Tom Lane wrote:
Kamil Srot <kamil.srot@nlogy.com> writes:
From above mentioned duplications of relatio names and what Tom wrote
recently (doesn't see like WA problem), it looks like the relation name
is/gets corrupted in some way and this corruption is internally taken
over to another instance of relation named the same but in another
database... but I know - it's too speculative.
The fact that the same thing happens concurrently in several independent
databases is to me the strongest indication that it's something the
client is doing. There isn't *any* mechanism within Postgres that would
somehow cause similarly-named tables in different DBs to go south at the
same time. I don't say that it couldn't be a PG bug, but I do say that
the client is doing something similar in each database to trigger it.
My advice is to turn on log_statement and acquire a trace (with
timestamps) of all the SQL commands being executed across the whole
system. Looking at what happens up to the instant the problem appears
should provide some clues.
OK, I did it... have good logging now, so if it'll happen again, I'll post relevant part of it to the list.
Thank you for your help!
--
Kamil