Question
Search Indexing Agent is not index few work items
Hi,
We are using Pega OOTB Case Manager portal search for work items. FTSIncrementalIndexer agent is up and running, but few work items are not indexed and so search is resulting no results. I could not find anything from the Pega logs. Could anyone help me to identify the rootcause of the issue.
Please note this is happening in Production.
Thanks in Advance
Regards,
Venugopal.
***Updated by Moderator: Vidyaranjan. Removed user added Ask the Expert tags .Apologies for confusion, shouldn't have been an end-user option***
**Moderation Team has archived post**
This post has been archived for educational purposes. Contents and links will no longer be updated. If you have the same/similar question, please write a new post.
Thank you Abhishek for your response.
I have run the query you have mentioned,but all the entries of same expected work class. The issue is very internmnent and very few work items are not indexed. We have already implemented the solution that is mentioned in the link. Even after very few work items are having this issue.
For the cases that have this issue, does a resave of the work item solve the problem?
Since the agent runs every one minute, do you only face this problem when the agent is yet to drain the queue?
Are there multiple nodes in your cluster? If yes, how many of them have been designated as the index host nodes on the search landing page? Are there times when one or more of those nodes are unavailable and work items updated in that duration are failing?
Are there any errors in the logs related to indexing either on the index host node (as configured on the search landing page) or other nodes?
Hi Rajiv,
Thanks for responding, Please find below details and let me know if you further details.
For the cases that have this issue, does a resave of the work item solve the problem? - [Venu] Yes, it did. As this is happened twice in Production, we are looking for a permenent solution.
Since the agent runs every one minute, do you only face this problem when the agent is yet to drain the queue?
Are there multiple nodes in your cluster? If yes, how many of them have been designated as the index host nodes on the search landing page? Are there times when one or more of those nodes are unavailable and work items updated in that duration are failing? - [Venu] we have 4 nodes and 2 nodes are configured for indexing. All nodes are up and running.
Are there any errors in the logs related to indexing either on the index host node (as configured on the search landing page) or other nodes? - [Venu] i could not find any failure index logs in any of the nodes.
To add one more point here, work item that is just created before and after of the work item are searchable. e.g W-23,W-24,W-25 W-23 and W-25 are searchable but W-24 is not. FTSIncrementalIndexer agent is up and running in all the nodes.
Regrds,
Venugopal.
Hi Venu,
Some more questions
- You mentioned that even though 3 work items were created, one of them got missed but the other 2 were searchable. Can we stop the FTSIncrementalIndexer agent and check the pr_sys_queue_ftsindex table to see if for all work items created, there are entries in this table? You can check the pxWorkKey and pxWorkClass columns to verify.
- When one of items is not available, does it become available after sometime or resave is the only option to make it available in search results?
- Please enable debug logging on com.pega.pegarules.search.internal.es package and share the logs whenever this occurs again.
-Rajiv
Thanks Rajiv.
i have enabled the debug logs in prod system. i will try the points 1 and 2 in non-prod and try to replicate this.
Hi,
Do you have any update for this issue?
We have the same issue in production environment. We can recreate search index by re-saving work object for a temporary treatment, but we need fundamental resolution.
Thanks,
Haruka
Hi,
Thank you for posting your query in the PSC. This looks like an inactive post and hence, we suggest you create a new post for your query. Click on the Write a Post button that’s available on the top right pane of this page. Once created, please reply back here with the URL of the new post.
You may also refer this discussion link as a reference in the new thread.
Hi Venugopal,
From Search Landing Page > Agent Information section, This section displays the status of the FTSIncrementalIndexer agent that incrementally update the search index files as changes are made to RULE-, DATA-, and/or WORK- instances. Display provides following information:
Agent name
The number of entries currently in the queue
The age of the oldest entry in the queue
There might be a case that some custom classes that are getting updated very frequently which are generating too many entries in the pr_sys_queue_ftsindexer table which can cause a mismatch between the indexes on the pr_sys_queue_ftsindexer table and the business logic code.
Please refer the below PDN link for more information :
https://pdn.pega.com/support-articles/ftsincrementalindexer-agent-queue-size-large
You can also run the query mentioned in above link to verify.
Hope this will help.
Thank you