Hello pg hackers!
Today we had to run a `REINDEX table CONCURRENTLY my_table;` in our production database due to considerable index
bloat.We used to deal with this problem in the past by using pg_repack but we stopped using it because our data
replicationtool doesn’t support “re creating” tables in the way that pg_repack does it (we are using Fivetran).
The reindex went fine in the primary database and in one of our standby. The other standby that we also operate for
somereason ended up in a state where all transactions were locked by the WAL process and the WAL process was not able
tomake any progress. In order to solve this issue we had to move traffic from the “bad” standby to the healthy one and
thenkill all transactions that were running in the “bad” standby. After that, replication was able to resume
successfully.
I’m just trying to understand what could have caused this issue. I was not able to identify any queries in the standby
thatwould be locking the WAL process. Any insight would be more than welcome!
We are running pg12 in our cluster and standbys are replicating using physical replication lots.
Best,
Martín
