"out of relcache_callback_list slots" after multiple calls to pg_logical_slot_get_binary_changes - Mailing list pgsql-hackers

Hi hackers,

After multiple calls to the function pg_logical_slot_get_binary_changes() in
single client backend (the output plugin of the slot is pgoutput), I got the
following error:

client backend FATAL:  out of relcache_callback_list slots
client backend CONTEXT:  slot "testslot", output plugin "pgoutput", in the startup callback
client backend STATEMENT:  SELECT data FROM pg_logical_slot_get_binary_changes('testslot', NULL, NULL, 'proto_version',
'3','streaming', 'off', 'publication_names', 'pub');
 

I tried to look into it and found that it's because every time the function
(pg_logical_slot_get_binary_changes) is called, relcache callback and syscache
callbacks are registered when initializing pgoutput (see pgoutput_startup() and
init_rel_sync_cache()), but they are not unregistered when it shutdowns. So,
after multiple calls to the function, MAX_RELCACHE_CALLBACKS is exceeded. This
is mentioned in the following comment.

    /*
     * We can get here if the plugin was used in SQL interface as the
     * RelSchemaSyncCache is destroyed when the decoding finishes, but there
     * is no way to unregister the relcache invalidation callback.
     */
    if (RelationSyncCache == NULL)
        return;

Could we fix it by adding two new function to unregister relcache callback and
syscache callback? I tried to do so in the attached patch.

Regards,
Shi Yu