DeleteTransactionInProgress - Mailing list pgsql-general

From Vilson farias
Subject DeleteTransactionInProgress
Date
Msg-id 003701c2f48f$6f7a4560$98a0a8c0@dgtac
Whole thread Raw
Responses Re: DeleteTransactionInProgress  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
Greetings,

I can't explain what happened but I'm getting some unusual messages that
I've never seen before and some operations in a specific table are not
working anymore (lock table and vacuum). I suppose that some kind of
curruption occurred in my database, but I'm not sure how can I deal with
this situation. Does someone know how can it be fixed? Is it possible to
'reset' TransactionProgress state or something like this? I even restarted
database and it's still the same situation? The problem has ocurred in a old
PostgreSQL 7.1.2 / RedHat 6.2 machine. I've been searching through mailing
lists but I couldn't find too much info about this.

Here are some of my psql logs :

bxs=# begin;
BEGIN
bxs=# lock table cham_atendimento;
NOTICE:  current transaction is aborted, queries ignored until end of
transaction block
*ABORT STATE*
bxs=#
bxs=# rollback;
ROLLBACK
bxs=# VACUUM VERBOSE ANALYZE cham_atendimento;
NOTICE:  --Relation cham_atendimento--
NOTICE:  Rel cham_atendimento: TID 304/76: DeleteTransactionInProgress 0 -
can't shrink relation
NOTICE:  Rel cham_atendimento: TID 305/2: DeleteTransactionInProgress 0 -
can't shrink relation
NOTICE:  Rel cham_atendimento: TID 305/4: DeleteTransactionInProgress 0 -
can't shrink relation
.....<it repeats last pattern a lot of times>
NOTICE:  Pages 316: Changed 0, reaped 316, Empty 0, New 0; Tup 25525: Vac 0,
Keep/VTL 0/0, Crash 0, UnUsed 1788, MinLen 84,
 MaxLen 93; Re-using: Free/Avail. Space 22880/0; EndEmpty/Avail. Pages 0/0.
CPU 0.08s/0.04u sec.
NOTICE:  Index xpkcham_atendiment: Pages 213; Tuples 25525: Deleted 0. CPU
0.04s/0.20u sec.
NOTICE:  Index xie1cham_atendimen: Pages 428; Tuples 25525: Deleted 0. CPU
0.09s/0.18u sec.
NOTICE:  Index xie2cham_atendimen: Pages 318; Tuples 25525: Deleted 0. CPU
0.13s/0.21u sec.
NOTICE:  Analyzing...
VACUUM
bxs=#
bxs=# SELECT count(*) FROM cham_atendimento;
 count
-------
 25525
(1 row)

Best regards,

----------------------------------------------------------------------------
----
José Vilson de Mello de Farias
Software Engineer

Dígitro Tecnologia Ltda - www.digitro.com.br
APC - Customer Oriented Applications
E-mail: vilson.farias@digitro.com.br
Tel.: +55 48 281 7158
ICQ 11866179


pgsql-general by date:

Previous
From: "Moritz Lennert"
Date:
Subject: Re: vacuum full fills up disk - solved
Next
From: Tom Lane
Date:
Subject: Re: Possible bug in date(timestamptz)?