ahouska

Thanks @shruthikc-gowda for pointing that out. PG 13 will be EOL in November, so I'll fix this issue when removing support for PG 13.

Thanks for the pull request. I'll handle the compatibility as soon as I know whether the [REPACK CONCURRENTLY](https://commitfest.postgresql.org/patch/5117/) command ended up in PG 18 or not. With this command, pg_squeeze...

Interesting, I thought it works with PG 17. I'll check next week. Anyway, version 1.2 should be used in the future - I'll create it next week. You can try...

It seems like you miss [this commit](https://github.com/cybertec-postgresql/pg_rewrite/commit/8a022134dd07c6d8c3a3c81dd06925d5fdf61a1e) but the branch REL1_1 does contain it. Nevertheless, [version 1.2 is available now](https://github.com/cybertec-postgresql/pg_rewrite/commits/REL1_2). (I'll create a tag short before the release of PG...

I wonder if this error is specific to `pg_squeeze`. Does the same error occur if you run the following? `psql -d 'replication=database dbname=postgres' -c 'CREATE_REPLICATION_SLOT pg_squeeze_test_slot LOGICAL pg_squeeze'` Please do...

Why do you think the error happens during the swap phase? Does server log indicate so? In such a case, the only idea I have so far is that `pg_class`...

Hm, `pg_class` actually has no TOAST relation. Can you please dump the table (without data), restore it in another database and try if the squeezing completes? Failure here would reveal...

This is rather a wild idea, anyway: I wonder if you managed to "accidentally" add TOAST table to some system catalog. It's probably not very likely, but might be worth...