Connection reset by peer / broken pipe - Mailing list pgsql-general

Much the same as this person here:

http://archives.postgresql.org/pgsql-general/2008-02/msg00689.php

I am running Postgres 8.2.3 and am seeing the following error messages in my logs:

LOG:  SSL SYSCALL error: Connection reset by peer
LOG:  could not receive data from client: Connection reset by peer
LOG:  unexpected EOF on client connection
LOG:  could not send data to client: Broken pipe

The end result is my users are connecting on TCPIP and their connection is "hanging" and they aren't able to finish their database transaction.  Sometimes this will also cause the database server to be extremely slow. 

I am using a fair number of default values in my postgresql.conf file. 

Do you have any advice?  My configuration is below.  Thanks in advance--

Jeff Wigal


CONFIGURATION

#---------------------------------------------------------------------------
# CONNECTIONS AND AUTHENTICATION
#---------------------------------------------------------------------------

# - Connection Settings -

max_connections = 100

#superuser_reserved_connections = 2
#unix_socket_directory = ''
#unix_socket_group = ''
#unix_socket_permissions = 0777      
#bonjour_name = ''          

#authentication_timeout = 60      
ssl = on
#password_encryption = on
#db_user_namespace = off

# - TCP Keepalives -
# see 'man 7 tcp' for details

#tcp_keepalives_idle = 0       
#tcp_keepalives_interval = 0      
#tcp_keepalives_count = 0     


#---------------------------------------------------------------------------
# RESOURCE USAGE (except WAL)
#---------------------------------------------------------------------------

# - Memory -

shared_buffers = 1000         
#temp_buffers = 1000           
#max_prepared_transactions = 5      
#work_mem = 1024           
#maintenance_work_mem = 16384      
#max_stack_depth = 2048           

# - Free Space Map -

#max_fsm_pages = 20000           
#max_fsm_relations = 1000      

# - Kernel Resource Usage -

#max_files_per_process = 1000      
#preload_libraries = ''

# - Cost-Based Vacuum Delay -

#vacuum_cost_delay = 0           
#vacuum_cost_page_hit = 1      
#vacuum_cost_page_miss = 10      
#vacuum_cost_page_dirty = 20      
#vacuum_cost_limit = 200      

# - Background writer -

#bgwriter_delay = 200            # 10-10000 milliseconds between rounds
#bgwriter_lru_percent = 1.0        # 0-100% of LRU buffers scanned/round
#bgwriter_lru_maxpages = 5        # 0-1000 buffers max written/round
#bgwriter_all_percent = 0.333        # 0-100% of all buffers scanned/round
#bgwriter_all_maxpages = 5        # 0-1000 buffers max written/round


#---------------------------------------------------------------------------
# AUTOVACUUM PARAMETERS
#---------------------------------------------------------------------------

autovacuum = on                # enable autovacuum subprocess?
#autovacuum_naptime = 60        # time between autovacuum runs, in secs
#autovacuum_vacuum_threshold = 1000    # min # of tuple updates before
                   # vacuum
#autovacuum_analyze_threshold = 500    # min # of tuple updates before
                   # analyze
#autovacuum_vacuum_scale_factor = 0.4    # fraction of rel size before
                   # vacuum
#autovacuum_analyze_scale_factor = 0.2    # fraction of rel size before
                   # analyze
#autovacuum_vacuum_cost_delay = -1    # default vacuum cost delay for
                   # autovac, -1 means use
                   # vacuum_cost_delay
#autovacuum_vacuum_cost_limit = -1    # default vacuum cost limit for
                   # autovac, -1 means use
                   # vacuum_cost_limit

#---------------------------------------------------------------------------
# LOCK MANAGEMENT
#---------------------------------------------------------------------------

#deadlock_timeout = 1000        # in milliseconds
#max_locks_per_transaction = 64        # min 10
# note: each lock table slot uses ~220 bytes of shared memory, and there are
# max_locks_per_transaction * (max_connections + max_prepared_transactions)
# lock table slots.

VMSTAT -S

       442368  total memory
        51800  used memory
            0  active memory
            0  inactive memory
       390568  free memory
            0  buffer memory
            0  swap cache
            0  total swap
            0  used swap
            0  free swap
      8882893 non-nice user cpu ticks
        58510 nice user cpu ticks
       952202 system cpu ticks
   1078995945 idle cpu ticks
      8062131 IO-wait cpu ticks
            0 IRQ cpu ticks
            0 softirq cpu ticks
            0 pages paged in
            0 pages paged out
            0 pages swapped in
            0 pages swapped out
            0 interrupts
   2452189757 CPU context switches
   1197496622 boot time
     48285711 forks

FREE -M

             total       used       free     shared    buffers     cached
Mem:           432         50        381          0          0          0
-/+ buffers/cache:         50        381
Swap:            0          0          0