BUG #2667: vacuuming a 70GB table causes a "crash"

From: "Stephen R(dot) van den Berg" <srb(at)cuci(dot)nl>
To: pgsql-bugs(at)postgresql(dot)org
Subject: BUG #2667: vacuuming a 70GB table causes a "crash"
Date: 2006-09-30 13:16:15
Message-ID: 200609301316.k8UDGFHT047182@wwwmaster.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs


The following bug has been logged online:

Bug reference: 2667
Logged by: Stephen R. van den Berg
Email address: srb(at)cuci(dot)nl
PostgreSQL version: 8.1.4
Operating system: Linux 2.6.16.20 Debian (pg: 8.1.4-6)
Description: vacuuming a 70GB table causes a "crash"
Details:

datalog=# vacuum verbose fnetsense;
INFO: vacuuming "public.fnetsense"
PANIC: right sibling is not next child in "fnetsense_utstamp_key"
server closed the connection unexpectedly

Previously, the database was fully packed, no deletetions, just insertions
at the end (kind of a logging DB).
Then I deleted around 50GB worth of data (the first 50GB).
That worked just fine, then I ran vacuum the first
time (on this table). It ran, and did 3 or 4 passes
successfully, then it bombed reporting the
error above. Then I tried running vacuum again, same
error as above again.

Restarted the server, didn't help. Error is persistent
now, DB is still working, don't know how to repair the
DB.

The table being vacuumed is inserted into in realtime
at a rate of 23 new entries per second.

This is the table definition of the table in question:
Table "public.fnetsense"
Column | Type | Modifiers
-------------------------------+---------+-----------
nsb | integer | not null
utstamp | integer | not null
ifoutoctets_1 | bigint | not null
ifoutucastpkts_1 | bigint | not null
ifinoctets_1 | bigint | not null
ifinucastpkts_1 | bigint | not null
opampsdownchannelcolor | "char" | not null
opampsdownchannelmains | "char" | not null
opampsupchannelcolor | "char" | not null
opampssigqcorrecteds | "char" | not null
opampssigquncorrectables | "char" | not null
opampssigqsignalwater | "char" | not null
opampssigqmicrobeachsand | "char" | not null
opampsfostatustxmains | "char" | not null
opampsfostatustttimeouts | "char" | not null
opampsfotsfostatusvalue | "char" | not null
opampsfotsfostatussignalwater | "char" | not null
opampsfotsfostatusrxmains | "char" | not null
fotsnloss | "char" | not null
fotsnpadj | "char" | not null
fotsrnginvminreqs | "char" | not null
Indexes:
"fnetsense_nsbutstamp_key" btree (nsb, utstamp)
"fnetsense_utstamp_key" btree (utstamp)
Foreign-key constraints:
"fnetsense_nsb_fkey" FOREIGN KEY (nsb) REFERENCES netsensebase(nsb)
ON UPDATE CASCADE ON DELETE CASCADE DEFERRABLE INITIALLY DEFERRED

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2006-09-30 21:01:38 Re: BUG #2667: vacuuming a 70GB table causes a "crash"
Previous Message how to cluster 2006-09-30 03:45:09 BUG #2666: how do cluster with 7.4.13 ?