Secondary Index Alarms

Secondary index alarms indicate issues that HPE Ezmeral Data Fabric Database might encounter while updating secondary indexes. It is important that you understand what the alarms indicate, and how to resolve the issue causing them.

Secondary Index Update Lag High

Logged As



This alarm displays the paths and names of the source tables for which replication lag is high. High lag times might be caused by these conditions:

  • High load on the source or replica
  • Low network bandwidth between the source and replicas
  • Miscellaneous error conditions that prevent replication from proceeding
  • Replication explicitly paused on the source cluster
If you have configured the threshold for this alarm too low, you can increase the theshhold. You configure this lag at the volume level by specifying the dbindexlagsecalarmthresh parameter. See volume modify for more information.

Secondary Index Update Error

Logged As

This alarm occurs if the JSON table regions cannot connect to any of the internal gateways used to update fields in a secondary index. This might be caused by these conditions:

  • The replication gateway failed.
  • You have configured too few replication gateway instances.
If the gateway failed, restart it. Otherwise, add more gateways. See Managing Gateways for further information.

Secondary Index Encoding Error

Logged As

This alarm occurs when any of the following encoding errors occur during index updates:

  • The indexed rowkey size is too big (> 32 Kb).
  • The indexed field contains a CAST function, and a failure occurs evaluating the CAST function.

You must either correct your underlying data or redefine the index to avoid missing rows. See Troubleshooting Secondary Index Encoding Errors for details about how to identify these errors and possible corrective actions.

You also must manually clear this alarm, even if you drop the index or the table.