Re: PostgreSQL 8.0.3 limiting max_connections to 64 ?

First of all, thanks in advance for all the replies.

Joshua D. Drake wrote:

>
>>> max_connections, shared_buffers, shmmax were tweaked, but the server
>>> does not seems to respect the 500 max_connections...
>>> i *know* i'm missing something obvious, but, what could be wrong
>>> ?... i'm lost... any help would be most appreciated... please.
>>
>
> A completely stupid response but don't take it pseronally.
>
> Did you remove the # from in front of the max_connections parameter and
> do a COMPLETE restart?
>
no worries, i didn't take personally. ;)

---
Tom Lane wrote:

>Alexandre Barros <alexandre(at)vectorx(dot)com(dot)br> writes:
>
>
>>max_connections, shared_buffers, shmmax were tweaked, but the server
>>does not seems to respect the 500 max_connections...
>>
>>
>
>Er ... did you restart the postmaster after increasing those parameters?
>
> regards, tom lane
>
>
>
Er... at least twice, once because a power failure with a defective
no-break,
and the second time i restarted all services ( and before that, i kept
using -- over and over -- the "kill -HUP" on postmaster, and trusted the
"reloading parameters" message on the logfile... );

---
Sebastian Hennebrueder wrote:

>500 parallel connections are very very much.
>

i agree... but i needed to test how ( and why ) things were ( not )
working... and the "/sorry too many clients already/" message was
driving me crazy...

>You should verify if one
>application is not closing connections or if you can create an
>connection pool.
>
>

the connection pool ( was testing pgpool for that ) was a possibility,
but i need things at least "barely working" before... and an obscene
value on "max_connections" was my best try.

>Use select * from pg_stat_activity to see wheach connections are open
>from which client and if they are iddle or not. See the postgreSQL doc
>for more information on these queries.
>
>
>
i have lots of idle connections showing on a "ps|grep" from squirrel (
apache with persistent connections ) amavisd, courier-authlib all keep
idle connections for a long time, but that pg_stat query only shows
stuff like that:
/
postfix=# select * from pg_stat_activity ;
datid | datname | procpid | usesysid | usename |
current_query | query_start

-------+--------------+---------+----------+--------------+------------------------------+-------------
/ /17230 | postfix | 29852 | 100 | postfix |
<command string not enabled> |
(...)
// (47 rows)

/i'm yet not sure what this means, we'll be looking over postgresql docs...

---
Sven Willenberger wrote:

>Can you post the relevent portions of your postgresql.conf file? Do you
>see any error messsages when starting the database up (perhaps about
>shared memory or such)?
>
everything not commented out from the postgresql.conf:

/max_connections = 500
# note: increasing max_connections costs about 500 bytes of shared
# memory per connection slot, in addition to costs from
shared_buffers
# and max_locks_per_transaction.
superuser_reserved_connections = 2
shared_buffers = 1004 # min 16, at least max_connections*2, 8KB each

debug_print_parse = false
debug_print_rewritten = false
debug_print_plan = false
debug_pretty_print = false
log_statement = 'none'

log_parser_stats = false
log_planner_stats = false
log_executor_stats = false
log_statement_stats = false

statement_timeout = 60000 # 0 is disabled, in milliseconds

/
/proc/sys/kernel/shmmax == 33554432

and no error messages at all on the startup logfile...

>We also run a postfix mailserver (with maildrop,
>courier-imap/vauthchkpw, etc) against a postgresql database with
>max_connections set to 512 (FreeBSD 5.x machine). On the delivery end of
>things we pool the connections from the postfix server using proxymap(8)
>(which helped ease the connections load).
>
>Sven
>
>
>
aha ! 512 connections ! someone as "exaggerated" as me, i feel happy now !
;)

i will also study proxymap, thanks !

Alexandre Barros