Assertion failure at standby promotion - Mailing list pgsql-hackers

From Fujii Masao
Subject Assertion failure at standby promotion
Date
Msg-id CAHGQGwEO1GF3=7LAeyn630+PNjjkC4fnwqx_L_Vyq6Y+OsB5jg@mail.gmail.com
Whole thread Raw
Responses Re: Assertion failure at standby promotion  (Amit Langote <amitlangote09@gmail.com>)
Re: Assertion failure at standby promotion  (Heikki Linnakangas <hlinnakangas@vmware.com>)
List pgsql-hackers
Hi,

I got the following assertion failure when I promoted the standby.

2013-05-04 00:12:31 JST sby1 LOG:  received promote request
2013-05-04 00:12:31 JST sby1 FATAL:  terminating walreceiver process
due to administrator command
2013-05-04 00:12:31 JST sby1 LOG:  redo done at 0/6FFE038
2013-05-04 00:12:31 JST sby1 LOG:  last completed transaction was at
log time 2013-05-04 00:12:25.878909+09
2013-05-04 00:12:31 JST sby1 LOG:  selected new timeline ID: 2
2013-05-04 00:12:31 JST sby1 LOG:  archive recovery complete
2013-05-04 00:12:31 JST sby1 LOG:  checkpoint starting:
TRAP: FailedAssertion("!(sentPtr <= sendTimeLineValidUpto)", File:
"walsender.c", Line: 1465)
2013-05-04 00:12:31 JST sby1 LOG:  autovacuum launcher started

The way to reproduce this is:

1. Create one master A, one standby B, and one cascade standby C.
2. Run pgbench -i -s 10
3. Promote the standby B before pgbench -i finishes

Regards,

-- 
Fujii Masao



pgsql-hackers by date:

Previous
From: Cédric Villemain
Date:
Subject: Re: Recovery target 'immediate'
Next
From: Andres Freund
Date:
Subject: Re: Remaining beta blockers