Portal Indexing failure -10.9

[Thomas_Puthusserry]

While trying an item index the following error comes: any suggestions where I need to fix this?

A bit of context: This is a HA deployment (10.9) with 2 Portals (Pri/Sec), GIS Server, Data store Primary/standby, Tilecache and fileshare to store portal/server contents.

The Sec portal was not in sync with Pri for a while and then started. 

Then the Sec portal was unregistered and joined to pri site. 

The User and Groups indexing worked, but failed with contents.

Indexer response. NOTE: Picked up JDK_JAVA_OPTIONS: --add-opens=java.base/java.lang=ALL-UNNAMED --add-opens=java.base/java.io=ALL-UNNAMED --add-opens=java.rmi/sun.rmi.transport=ALL-UNNAMED Feb 01, 2024 2:29:31 PM com.esri.arcgis.portal.Indexer <init> INFO: Constructing new Indexer... Feb 01, 2024 2:29:31 PM com.esri.arcgis.portal.Indexer <init> INFO: Number of threads: 2 Feb 01, 2024 2:29:32 PM com.esri.arcgis.portal.ConfigPropertiesPopulator <init> INFO: Loading the configuration store connection. Feb 01, 2024 2:29:32 PM com.esri.arcgis.portal.ConfigPropertiesPopulator processProperties INFO: Reading GW configuration properties. Feb 01, 2024 2:29:32 PM com.esri.arcgis.portal.ConfigPropertiesPopulator processProperties INFO: Reading Elastic Search configuration properties. Feb 01, 2024 2:29:32 PM com.esri.arcgis.portal.ConfigPropertiesPopulator processProperties INFO: Injecting properties into Spring framework. ERROR StatusLogger Log4j2 could not find a logging implementation. Please add log4j-core to the classpath. Using SimpleLogger to log to the console... SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder". SLF4J: Defaulting to no-operation (NOP) logger implementation SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details. 1 [WARN] RestClient: - request [GET http://localhost:7220/ago-items/_mapping?master_timeout=30s&include_type_name=true&ignore_unavailabl...] returned 1 warnings: [299 Elasticsearch-7.9.3-c4138e51121ef06a6404866cddc601906fe5c868 "[types removal] Using include_type_name in get mapping requests is deprecated. The parameter will be removed in the next major version."] 156 [WARN] RestClient: - request [GET http://localhost:7220/ago-groups/_mapping?master_timeout=30s&include_type_name=true&ignore_unavailab...] returned 1 warnings: [299 Elasticsearch-7.9.3-c4138e51121ef06a6404866cddc601906fe5c868 "[types removal] Using include_type_name in get mapping requests is deprecated. The parameter will be removed in the next major version."] 228 [WARN] RestClient: - request [GET http://localhost:7220/ago-users/_mapping?master_timeout=30s&include_type_name=true&ignore_unavailabl...] returned 1 warnings: [299 Elasticsearch-7.9.3-c4138e51121ef06a6404866cddc601906fe5c868 "[types removal] Using include_type_name in get mapping requests is deprecated. The parameter will be removed in the next major version."] Status:{"users":{"databaseCount":130,"indexCount":130,"name":"users"},"groups":{"databaseCount":552,"indexCount":552},"search":{"databaseCount":22028,"indexCount":0}} Time for Indexer application in milliseconds 21840

Thanks for suggestions

Thomas

[RanaDas · Accepted solution]

Hi @Thomas_Puthusserry 

Unfortunately Portal 10.9.x does not support manual rebuild of indexes. Since the secondary portal was disconnected for a log time, the json files for portal contents might not have updated. You can try the below workaround and see if it works. (do take machine level backups before proceeding)

• Unregister the secondary portal ( which was disconnected)
• Run reindex form the Primary portal (only remaining node)
• On successful reindexing join the second site to portal
• Check the index status
• Run reindex if required

Hope this helps.

[Thomas_Puthusserry]

Thanks very much  @RanaDas that worked.

[RanaDas]

@Thomas_Puthusserry  Glad it worked. You're most welcome.