pyFTSIncrementalIndexer Automatically Stops - Work Object Search Issue

General
, , ,
1

When I restart Queue Processor “pyFTSIncrementalIndexer” manually from Admin Studio, it automatically stops after sometime. On further investigation we found that the Data flow pyFTSIncrementalIndexer is stopped and results in Work Object Search Issue for newly created objects.

Please let me know, if anyone has faced similar issue and found suitable solution for the same.

Environment : Pega Platform 8.4.5

2

HI @ShivRanjan_PCLSA8,

In general, below are some of the DSS used for on/off search option in Pega. Can you check if the following are enabled or disabled. These DSS can over write Search Functionality.

Pega-SearchEngine • indexing/distributed/search_enabled = true (To enalbe search)

Pega-RULES • indexing/enabled = true (To enable Indexing)

Pega-SearchEngine • indexing/distributed/index_enabled = true (To enable indexing thru Elastic Search)

Hope this helps. Thanks.

3

@ShivRanjan_PCLSA8 Hi, we had a similar issue from time to time and the only solution was to completely remove index files, restart utility nodes and start indexing fresh. Pega Cloud support is usually helpful with that if you are hosted in the cloud.

4

@NishanthVanka Hi Nishanth, these DSS configurations are already enabled.

5

@Oleg.Shepelev

Hi Oleg, thanks for your response.

We are not on cloud.

But we have performed cleanup with following steps below:

  1. Stop all the nodes.
  2. Clean the indexing temp directory.
  3. Start the nodes
  4. Perform re-indexing

But the issue still persists.

6

@ShivRanjan_PCLSA8

Hi! I have a bit longer list of steps captured from the times when we had such an issue, I would recommend to try that as the restart with the search folder cleanup didn’t help:

  1. Shutdown all nodes of all types in a cluster Completely

Do it in following order: AppTier >> DecisionHub >> Dataflow >> VBD

ADM >> Stream >> DDS >> Util // Kindly make sure the Shutdown is Graceful //

  1. Delete the pega Search Index Directory in all the Search nodes . It should reside in location pointed by index.directory system variable or location found in

pr_sys_statusnodes table in column pyIndexDirectory for each node

  1. Truncate the below Table

pr_sys_statusnodes

  1. Once the Above is Complete , follow the below Restart procedure

  2. First only One Util Node ( Search node) Should be Started out of the 3 Util Node

( Verify from the Startup Logs that there are No errors after Startup )

  1. Bring Only the Other App( Web) Tiers along with Other Stream Nodes * DO NOT Start the other two Util Nodes Until the Indexing is completed in the first Util node

  2. Login to Prpc and Go to the Search landing Page .The Re-Indexing Operation Should be Automatically in progress.

7

@ShivRanjan_PCLSA8 I think the last reply of Oleg should indeed cover all the steps for on premise. As a reference: I received the following steps from Pega support;

1.Shutdown all util nodes

2.Delete search index directories

3.See if pr_sys_statusnodes column pyindexdirectory should have no value

4.Do a rolling restart of util nodes.