From: Jerry Sievers <gsievers19@comcast.net>
To: pgsql-admin@lists.postgresql.org
Subject: ignore_system_indexes=on no help for missing/corrupt
 pg_class_oid_index?
Date: Tue, 08 Aug 2023 17:47:27 -0500
Message-ID: <m0o7jhqhhc.fsf@comcast.net>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/28.1 (darwin)
MIME-Version: 1.0
Content-Type: text/plain
Archived-At: 
 <https://www.postgresql.org/message-id/m0o7jhqhhc.fsf%40comcast.net>
Precedence: bulk

Greetings admins!

I found that the ignore_system_indexes flag apparently doesn't get me
past this case of a corrupt index as seen below.

Or am I doing something silly here?

The particular DB is a scratch so we'll just drop it but anyway,
comments?

I've not had to resort to this anytime lately but IIRC the
ignore_system_indexes foo did the expected for us $sometime before.

I believe that index 2662::regclass is probably this one but of course I
had to probe for it in some other DB.  The same one exists in template1
and I presume all other DBS as well.

Please advise.  Thx!

----


$ pg --single -D $PWD -c config_file=$cf -c hba_file=$hba -P broken_db
,,,,1969183,2023-08-08 22:25:17.844 GMT,,XX000,64d2c0cd.1e0c1f,1,2023-08-08 22:25:17 GMT,1/1,0 || PANIC:  could not open critical system index 2662
Aborted

$ pg --single -D $PWD -c config_file=$cf -c hba_file=$hba -P postgres
,,,,1969691,2023-08-08 22:25:51.027 GMT,,00000,64d2c0ee.1e0e1b,1,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  database system was interrupted; last known up at 2023-08-08 22:25:17 GMT
,,,,1969691,2023-08-08 22:25:51.045 GMT,,00000,64d2c0ee.1e0e1b,2,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  database system was not properly shut down; automatic recovery in progress
,,,,1969691,2023-08-08 22:25:51.048 GMT,,00000,64d2c0ee.1e0e1b,3,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  invalid record length at 0/5F175310: wanted 24, got 0
,,,,1969691,2023-08-08 22:25:51.048 GMT,,00000,64d2c0ee.1e0e1b,4,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  redo is not required
,,,,1969691,2023-08-08 22:25:51.055 GMT,,00000,64d2c0ee.1e0e1b,5,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  checkpoint starting: end-of-recovery immediate wait
,,,,1969691,2023-08-08 22:25:51.063 GMT,,00000,64d2c0ee.1e0e1b,6,2023-08-08 22:25:50 GMT,1/0,0 || LOG:  checkpoint complete: wrote 3 buffers (0.0%); 0 WAL file(s) added, 0 removed, 0 recycled; write=0.004 s, sync=0.001 s, total=0.010 s; sync files=2, longest=0.001 s, average=0.001 s; distance=0 kB, estimate=0 kB

PostgreSQL stand-alone backend 15.3 (Ubuntu 15.3-1.pgdg20.04+1)
backend> select relname from pg_class where oid=2662;
,,,,1969691,2023-08-08 22:26:11.183 GMT,,01000,64d2c0ee.1e0e1b,7,2023-08-08 22:25:50 GMT,1/2,0 || WARNING:  using index "pg_toast_2619_index" despite IgnoreSystemIndexes
	 1: relname	(typeid = 19, len = 64, typmod = -1, byval = f)
	----
	 1: relname = "pg_class_oid_index"	(typeid = 19, len = 64, typmod = -1, byval = f)
	----
backend> select version()
	 1: version	(typeid = 25, len = -1, typmod = -1, byval = f)
	----
	 1: version = "PostgreSQL 15.3 (Ubuntu 15.3-1.pgdg20.04+1) on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0, 64-bit"	(typeid = 25, len = -1, typmod = -1, byval = f)
	----
backend> show ignore_system_indexes
	 1: ignore_system_indexes	(typeid = 25, len = -1, typmod = -1, byval = f)
	----
	 1: ignore_system_indexes = "on"	(typeid = 25, len = -1, typmod = -1, byval = f)
	----
backend>