Problem

From: Merlijn van der Mee <M(dot)vanderMee(at)amc(dot)uva(dot)nl>
To: pgsql-general(at)postgresql(dot)org
Subject: Problem
Date: 2000-07-27 12:03:49
Message-ID: 39802525.73FD8615@amc.uva.nl
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


This morning one of my postgres databases wasn't working anymore.

I can connect to this database, do a select on a table, but when
I do a vacuum, a select on a internal table (pg_...) or just '\d'
the backend gives a sigfault.

It has always worked, and the other databases running on the same port
are still functioning.

What am I doing wrong?

Example. (in psql)
select * from pg_attribute;
Backend message type 0x44 arrived while idle
Backend message type 0x44 arrived while idle
We have lost the connection to the backend, so further processing is
impossible. Terminating.

---
vacuum;
pqReadData() -- backend closed the channel unexpectedly.
This probably means the backend terminated abnormally
before or while processing the request.
We have lost the connection to the backend, so further processing is
impossible. Terminating.

---

Example (with truss postgres [databasename])

vacuum pg_class
read(0, " v a c u u m p g _ c l".., 1024) = 16
time() = 964698637
ioctl(0, TCGETA, 0xEFFF6A84) = 0
NOTICE: CreatePortal: portal <vacuum> already exists
write(2, " N O T I C E : C r e".., 54) = 54
NOTICE: CreatePortal: portal <vacuum> already exists
write(2, " N O T I C E : C r e".., 54) = 54
open("pg_vlock", O_RDONLY|O_CREAT|O_EXCL, 0600) = 12
close(12) = 0
time() = 964698637
lseek(7, 0, SEEK_END) = 180224
lseek(7, 0, SEEK_END) = 180224
lseek(7, 8192, SEEK_SET) = 8192
read(7, "01 (01 @ \0 \0 E08090".., 8192) = 8192
lseek(7, 16384, SEEK_SET) = 16384
read(7, "01 40188 \0 \0 ? 80 n".., 8192) = 8192
lseek(7, 24576, SEEK_SET) = 24576
read(7, "01 (01 8 \0 \0 71080 n".., 8192) = 8192
lseek(7, 32768, SEEK_SET) = 32768
read(7, "01 01 H \0 \0 ; 080 n".., 8192) = 8192
lseek(7, 40960, SEEK_SET) = 40960
read(7, "01 01 0 \0 \010808090".., 8192) = 8192
lseek(7, 49152, SEEK_SET) = 49152
read(7, "01 $0180 \0 \012\08090".., 8192) = 8192
lseek(7, 57344, SEEK_SET) = 57344
read(7, "011801 ( \0 \0 108090".., 8192) = 8192
lseek(7, 65536, SEEK_SET) = 65536
read(7, "01 01 P \0 \0 ? 80 n".., 8192) = 8192
lseek(7, 73728, SEEK_SET) = 73728
read(7, "01 01 0 \0 \0 >E08090".., 8192) = 8192
lseek(7, 81920, SEEK_SET) = 81920
read(7, "01 01 ` \0 \0 2 80 n".., 8192) = 8192
lseek(7, 90112, SEEK_SET) = 90112
read(7, "01 01 H \0 \0\t8080 n".., 8192) = 8192
lseek(7, 98304, SEEK_SET) = 98304
read(7, "011C01 ` \0 \0 ' @80 l".., 8192) = 8192
lseek(7, 106496, SEEK_SET) = 106496
read(7, "01140180 \0 \0 ? 80 l".., 8192) = 8192
lseek(7, 114688, SEEK_SET) = 114688
read(7, "011C01 0 \0 \0 ? 80 n".., 8192) = 8192
lseek(7, 122880, SEEK_SET) = 122880
read(7, "011401 x \0 \0 ? 80 l".., 8192) = 8192
lseek(7, 131072, SEEK_SET) = 131072
read(7, "011401 X \0 \0 ? 80 n".., 8192) = 8192
lseek(7, 139264, SEEK_SET) = 139264
read(7, "01140180 \0 \0 ? 080 f".., 8192) = 8192
lseek(7, 147456, SEEK_SET) = 147456
read(7, "011401 x \0 \0 ? 80 n".., 8192) = 8192
lseek(7, 155648, SEEK_SET) = 155648
read(7, "011401C8 \0 \0 >E08090".., 8192) = 8192
lseek(7, 163840, SEEK_SET) = 163840
read(7, "\0\0\0\0\0\0\0\0\0\0\0\0".., 8192) = 8192
Incurred fault #5, FLTACCESS %pc = 0x00047CA4
siginfo: SIGBUS BUS_ADRALN addr=0x002C19B1
Received signal #10, SIGBUS [default]
siginfo: SIGBUS BUS_ADRALN addr=0x002C19B1
*** process killed ***

Anybody?!?

Merlijn

--
Merlijn van der Mee - Bioinformatician
Academic Medical Centre - Amsterdam
Bioinformatics Laboratory
Email: M(dot)vanderMee(at)amc(dot)uva(dot)nl tel. +31-20-5668601
-
Version: 3.1
GCS d- s+:- a- C+++ UL++ P++++ L+++ E---- W++ N o-- K- w---
O M-- V-- PS+++ PE- Y+ PGP+ t+ 5 X- R+ tv-- b+ DI-- D++
G e* h- r--- y?
--

Browse pgsql-general by date

  From Date Subject
Next Message igor 2000-07-27 12:17:32 upgrade from 6.4 to 7.02 problem
Previous Message Andrew Snow 2000-07-27 10:58:13 RE: 4 billion record limit?