triggering a wraparound by forcing some page corruption of table on disk, making autovacuum fail

gistfile1.txt

this is just for fun.
we corrupt some page of a table that is heavily bloated and autovaccum has not yet run. now the bloat is enough for autovacuum to see the threshold breach
and trigger its run on the bloated table. it runs, then it finds the corrupted page and it fails silently. then it sleeps but it finds the 
same table again for threshold breach and runs it again only to fail again. this goes on forever, monitoring shows no new bloat, but the table 
does not have enough bloat to trouble other queries. (maybe a cold table). but then, txid starts growing and we now see
a txid exhaustion limit approaching. it triggers forced autovaccum, but still it keeps growing. 

postgres@ubuntu:/mnt/VHD$ pg_ctl -D db1 -l logfile start
waiting for server to start.... done
server started
postgres@ubuntu:/mnt/VHD$ psql
psql (18.1 (Ubuntu 18.1-1.pgdg24.04+2))
Type "help" for help.

postgres=# create table t(col1 int primary key, col2 int);
CREATE TABLE
postgres=# alter table t set (autovacuum_enabled = 0);
ALTER TABLE
postgres=# select pg_relation_filepath('t');
 pg_relation_filepath
----------------------
 base/5/16384
(1 row)

postgres=# insert into t select x,x from generate_series(1, 10000) x;
INSERT 0 10000
postgres=# do $$
begin
for i in 1..20 loop
update t set col2 = i::int; commit;
end loop;
end; $$;
DO
postgres=# select * from pg_stat_user_tables where relname = 't';
postgres=# analyze t;
ANALYZE
postgres=# \x
Expanded display is on.
postgres=# select * from pg_stat_user_tables where relname = 't';
-[ RECORD 1 ]----------+------------------------------
relid                  | 16384
schemaname             | public
relname                | t
seq_scan               | 21
last_seq_scan          | 2025-12-16 06:23:19.442824+00
seq_tup_read           | 200000
idx_scan               | 0
last_idx_scan          |
idx_tup_fetch          | 0
n_tup_ins              | 10000
n_tup_upd              | 200000
n_tup_del              | 0
n_tup_hot_upd          | 1784
n_tup_newpage_upd      | 198216
n_live_tup             | 10000
n_dead_tup             | 198522
n_mod_since_analyze    | 0
n_ins_since_vacuum     | 10000
last_vacuum            |
last_autovacuum        |
last_analyze           | 2025-12-16 06:23:39.652082+00
last_autoanalyze       |
vacuum_count           | 0
autovacuum_count       | 0
analyze_count          | 1
autoanalyze_count      | 0
total_vacuum_time      | 0
total_autovacuum_time  | 0
total_analyze_time     | 3
total_autoanalyze_time | 0

postgres=# create extension pageinspect;
CREATE EXTENSION
postgres=# \q
postgres@ubuntu:/mnt/VHD$ ls -l db1/base/5/16384
-rw------- 1 postgres postgres 7553024 Dec 16 06:23 db1/base/5/16384
postgres@ubuntu:/mnt/VHD$ perl -le 'print 7553024/8192'
922
postgres@ubuntu:/mnt/VHD$ #dd if=/dev/urandom seek=000 bs=8k of=db1/base/5/16384 conv=notrunc,noerror count=1
postgres@ubuntu:/mnt/VHD$ psql
psql (18.1 (Ubuntu 18.1-1.pgdg24.04+2))
Type "help" for help.

postgres=# alter system set log_autovacuum_min_duration TO 0;
ALTER SYSTEM
postgres=# select pg_reload_conf();
 pg_reload_conf
----------------
 t
(1 row)

postgres=# \q
postgres@ubuntu:/mnt/VHD$ pg_ctl -D db1 -l logfile stop
waiting for server to shut down.... done
server stopped
postgres@ubuntu:/mnt/VHD$ dd if=/dev/urandom seek=900 bs=8k of=db1/base/5/16384 conv=notrunc,noerror count=1
1+0 records in
1+0 records out
8192 bytes (8.2 kB, 8.0 KiB) copied, 8.6619e-05 s, 94.6 MB/s
postgres@ubuntu:/mnt/VHD$ pg_ctl -D db1 -l logfile start
waiting for server to start.... done
server started
postgres@ubuntu:/mnt/VHD$ psql
psql (18.1 (Ubuntu 18.1-1.pgdg24.04+2))
Type "help" for help.

postgres=# alter table t set (autovacuum_enabled = 1);
ALTER TABLE
postgres=# alter table t set (autovacuum_
autovacuum_analyze_scale_factor        autovacuum_freeze_min_age              autovacuum_multixact_freeze_table_age  autovacuum_vacuum_insert_threshold
autovacuum_analyze_threshold           autovacuum_freeze_table_age            autovacuum_vacuum_cost_delay           autovacuum_vacuum_max_threshold
autovacuum_enabled                     autovacuum_multixact_freeze_max_age    autovacuum_vacuum_cost_limit           autovacuum_vacuum_scale_factor
autovacuum_freeze_max_age              autovacuum_multixact_freeze_min_age    autovacuum_vacuum_insert_scale_factor  autovacuum_vacuum_threshold
postgres=# alter table t set (autovacuum_vacuum_threshold = 1);
ALTER TABLE
postgres=# alter table t set (autovacuum_vacuum_scale_factor = 0);
ALTER TABLE
postgres=# \x
Expanded display is on.
postgres=# select * from pg_stat_user_tables where relname = 't';
-[ RECORD 1 ]----------+------------------------------
relid                  | 16384
schemaname             | public
relname                | t
seq_scan               | 21
last_seq_scan          | 2025-12-16 06:23:19.442824+00
seq_tup_read           | 200000
idx_scan               | 0
last_idx_scan          |
idx_tup_fetch          | 0
n_tup_ins              | 10000
n_tup_upd              | 200000
n_tup_del              | 0
n_tup_hot_upd          | 1784
n_tup_newpage_upd      | 198216
n_live_tup             | 10000
n_dead_tup             | 198522
n_mod_since_analyze    | 0
n_ins_since_vacuum     | 10000
last_vacuum            |
last_autovacuum        |
last_analyze           | 2025-12-16 06:23:39.652082+00
last_autoanalyze       |
vacuum_count           | 0
autovacuum_count       | 0
analyze_count          | 1
autoanalyze_count      | 0
total_vacuum_time      | 0
total_autovacuum_time  | 0
total_analyze_time     | 3
total_autoanalyze_time | 0

postgres=# vacuum (verbose) t;
INFO:  vacuuming "postgres.public.t"
ERROR:  invalid page in block 900 of relation "base/5/16384"
CONTEXT:  while scanning block 899 of relation "public.t"