PostgreSQL Connection Error

[dujas]

Hi All,

I have a Jira Service Management cluster running at 8.20.2 and the infra is like:

1. Two virtual machines on Azure are serving the application, aka jira service management

2. One virtual machine on Azure is serving PostgreSQL 12.9

3. NFS is offered by Azure Storage Account

The database was switched to PostgreSQL from Microsoft SQL Server, which was running smoothly. Once I switched it to PostgreSQL, error like "Cluster Index Replication" complained out but it will be ceased out on its own. However, one of the 2 nodes will stop serving owning to error "FATAL: remaining connection slots are reserved fro non-replication superuser connections.".

I searched the related error and most of the suggestion is turn up the max_connections in PostgreSQL (one of the blogs I followed is https://confluence.atlassian.com/jirakb/jira-encounters-postgres-error-fatal-remaining-connection-slots-are-reserved-for-non-replication-superuser-connections-1063178451.html), I did turn it up from 100 to 120, but issue kept still. When it was at 100, the idle session is 95 and now it is up to 110, I don't think this would be a good workaround.

As a comparison, I have another identical environment running locally, everything is fine there.

Your suggestions would be a great help.

Thanks.

Jason

[Fabian Lim]

Hi @dujas

I recommend you open a ticket with atlassian support and provide the support zip files for analysis.

Regards