ERROR Configuration from app=<appname> does not support reload: server.conf/[clustering]/master_uri

[justynap_ldz]

Hi Guys,

we keep getting following errors in Splunk 8.1.2:

ERROR ApplicationUpdater - Error reloading <appname>: handler for server (http_post /replication/configuration/whitelist-reload): Configuration from app=<appname> does not support reload: server.conf/[clustering]/master_uri

The app was created manually long time ago and defines basic configuration for our Search Head Cluster.
It is installed on the Master server that we use to deploy changes to our Search Head Cluster.

We hoped it is just a typo so we double checked server name in [clustering] master_uri but that is not the case.

The whole server.conf in <appname> is as follows: 

[license]
master_uri = https://<license_master_server>:8089

[clustering]
master_uri = https://<master_server>:8089
mode = searchhead
pass4SymmKey = xyz

[replication_port://8815]

[shclustering]
conf_deploy_fetch_url = https://<master_server>:8089
disabled = 0
pass4SymmKey = xyz
replication_factor = 2
shcluster_label = Splunk_AB
id = xyz

Do you have any idea what may is wrong here?

Your help will be much appreciated! 

Greetings!

[codebuilder]

The Master does not deploy apps nor changes to your search head cluster, only to the indexers.

If you want to push out something to the SHC you need to use a deployer:
https://docs.splunk.com/Documentation/Splunk/8.2.1/DistSearch/PropagateSHCconfigurationchanges

Any changes made to server.conf require cycling Splunk itself, or a rolling restart if part of a cluster.

[justynap_ldz]

Thank you @codebuilder ! 

[codebuilder]

Sure.

If the reply helped you, an upvote would be appreciated. 

[justynap_ldz]

Hi @codebuilder,
We have checked your suggestion again, but I am afraid it is not applicable in our case.

Our Master functions as SH cluster deployer, too. It has been working like that since years.
Strange, but although we have the same config in all 3 stages, in prod there are no errors, but in pre-prod and test the errors occur. 

The app=<appname> is created on the Master in /opt/splunk/etc/shcluster/apps/<appname> and this is where we have basic config for our SearchHead clusters. Also, whenever we push any shcluster bundle changes we run the command from the Master. 

Generally, SH clusters work, but we want to know the reason of the errors and get rid off them if possible

[codebuilder]

Generally speaking, it would be better practice not to use the master as a SHC deployer.

With that said, what command are you issuing that triggers the error you posted?

[justynap_ldz]

Hi there, 

any update on this? 

Regards,

Justyna