2

As part of preliminary tests for running a scaled-out search infrastructure, I have done the following:

On one CM server (let's call it server1), I have configured the search host to be the master, by running the following:

configure-search-host.ps1 master
restart-service tcmsearchhost

On the other CM server (server2) I have configured a slave as follows:

configure-search-host.ps1 slave server1 domain\user password
start-service tcmsearchhost 

Visiting http://server1:8983/#/tridion/replication I can see that replication is enabled, but when I visit http://server2:8983/#/tridion/replication I can immediately see a bright red banner that says invalid_master

This doesn't seem a good portent for success. Anthropomorphism aside, why would my slave think it has an invalid_master, and what should I do to remedy this.

Tridion 2013 SP1 HR1

2
  • Hi Dominic, A friendly word of warning, We scaled out the search but soon had problems with the replication. It generated full index files for each replication on the slave, soon filling the disk and killing the server. Let me know how you go. Jun 11, 2015 at 10:32
  • @ChrisMills I will highly encourage you to post your question on SE. The outscaled search scenario (master-slave) is not a new concept and has been used for a long time with Solr (independent of Tridion) as well as with Tridion. FYI, we are using outscaled search with Tridion and have not faced the problem that you are facing Jun 17, 2015 at 14:37

1 Answer 1

3

It seems like you have not given the command correctly. The command on slave needs master server with port (the port was missing in your command)

configure-search-host.ps1 slave masterserver:8983 domain\user password

This could be a probable reason. Also, for communication across servers on the port, FW should be open

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.