> From: Tom Lane [mailto:[EMAIL PROTECTED] 
> "George Pavlov" <[EMAIL PROTECTED]> writes:
> > On 5/29/2007 10:19 AM, Ed L. wrote:
> >> FWIW, I've also been seeing this sort of query log corruption for
> >> as long as I can remember, 7.1 through 8.2, HPUX (parisc, ia64),
> >> Linux on intel, amd...
> 
> What *exactly* is the logging setup you guys use, and have you tried
> alternatives?

Have not tried many alternatives as I am using Pgfouine and it is fairly
rigid re. the log format it expects. As I had originally posted my
logging setup is:

> the relevant config settings are (show all | grep ' log_' output):
> 
>  log_connections                 | on                  
>  log_destination                 | stderr              
>  log_directory                   | pg_log              
>  log_disconnections              | on                  
>  log_duration                    | on                  
>  log_error_verbosity             | default             
>  log_executor_stats              | off                 
>  log_filename                    | postgresql-%a-%H.log
>  log_hostname                    | off                 
>  log_line_prefix                 | %t [%p]: [%l-1]     
>  log_min_duration_statement      | -1                  
>  log_min_error_statement         | panic               
>  log_min_messages                | notice              
>  log_parser_stats                | off                 
>  log_planner_stats               | off                 
>  log_rotation_age                | 60                  
>  log_rotation_size               | 0                   
>  log_statement                   | all                 
>  log_statement_stats             | off                 
>  log_truncate_on_rotation        | on                  

Also redirect_stderr = on. Let me know if there are any other config
settings that might be important.

I am currently on: "PostgreSQL 8.1.8 on x86_64-redhat-linux-gnu,
compiled by GCC gcc (GCC) 3.4.6 20060404 (Red Hat 3.4.6-3)", but this
has happened with other 8.1.* versions and Ed L. indicates with other
major versions.

This happens even with very short messages (definitely less than 4096
bytes) -- I had pasted some examples originally; it does not seem to be
related to message/query length.

George

---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
       choose an index scan if your joining column's datatypes do not
       match

Reply via email to