Thread: a error about PostgreSQL7.3.4's Log setting
Hi everyone, My name is Ken working in Tokyo,Japan. Please forgive my poor English. We have been using PostgreSQL7.3.4 on a Map GIS engine since last year.To tune the system, I need to get the log from PostgreSQL7.3.4 and analyze it. Here comes the problem. When a long SQL is excuted. I got the error as follow error:SQL Execute error 0 = Too much data was received. when I removed the log setting from my postgresql.conf file the sql was excuted without any problem and ended normally. I appand my postgresql.conf and start shell as follow. postgresql.conf ---------------------------------------------------------------- # # Message display # server_min_messages = log # Values, in order of decreasing detail: # debug5, debug4, debug3, debug2, debug1, # info, notice, warning, error, log, fatal, # panic client_min_messages = log # Values, in order of decreasing detail: # debug5, debug4, debug3, debug2, debug1, # log, info, notice, warning, error silent_mode = false log_connections = false log_pid = true log_statement = true log_duration = true log_timestamp = true # # Syslog # syslog = 1 # range 0-2 syslog_facility = 'LOCAL0' syslog_ident = 'postgres' LC_MESSAGES = 'C' LC_MONETARY = 'C' LC_NUMERIC = 'C' LC_TIME = 'C' ---------------------------------------------------------------- start_pgsql.sh ---------------------------------------------------------------- pg_ctl start -D /data/pgsql -o "-i" -l /data/pgsql/postgresql.log ---------------------------------------------------------------- develop environment OS: solaris2.9 Apserver: Sun application Server 7 JDBC Driver: pg74.215.jdbc3.jar can any friend help me, thanks a lot
acemeicn <acemeijp@yahoo.co.jp> writes: > Here comes the problem. When a long SQL is excuted. I got > the error as follow > error:SQL Execute error 0 = Too much data was received. The best guess I have is that this is a failure in Solaris' syslog. There is certainly no such message anywhere in the Postgres backend code. Try changing your syslog setting to 0 (log only to stderr) and see if logging to a plain file works better. BTW, PG 7.3.4 is pretty darn old, you really ought to update. regards, tom lane
> The best guess I have is that this is a failure in > Solaris' syslog. > There is certainly no such message anywhere in the > Postgres backend > code. Try changing your syslog setting to 0 (log > only to stderr) > and see if logging to a plain file works better. > > BTW, PG 7.3.4 is pretty darn old, you really ought > to update. > Dear tom lane Thank you for your advice. I tried it, but nothing change. I tried excuting the long sql on the console and a DB tool called CSE(Common SQL Environment), it worked. but in a java program using JDBC connection it doesn't work. Ken
acemeicn <acemeijp@yahoo.co.jp> writes: > I tried excuting the long sql on the console and a DB tool > called CSE(Common SQL Environment), it worked. > but in a java program using JDBC connection it doesn't > work. Possibly the folks on pgsql-jdbc could help you with that. regards, tom lane
> Possibly the folks on pgsql-jdbc could help you with > that. > > regards, tom lane sincere thanks to you tom lane I checked all the shell programs in the system which running standalone from SUN application server. Finally find the answer. One of those shell programs uses an old JDBC driver(jdbc7.1-1.2.jar). anyway thank you again for your help.