue
  • Use this article to obtain the SID of bad node and generate SQL scripts, 
  • https://support.citrix.com/article/CTX139505
  • Run the commands at the SQL level to do an evict of the Bad Node (Upgraded Node)
  • Verify bad node no longer shows as an Active DDC within Powershell on a good node
  • Restore good backup of all 3 Citrix DB's at the SQL level.
  • Open Studio and/or reboot box and open Studio.
  • Remove software on bad node - If you leave CU7 on here after restore, it will be detected and go into a bad state
  • May need to remove the bad node manually in Studio after you have removed software from DDC
  • Verify all is functional minus bad node.

Then go back to bad node and fix the issue by reinstalling CU5.
ASKER CERTIFIED SOLUTION
Avatar of jnordeng
jnordeng

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial