[BUGS] BUG #3969: pg_ctl cannot detect server startup
The following bug has been logged online: Bug reference: 3969 Logged by: ITAGAKI Takahiro Email address: [EMAIL PROTECTED] PostgreSQL version: 8.2.6, 8.3.0 Operating system: CentOS 5, Windows XP Description:pg_ctl cannot detect server startup Details: pg_ctl -w -o "-p 5432 -c max_connections=100" start waits for server startup forever and timeout, but server has successfully started actually. The following work as expected. * pg_ctl -w -o "-p 5432" start * pg_ctl -w -o "-c max_connections=100" start * pg_ctl -w -o "-c max_connections=100 -p 5432" start Are there any limitation in ordering of -o arguments? ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [BUGS] BUG #3964: Parsing error in Stack Builder with LATIN1 client encoding
On Feb 18, 2008 4:53 PM, <[EMAIL PROTECTED]> wrote: > Hi there, > > > Betreff: Re: [BUGS] BUG #3964: Parsing error in Stack Builder with LATIN1 > > client encoding > > > > On Fri, Feb 15, 2008 at 9:41 AM, Jan-Peter Seifert > > <[EMAIL PROTECTED]> wrote: > > > > > > The following bug has been logged online: > > > > > > Bug reference: 3964 > > > Logged by: Jan-Peter Seifert > > > Email address: [EMAIL PROTECTED] > > > PostgreSQL version: 8.3.0 > > > Operating system: Windows xp sp2 German > > > Description:Parsing error in Stack Builder with LATIN1 client > > > encoding > > > Details: > > > > > > When choosing LATIN1 instead of the default CP1252 for the client > > encoding I > > > get this error in PostgreSQL & EnterpriseDB Stack Builder: > > > > > > 13:42:54: Fehler beim Lesen des XML: 'syntax error' in Zeile 1 > > > 13:42:54: Konnte die Anwendungsliste nicht parsen: > > > http://www.postgresql.org/applications.xml > > > > > > It doesn't matter whether the server encoding is LATIN1 or not. > > > > What exactly are you doing? Stackbuilder never talks to the database > > server, and has no encoding options at all. > > Well, if I choose LATIN1 as the encoding for the client in the setup / > cluster initialization the parser in the Stack Builder exits with the error > mentioned above. If I choose the default WIN1252 the Stack Builder works > correctly. I did make installs on copies of a German Windows xp sp2 VMWare > image. However, now it seems to be working after it has once been installed > correctly. Strange. I honestly don't see how the client encoding option could affect StackBuilder - all it does is change the command line options that are written to the start menu shortcut for psql, and sets PGCLIENTENCODING for the command prompt shortcut. StackBuilder itself doesn't know anything about PostgreSQL at all - it's an entirely standalone application!! FWIW, I tested this anyway, and it works fine here (except that the server complains that conversion between win1252 and latin1 isn't supported when I try to run psql). -- Dave Page EnterpriseDB UK: http://www.enterprisedb.com The Oracle-compatible database company ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
[BUGS] BUG #3970: ODBC Unknown Sizes Bug?
The following bug has been logged online: Bug reference: 3970 Logged by: Ricardo David Email address: [EMAIL PROTECTED] PostgreSQL version: 8.3 Operating system: Microsoft Windows XP SP2 (32bits) Description:ODBC Unknown Sizes Bug? Details: Until version psqlodbc-08_01_0200 the driver could work with Borland (Code Gear) products like Delphi. >From version psqlodbc-08_02_0100 up to psqlodbc-08_03_0100, when we mark: * Unknown Sizes = Don't Know; * Data Type Option = Text as LongVarChar, Unknowns as LongVarChar; all TEXT fields don't get recognized by the application as MEMO as they do before. We are still using the psqlodbc-08_01_0200 version with no problem. Thanks. ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [BUGS] BUG #3969: pg_ctl cannot detect server startup
"ITAGAKI Takahiro" <[EMAIL PROTECTED]> wrote: > Bug reference: 3969 > Description:pg_ctl cannot detect server startup > Details: > pg_ctl -w -o "-p 5432 -c max_connections=100" start > waits for server startup forever and timeout, > but server has successfully started actually. I found this bug comes from the definition of WHITESPACE characters in pg_ctl.c. WHITESPACE is defined as folows: #define WHITESPACE "\f\n\r\t\v" In fact, WHITESPACE does not contain whilespace (0x20) :-( I attach a patch to fix it. BTW, I also found similar definitions in some places. (Please grep with "\t\n\r".) They are a bit different from each other. For example, whitespaces is defined as " \t\n\r" in tzparser.c. Is it ok in the inconsistency? Or, should we always use " \f\n\r\t\v" ? Index: src/bin/pg_ctl/pg_ctl.c === --- src/bin/pg_ctl/pg_ctl.c (HEAD) +++ src/bin/pg_ctl/pg_ctl.c (working copy) @@ -49,7 +49,7 @@ typedef long pgpid_t; -#define WHITESPACE "\f\n\r\t\v" /* as defined by isspace() */ +#define WHITESPACE " \f\n\r\t\v" /* as defined by isspace() */ /* postgres version ident string */ #define PM_VERSIONSTR "postgres (PostgreSQL) " PG_VERSION "\n" Regards, --- ITAGAKI Takahiro NTT Open Source Software Center WHITESPACE.patch Description: Binary data ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
