CommVault Content Indexing operations are failing after performing a search engine optimization

Last week, I downloaded a custom workflow which executes a Search Engine optimization from CommVault Cloud.

The Search Engine optimization workflow brings the Search Engine offline, performs some maintenance activities on the Apache Lucene SOLR engine. Once completed, the search engine is brought back online. Below you can find a screenshot of the workflow.
SOLR Content Indexing Failing - 1

When been executed, I noticed it stayed in the “run optimize” phase for about 25 minutes and the logs files remained remarkable empty. I decided to kill the job and check it on a later moment. (sorry I did not had the time to wait :) )

When the Content Indexing operations are executed as scheduled, they got in a waiting status with the following delay reason: “Error Code: [62:2397] Description: Cannot find Search Engine to Index data for this storage policy. Please add a new node or use other Search Engine to content index data“.

I checked the killed job and noticed the failed phases on the “attempts” tab:

  • Run Optimize (“killed”)
  • Enable Engine (“killed”)

It quickly became clear the Search Engine was put offline and is causing this error. So logically I checked the GUI, but I noticed it’s not possible to bring the Search Engine back online from within the GUI.

After investigating the workflow, I found the following SQL statement was executed in the Enable Engine phase:
“UPDATE DM2SearchServerCoreInfo
SET Enabled = 1
WHERE ServiceLink = ‘xpath:{/workflow/PopupInput_2/URL}'”

After opening a SQL Management Studio, I collected the path to my SOLR engine and it’s status by executing the following script:
“use CommServ;
select Servicelink,Enabled from DM2SearchServerCoreInfo”
SOLR Content Indexing Failing - 2

In the output I noticed, the enabled status was set to ‘0’ (disabled). I executed the following query to enable the engine again:
“UPDATE DM2SearchServerCoreInfo
SET Enabled = 1
WHERE ServiceLink = ‘http://MyCollectedLink:27000′”

Afterwards I restarted the services on the Content Indexing Search Node and relaunched a Content Indexing job. Now everything is running properly again:
SOLR Content Indexing Failing - 3

The performance of the Content Indexing jobs still dazzles me when we speak. The job completed after approximately 2 hours and indexed 1.13TB of new data. A blazing 175MB/sec!

Thanks for reading,

Ruben.

Leave a Reply