RE: Questionable coding in proc.c & lock.c

> -----Original Message-----
> From: pgsql-hackers-owner@hub.org [mailto:pgsql-hackers-owner@hub.org]On
> Behalf Of Tom Lane
> 
> I spent some time looking around for possible causes of the recently
> reported deadlock conditions.  I couldn't find any smoking gun, but
> I found a couple of things that look fishy:
> 

Oops,I've forgotten another freezing issue reported by Alfred Perlstein.
We know the cause(db access in abort transaction state) of it.
Seems xact.c has been pretty changed after I examined it 2 months
ago. Have you already fixed it ? If not,I would examine it again and
fix the bug. OK ?

Here's a reproducible example.

Session-1# begin;BEGIN=# lock t;LOCK TABLE

Session-2=# begin;BEGIN=# lock t; [blocked] ^CCancel request sentERROR:  Query cancel requested while waiting
lockreindex=#select * from t;[blocked]
 

Session-1=# commit;COMMIT

Session-2ERROR:  LockRelation: LockAcquire failed=# abort;ROLLBACK=# lock t;[blocked]

Regards.

Hiroshi Inoue