I have several directories such as 1, 17229, 17230, 333909 under
base. but none of them is generated during crash time. Under thsoe
directories, there are a bunch of files with NNNN(number) filename.
Some of them are over size of 6M.
I guess i will set the max_stack_depth in postgres.conf to get the right
coredump?
Nope, no other add-on or modules. we are just using tables, indices and
triggers from slony.
Thanks!!
Michael Fuhr wrote:
On Wed, Oct 26, 2005 at 09:27:52AM +0100, Telin Lin wrote:
My postgresql keeps crashing by signal 11 after a number of concurrent
inserts + updates.
Did you get a core dump? If so, can you get a stack trace from it?
In released versions of PostgreSQL core dumps are usually found
under the $PGDATA/base/<database oid> directory unless your system
is configured to put them elsewhere. If you got no core dump then
you might need to adjust the coredumpsize resource limit that the
postmaster runs with.
You mentioned that you were using Slony. Do you have any other
add-ons? Custom data types, third-party modules, etc.?
---------------------------(end of broadcast)---------------------------
TIP 3: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faq
