Database locks due to Qlik replication task ?

[gkaur]

Since we have started using Qlik Replicate to read data from PostgreSql and writing to Snowflake, we have been experiencing lot of PG database locks. 

There are days with no issue but then suddenly database start running slow, locking and impacting live applications.

Also, observed that even though we have only 3 Tasks running against same PG source, but at a time there are alteast 20+ DB connections established by Qlik.  My guess was to have only 3 connections per slot but why 20+ ?

Please advice if anyone else is experiencing the same, should we be changing any configuration on Postgres/ Qlik task etc...?

[DesmondWOO]

Hi @gkaur ,

For CDC, the number of connection can be up to 6 connections for each task. Please find this article for details.

Regarding the PG database locks, could you send us the error message that Replicate reported?

Thanks,
Desmond

[john_wang]

Hello @gkaur ,

Besides @DesmondWOO comments, also please let us know if the PostgreSQL source database is a primary node, or standby node? Any helpful information about the locks are welcome, eg what's the locks type, how about if you execute query the locks by:

select * from pg_locks();

thanks,

John.