4

Due to an unforeseen hardware error, I had to take our secondary replicas down for maintenance. This was really due to a storage issue and, it was decided at a higher level, to purchase a new storage system for the secondary instance. This is several weeks out from delivery and installation at this point.

This was on a system that also uses FCI with 2 active instances.

I was able to successfully remove the secondary replicas from both Availability Groups and I thought I had done this correctly (they no longer show up in the UI or when I look at the AG Dashboard. However, one one of the AG's, all of my log files are growing. When I query the sys.databases and look at the log_reuse_wait_desc for that instance it says 'AVAILABILITY_REPLICA'

I'd rather not remove the AG as the listener is actively used and would require reconfiguration of numerous web.configs and other configuration entries. But, I also am slightly worried about the log file growth.

Why, on one AG does it not appear to be growing or have the log_reuse_wait_desc of AVAILABILITY_REPLICA but on the other it does?

András Váczi
  • 31,778
  • 13
  • 102
  • 151
Brent Bain
  • 41
  • 1
  • 2

2 Answers2

1

I think the question has a wrong assumption.

So is there a way to keep a single (primary) replica server active in the AG for the databases without removing them but keeping the log file from growing or being stuck in 'AVAILABILITY_REPLICA'?

...

I'd rather not remove the AG as the listener is actively used and would require reconfiguration of numerous web.configs and other configuration entries. But, I also am slightly worried about the log file growth.

If you have an AG running on a single node, you can remove the database from the AG completely AND still connect to the database via the listener (as long as the clustering part is online either forced or using Windows 2012+ with dynamic quorum).

For a normal connection without any application intent the listener just functions as a dumb alias for the server. It doesn't check something like whether the database you're connecting to is part of its AG or not.

Cody Konior
  • 3,538
  • 14
  • 23
0

Work around would be to remove the DB from AG and then clear the log and add it back to AG

Hareesh Gottipati
  • 1