Thread: Fix for log_min_error_messages
Due to my error, log_min_error messages went into 7.3.X with the wrong default, and the wrong listed options. This fixes the bug in CVS current and 7.3.X. Patch attached. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073 Index: src/backend/utils/misc/guc.c =================================================================== RCS file: /cvsroot/pgsql-server/src/backend/utils/misc/guc.c,v retrieving revision 1.111 diff -c -c -r1.111 guc.c *** src/backend/utils/misc/guc.c 25 Jan 2003 23:10:27 -0000 1.111 --- src/backend/utils/misc/guc.c 27 Jan 2003 23:39:15 -0000 *************** *** 101,107 **** bool Password_encryption = true; ! int log_min_error_statement = ERROR; char *log_min_error_statement_str = NULL; const char log_min_error_statement_str_default[] = "panic"; --- 101,107 ---- bool Password_encryption = true; ! int log_min_error_statement = PANIC; char *log_min_error_statement_str = NULL; const char log_min_error_statement_str_default[] = "panic"; Index: src/backend/utils/misc/postgresql.conf.sample =================================================================== RCS file: /cvsroot/pgsql-server/src/backend/utils/misc/postgresql.conf.sample,v retrieving revision 1.66 diff -c -c -r1.66 postgresql.conf.sample *** src/backend/utils/misc/postgresql.conf.sample 25 Jan 2003 23:10:29 -0000 1.66 --- src/backend/utils/misc/postgresql.conf.sample 27 Jan 2003 23:39:15 -0000 *************** *** 134,140 **** #log_min_error_statement = error # Values in order of increasing severity: # debug5, debug4, debug3, debug2, debug1, ! # info, notice, warning, error #debug_print_parse = false #debug_print_rewritten = false --- 134,140 ---- #log_min_error_statement = error # Values in order of increasing severity: # debug5, debug4, debug3, debug2, debug1, ! # info, notice, warning, error, panic(off) #debug_print_parse = false #debug_print_rewritten = false
Bruce Momjian writes: > Due to my error, log_min_error messages went into 7.3.X with the wrong > default, and the wrong listed options. > > This fixes the bug in CVS current and 7.3.X. Patch attached. Why does "panic" mean "off"? If you set it to "panic" don't you still get the "panic" messages? I'm sure the implementation makes sense, but the way it's documented seems confusing. -- Peter Eisentraut peter_e@gmx.net
Peter Eisentraut wrote: > Bruce Momjian writes: > > > Due to my error, log_min_error messages went into 7.3.X with the wrong > > default, and the wrong listed options. > > > > This fixes the bug in CVS current and 7.3.X. Patch attached. > > Why does "panic" mean "off"? If you set it to "panic" don't you still get > the "panic" messages? I'm sure the implementation makes sense, but the > way it's documented seems confusing. Yep, it is really weird. Gavin didn't want to have separate routines to validate the various flags, so all the "level" messages can take all the possible values, and there is no 'off' so "panic" seems the best. only meaningful ones. I now realize panic isn't really off, but I don't expect panic to happen too often. :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania19073
On Tue, 28 Jan 2003, Bruce Momjian wrote: > Peter Eisentraut wrote: > > Bruce Momjian writes: > > > > > Due to my error, log_min_error messages went into 7.3.X with the wrong > > > default, and the wrong listed options. > > > > > > This fixes the bug in CVS current and 7.3.X. Patch attached. > > > > Why does "panic" mean "off"? If you set it to "panic" don't you still get > > the "panic" messages? I'm sure the implementation makes sense, but the > > way it's documented seems confusing. > > Yep, it is really weird. Gavin didn't want to have separate routines to > validate the various flags, so all the "level" messages can take all the > possible values, and there is no 'off' so "panic" seems the best. > only meaningful ones. I offered, from memory, three different possible solutions. setting it to PANIC seemed the most popular. Gavin
Bruce Momjian writes: > I now realize panic isn't really off, but I don't expect panic to happen > too often. :-) Just add a level past panic that actually says "off" and really is off. -- Peter Eisentraut peter_e@gmx.net
Peter Eisentraut <peter_e@gmx.net> writes:
> Bruce Momjian writes:
>> I now realize panic isn't really off, but I don't expect panic to happen
>> too often. :-)
> Just add a level past panic that actually says "off" and really is off.
Would anyone actually use it? *Should* anyone actually use it?
I cannot imagine a situation where you don't want to see panic messages
(besides, there's always the option of routing the log to /dev/null...)
regards, tom lane