Litao Wu <litaowu(at)yahoo(dot)com> writes:
> Since there are multiple databases and
> there are 170 postgres processes this morning,
> 60 of them are access the problem database,
> and 57 of 60 are non-idle.
> We only need to gdb those 57 processes, or
> we need gdb 60 or 170?
Potentially the deadlock could be anywhere :-(. You should definitely
not assume it must be one of the processes connected to the problem
database, because the buffer pool is cluster-wide.
Might be worth setting up a shell script to help.
regards, tom lane
In response to
pgsql-performance by date
|Next:||From: Joseph Shraibman||Date: 2004-07-01 03:14:58|
|Subject: planner and worst case scenario|
|Previous:||From: Litao Wu||Date: 2004-06-30 18:45:18|
|Subject: Re: reindex and copy - deadlock? |