Inconsistent Portal Reindex Counts in HA ArcGIS Enterprise Setup

RahulRaj2025

Hi,

I have a High Availability (HA) ArcGIS Enterprise setup (11.4v), and I'm encountering an issue with portal reindexing.

When I check the reindex status using the private URL of one portal machine, the store and index counts are consistent and correct. However, when I check the same from the other portal machine’s private URL, the index count shows as zero, even though the store count is accurate.

The environment is otherwise healthy, and both portal machines are part of the same HA setup. This inconsistency is concerning, especially since it could affect search and content discovery.

Has anyone experienced a similar issue?

What could be the possible reasons for this discrepancy?
Could it be related to synchronization issues between the portal machines or something else?
What steps can I take to resolve or further diagnose this?

Any insights or suggestions would be greatly appreciated.

Thanks in advance!

VenkataKondepati

Hi,

This issue typically points to index synchronization failure in your HA Portal setup. In ArcGIS Enterprise 11.4, only one portal machine is the index writer—the other is read-only and syncs from it. If one shows correct counts and the other shows zero, it’s likely the sync has failed.

Quick Checks:

Check index roles
Go to:
https://<portal-url>:7443/arcgis/portaladmin/system/indexer/machines
Confirm the writer vs. standby roles and both have STARTED status.
Force replication
On the secondary:
https://<portal-url>:7443/arcgis/portaladmin/system/indexer/replicateIndex
Review logs
Look under portal-index.log for sync errors on the standby node.
Last resort: Full reindex
Run from the primary node if needed.

Also ensure internal DNS, machine names, and SSL certs are aligned across nodes.

Let me know if you want a script or deeper dive.

Regards,
Venkat

RahulRaj2025

Hi,

Thanks for the explanation @VenkataKondepati .
I will look into this.

Regards,
Rahul