| Lists: | pgsql-interfaces |
|---|
| From: | "Romanenko Mikhail" <mikhail(at)angg(dot)ru> |
|---|---|
| To: | <pgsql-interfaces(at)postgresql(dot)org> |
| Subject: | Error message: "my bits moved right off the end of the world!" |
| Date: | 2000-12-07 02:51:41 |
| Message-ID: | 006501c05ff8$a3697720006501c05ff8$a3697720$01011cac@angg.ru011cac@angg.ru |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-interfaces |
I'm developing application with Borland C++ Builder as a client and PostgreSQL as a server. Sometime (Almost every other time) the application failed to get loaded and I get an error message in psqlodbc_xxxxxxxxxxxx.log file:
[...]
conn=58919036, query='CREATE TABLE course_dl_tmp10183 (brig_id int2, well varchar(10), pad varchar(7), date_dl date, course_id int2, what varchar(4), course_what varchar(60), course_value time, PRIMARY KEY (course_id));'
NOTICE from backend during send_query: 'NOTICE: CREATE TABLE/PRIMARY KEY will create implicit index 'course_dl_tmp10183_pkey' for table 'course_dl_tmp10183''
ERROR from backend during send_query: 'FATAL 1: my bits moved right off the end of the world! Recreate index pg_attribute_relid_attnum_index.'
conn=58919036, query='COMMIT'
Could anybody tell me what does this error message "my bits moved right off the end of the world!" mean.
Client software:
MS Windows 98
Borland C++ Builder 5.01
BDE 5.1.1
ODBC 3.510.3711.0
PsqlODBC 6.50.0000
Server software
Linux (Dedian woody, kernel 2.2.17, libc6 2.2)
PostgreSQL 7.0.3
Thank you, Mikhail.
| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "Romanenko Mikhail" <mikhail(at)angg(dot)ru> |
| Cc: | pgsql-interfaces(at)postgresql(dot)org |
| Subject: | Re: Error message: "my bits moved right off the end of the world!" |
| Date: | 2000-12-12 17:55:17 |
| Message-ID: | 2293.976643717@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-interfaces |
"Romanenko Mikhail" <mikhail(at)angg(dot)ru> writes:
> Could anybody tell me what does this error message "my bits moved
> right off the end of the world!" mean.
It's an internal error in the btree index routines.
I would *love* to see an example that reproduces this error from a
standing start (ie, newly created database). I suspect there's a
corner-case bug somewhere in btree, but it's never been possible
to find it without a test case...
regards, tom lane