Terracotta IS clustering issue at IS startup

,
1

I have 2 9.8 IS’s installed along with Terracotta for clustering purposes only. I have Terracotta starting up just fine on both IS’s. However during IS startup on both IS’s I’m getting the following:

Caused by: java.lang.RuntimeException: com.tc.config.schema.setup.ConfigurationSetupException:

Could not fetch configuration data from 2 different configuration sources. The sources we tried were: the server at ‘yvm6batchstg01..com:9510’, and the server at 'yvm6batchstg02..com:9510’. Fetch attempt duration: 60 seconds.

Anyone see this before?

2

Make sure the Terracotta server is up and running before starting the Integration Server.

Let me know if you have any questions.

3

Yes Terracotta is up and running on both cluster servers

I first get this on the server.log:

2015-07-27 16:51:07 CDT [ISS.0017.0001E] An error was encountered parsing SoftwareAG-IS-Core.xml: net.sf.ehcache.CacheException: Unable to load class net.sf.ehcache.terracotta.ExpressEnterpriseTerracottaClusteredInstanceFactory. Initial cause was org.terracotta.toolkit.ToolkitInstantiationException: java.lang.RuntimeException: Unable to create toolkit.

then eventually in the stack trace you see this:
Caused by: java.lang.RuntimeException: com.tc.config.schema.setup.ConfigurationSetupException:

Could not fetch configuration data from 2 different configuration sources. The sources we tried were: the server at ‘wm9batchstg01..com:9510’, and the server at 'wm9batchstg02..com:9510’. Fetch attempt duration: 60 seconds.

To correct this problem specify a valid configuration location using the -f/–config command-line options.

4

Ok, can you reboot your IS keeping the terracotta servers up and running.

5

Recycled both cluster IS’s without shutting down Terracotta and get the same stack trace

6

I have this working on another sandbox cluster and tried telneting to that server at port 9510 and it was successful. Then I tried doing the same for the servers giving me issues and I get Connection refused. These are brand new Linux VMs, so just wondering if something could be missing when they built them causing the IS to not be able to communicate with Terracotta? Or am I way off on this?

7

If you are unable to telnet on new linux vm’s there might be some network block or port needs to be opened (meanwhile include the network admin team also)

Also check and confirm the hotspot JDK version on your OS.

And also restarting the terracotta server might help you.

Let me know if you resolve it.

8

What is your TS Bigmemory verison?

Please also check the tc-config.xml and make sure the ip assocaiated to the 9510 port is open on the network/firewall and alos check the logs.

folder_home/bigmemory-max-4.1.5/server-logs

HTH,
RMG

9

Thanks for all your replies, what happened was that we had tar’d up an existing cluster IS that was working and laid it down on a new VM, modified everything for the new VM and then started up Terracotta and then the IS’s. Discovering the hard way that server-data directory contents in the bin directory for Terracotta are built for the IS at startup time the first time and it obviously contains pertinent data for that IS. Once I deleted that directory and started things back up the clustering worked.

thanks again!

1 Like
10

Thanks for you sharing your resolution. It is really helpful, good day.