From: | C GG <cgg0007(at)gmail(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | PANIC: could not open critical system index 2662 |
Date: | 2018-08-04 14:45:59 |
Message-ID: | CAJXW-z8kkC2=WepG8Jf2c0WZyUma4BThLofG2v=Oji6+z2H89w@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
This comes at the end of a long cascade of problems stemming from a
lightning strike to our datacenter causing our battery backup systems to
fail, and overvoltage from the commercial power company to damage our
storage arrays which run some services and contain our on-site backups.
No problem, right-- just get the backups from the DR site. Even though we
have exercised our DR backups previously with no problems, this go we have
a data corruption issue. Murphy's law, right?
When trying to restart the PostgreSQL 9.2 database for our JIRA
installation, we're getting
"PANIC: could not open critical system index 2662" ... I've tried various
things like turning on zero_damaged_pages but postmaster still crashes on
connect when attempting to pg_dump.
PANIC: could not open critical system index 2662
LOG: server process (PID 44753) was terminated by signal 6: Aborted
LOG: terminating any other active server processes
WARNING: terminating connection because of crash of another server process
DETAIL: The postmaster has commanded this server process to roll back the
current transaction and exit, because another server process exited
abnormally and possibly corrupted shared memory.
HINT: In a moment you should be able to reconnect to the database and
repeat your command.
LOG: all server processes terminated; reinitializing
LOG: database system was interrupted; last known up at 2018-08-04 10:04:19
EDT
LOG: database system was not properly shut down; automatic recovery in
progress
LOG: record with zero length at C/8C7E28D0
LOG: redo is not required
LOG: database system is ready to accept connections
LOG: autovacuum launcher started
Attempt to connect again, and you get the same set of messages. What other
ways can you think of exist for me to tackle getting a dump of the data?
I have a disk backup of the data dir prior to performing zero_damaged_pages
so we can do our worst, scrap the failed attempts and try as many other
things as we can think of.
If there are any low-level or recovery tools that would potentially work on
a 9.2 install, I'd love to hear about those as well. I looked at pg_hexdump
but that is 9.4+ ...
I could really use a miracle right now. Thanks all.
CG
From | Date | Subject | |
---|---|---|---|
Next Message | Christophe Pettus | 2018-08-04 14:54:36 | Re: Pg_rewind cannot load history wal |
Previous Message | Michael Paquier | 2018-08-04 13:13:12 | Re: Pg_rewind cannot load history wal |