Horton

Thread

  1. CLUSTER bug

    Darren Steven <dsteven@nettas.telstra.com.au> — 2000-05-11T00:51:47Z

    Sorry for a downer on an excellent piece of software.

--
Darren Steven
Applications Specialist
Networking Tasmania
Telstra Australia
Ph.1800 813 302

  2. Re: CLUSTER bug

    Bruce Momjian <pgman@candle.pha.pa.us> — 2000-05-11T01:01:25Z

    We have gotten sporatic reports of problems with CLUSTER.  Perhaps this
will prompt someone to find the problem and FIX IT FOR 7.0.1.  :-)


> Sorry for a downer on an excellent piece of software.
> 
> --
> Darren Steven
> Applications Specialist
> Networking Tasmania
> Telstra Australia
> Ph.1800 813 302
> 
> 

> If PostgreSQL failed to compile on your computer or you found a bug that
> is likely to be specific to one platform then please fill out this form
> and e-mail it to pgsql-ports@postgresql.org.
> 
> To report any other bug, fill out the form below and e-mail it to
> pgsql-bugs@postgresql.org.
> 
> If you not only found the problem but solved it and generated a patch
> then e-mail it to pgsql-patches@postgresql.org instead.  Please use the
> command "diff -c" to generate the patch.
> 
> You may also enter a bug report at http://www.postgresql.org/ instead of
> e-mail-ing this form.
> 
> ============================================================================
>                         POSTGRESQL BUG REPORT TEMPLATE
> ============================================================================
> 
> 
> Your name		:	Darren Steven
> Your email address	:	dsteven@nettas.telstra.com.au
> 
> 
> System Configuration
> ---------------------
>   Architecture (example: Intel Pentium)  	: Intel PIII
> 
>   Operating System (example: Linux 2.0.26 ELF) 	: Linux 2.2.14
> 
>   PostgreSQL version (example: PostgreSQL-6.5.1):   PostgreSQL-7.0
> 
>   Compiler used (example:  gcc 2.8.0)		: egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
> 
> 
> Please enter a FULL description of your problem:
> ------------------------------------------------
> 
> CLUSTER fails and appears to corrupt database. I don't know if this happened on other releases, found it by mistake while playing with 7.0
> 
> PS Great DB by the way.
> 
> 
> 
> Please describe a way to repeat the problem.   Please try to provide a
> concise reproducible example, if at all possible: 
> ----------------------------------------------------------------------
> 
> Create a database, create a table. insert some data. create index on one of fields. CLUSTER indexname ON tablename;
> 
> With a large number of rows (so I can see it happeneing), I see the table being copied to a temp, and the renamed back to its original name. the problem is with the index :
> 
> dws=# create index testind on accesslog (alclient);
> CREATE
> dws=# cluster testind on accesslog;
> ERROR:  temp_674e00 is an index relation 
> 
> 
> temp_674e00 appears to be a copy of the index file.
> 
> psql now tells me the relation accesslog no longer exists
> 
> the same error also occurs on small tables.
> 
> If you know how this problem might be fixed, list the solution below:
> ---------------------------------------------------------------------
> 
> Sorry. I've looked at cluster.c and if I fix it I'll post it.


-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

  3. Re: CLUSTER bug

    Bruce Momjian <pgman@candle.pha.pa.us> — 2000-05-11T01:10:40Z

    Here is a reproducable failure for CLUSTER.  Please, someone, find out
why this is happening so we can get it off our TODO list.

I now get:

	test=> cluster i_te on prauflage;
	ERROR:  temp_4c92 is an index relation

---------------------------------------------------------------------------

It works. But why doesn't it work on my tables:

create table prauflage(
nr int4,
au int4,
pl text,
be text,
st int4,
vh bool,
ex int4,
primary key(nr,au));

prd=> create unique index i_te on prauflage(nr,au);
CREATE
prd=> cluster i_te on prauflage;
ERROR:  Cannot create unique index. Table contains non-unique values           

-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

  4. Re: CLUSTER bug

    Tom Lane <tgl@sss.pgh.pa.us> — 2000-05-11T04:01:24Z

    Bruce Momjian <pgman@candle.pha.pa.us> writes:
> We have gotten sporatic reports of problems with CLUSTER.  Perhaps this
> will prompt someone to find the problem and FIX IT FOR 7.0.1.  :-)

The "sporadic" aspect probably had something to do with referencing
already-freed storage, which looks like a bug that'd been there awhile.
However, CLUSTER was definitely broken completely by changes elsewhere
in 7.0.  Anyone feel like adding a regression test for it?

			regards, tom lane

  5. Re: CLUSTER bug

    Bruce Momjian <pgman@candle.pha.pa.us> — 2000-06-13T00:21:26Z

    This was fixed in 7.0.1.

> Sorry for a downer on an excellent piece of software.
> 
> --
> Darren Steven
> Applications Specialist
> Networking Tasmania
> Telstra Australia
> Ph.1800 813 302
> 
> 

> If PostgreSQL failed to compile on your computer or you found a bug that
> is likely to be specific to one platform then please fill out this form
> and e-mail it to pgsql-ports@postgresql.org.
> 
> To report any other bug, fill out the form below and e-mail it to
> pgsql-bugs@postgresql.org.
> 
> If you not only found the problem but solved it and generated a patch
> then e-mail it to pgsql-patches@postgresql.org instead.  Please use the
> command "diff -c" to generate the patch.
> 
> You may also enter a bug report at http://www.postgresql.org/ instead of
> e-mail-ing this form.
> 
> ============================================================================
>                         POSTGRESQL BUG REPORT TEMPLATE
> ============================================================================
> 
> 
> Your name		:	Darren Steven
> Your email address	:	dsteven@nettas.telstra.com.au
> 
> 
> System Configuration
> ---------------------
>   Architecture (example: Intel Pentium)  	: Intel PIII
> 
>   Operating System (example: Linux 2.0.26 ELF) 	: Linux 2.2.14
> 
>   PostgreSQL version (example: PostgreSQL-6.5.1):   PostgreSQL-7.0
> 
>   Compiler used (example:  gcc 2.8.0)		: egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
> 
> 
> Please enter a FULL description of your problem:
> ------------------------------------------------
> 
> CLUSTER fails and appears to corrupt database. I don't know if this happened on other releases, found it by mistake while playing with 7.0
> 
> PS Great DB by the way.
> 
> 
> 
> Please describe a way to repeat the problem.   Please try to provide a
> concise reproducible example, if at all possible: 
> ----------------------------------------------------------------------
> 
> Create a database, create a table. insert some data. create index on one of fields. CLUSTER indexname ON tablename;
> 
> With a large number of rows (so I can see it happeneing), I see the table being copied to a temp, and the renamed back to its original name. the problem is with the index :
> 
> dws=# create index testind on accesslog (alclient);
> CREATE
> dws=# cluster testind on accesslog;
> ERROR:  temp_674e00 is an index relation 
> 
> 
> temp_674e00 appears to be a copy of the index file.
> 
> psql now tells me the relation accesslog no longer exists
> 
> the same error also occurs on small tables.
> 
> If you know how this problem might be fixed, list the solution below:
> ---------------------------------------------------------------------
> 
> Sorry. I've looked at cluster.c and if I fix it I'll post it.


-- 
  Bruce Momjian                        |  http://www.op.net/~candle
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026