Jump to content
Welcome to our new Citrix community!
  • 0

Internal Error: Cluster already running; Try destroying first (???)


Carlo Piva1709156772

Question

I have been having nothing but issues trying to get clustering and HA working.

I finally got the two hosts working together.

I had enabled multipathing

I got them talking to the SAN

I got the HA iscsi DB connected to its own lun

I HAD enabled cluster mode

I enabled HA mode

HA mode keeps saying it is not enabled.

I checked cluster mode, and it is showing as disabled now and I can't seem to re-enable it.

What makes this worse is that I can't find any real step by step documentation on this.

Citrix support on the phone has been basically useless and has resulted in me having to reinstall Xenserver a few times on these two hosts.

 

Background:

I know for quorum I should have three hosts minimum.

Unfortunately, at this specific moment, that is not an option for me.

I have been given 2 hosts (repurposed from Hyper-V environment that my manager didn't want to have in the new Hyper-V cluster)

I have installed XenServer 8.1 on both and configured the pool, and san connections.

He wants me to copy the vm's from the other two standalone servers (running 7.1) to the new hosts and san, but he does not want me to upgrade and incorporate any of those standalone servers into the pool, and to leave the live servers on them until I have shown him that with the two "new" servers that everything is working well and stable (which is understandable and reasonable except the part where I need three hosts....).

Which is frustrating because he isn't understanding we need three servers to satisfy quorum.

So what happened when I was performing a failover test with a single running VM (I put the slave into maintenance mode), something went wrong, the slave host rebooted and the master self fenced (which was rather predictable :34_rolling_eyes:, except the slave host rebooting - that was unexpected)

I managed to unfence the master host, I was able to start vm's on both hosts then shut them down.

I went to back track settings so that I can try to figure out what had occured, only to find that Cluster mode had disabled itself and I can't re-enable.

 

HELP :57_cry:

 

Link to comment

8 answers to this question

Recommended Posts

  • 0

Thanks Alan,

I'll look into that shortly but i'll talk to my manager first.

I dont want to get into a situation where we are relying on HA lizard, once we are able to add one or both of the other servers (currently stand alone) to the mix.

 

I guess I have three other questions - should I have enabled HA before enabling Cluster Mode?

Or maybe I screwed it up by enabling Cluster Mode before HA mode?

Also, can I recover from this issue without having to do a full reinstall?

Link to comment
  • 0

Maybe a terminology thing, but I just assumed by clustering you meant servers joined to a pool.

I would want all of my servers in the pool before enabling HA. HA should be the last thing you

enable and once enabled it needs to be the first thing disabled before doing most XE commands.

 

--Aaln--

 

 

Link to comment
  • 0
20 minutes ago, Alan Lantz said:

Maybe a terminology thing, but I just assumed by clustering you meant servers joined to a pool.

I would want all of my servers in the pool before enabling HA. HA should be the last thing you

enable and once enabled it needs to be the first thing disabled before doing most XE commands.

 

--Aaln--

 

 

If you right click on the Pool name in the resource column,  there is the option to enable clustering

image.thumb.png.8adeb1511130bbecf97664c6779a4e5c.png

 

I had this enabled before the failover test (which failed dramatically) but it seems to have disabled itself, and I can't seem to re-enable.

image.png.e43fe4eead8de26775952c4e72fe4861.png

 

and HA is currently disabled

image.thumb.png.a44f2a68c0a214a2bd9b538575a415f9.png

 

In further trying to test just now, I tried to enable HA, but it is not enabling.

Error is as such:

image.png.875e5b36f8b6c486e462051c0139adb1.png

Link to comment
  • 0

Thanks both of you.

I'll be taking both of your advice for this.

My manager is headed to the data center now with a unit of the same make and model as the one that started the problem (Dell R710) and the hope is we can try to everything at least temporarily working as a three host cluster. The one he is bringing down - we wont be able to configure HA as it has 1/6 the ram as the other two, and only has 4 nic's on it. I had used it for the the last few years as a test server inside the branch. 

I'll reinstall everything fresh on these servers, not enable cluster mode, and just get everything working so that I can copy the vm's to the san.  When we are ready to upgrade the true three live servers (2 Dell R630's and a Dell R730 - I am told there will be another R630 or R730 available for me for this in a few months, bring the cluster to 4) I'll configure the HA mode. As I'm not going to use GFS2 any longer (to many problems right now) I won't enable "cluster mode".

 

In retrospect I think enabling "cluster mode" was probably causing much of my issues yesterday.

As I was setting up a cluster, I just assumed (bad I know) that I needed to enable "Cluster mode".

 

Again, thanks both of you, your advice is much appreciated!

 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...