RE: pg_recvlogical requires -d but not described on the documentation - Mailing list pgsql-hackers

Dear Fujii-san, David,

> > BTW, I'm curious why --dbname isn't required for the --drop-slot action.
> 
> I'm analyzing around here...
>

Actually, replication slots can be dropped from another database where it created,
or even from the streaming replication connection.
I forked the new thread which fixes the description [1].

Based on the fact, there are two approaches to fix:

1. Fix not to raise fatal error like:

```
@@ -950,7 +950,7 @@ main(int argc, char **argv)
        if (!RunIdentifySystem(conn, NULL, NULL, NULL, &db_name))
                exit(1);
 
-       if (db_name == NULL)
+       if (!do_drop_slot && db_name == NULL)
                pg_fatal("could not establish database-specific replication connection");
```

db_name == NULL means that streaming replication connection has been established,
so other operations are not allowed.

2. Fix documentation

We keep the current restriction and clarify it. For the reportability, it is
OK for me to also modify the code like:

```
@@ -881,7 +881,7 @@ main(int argc, char **argv)
                exit(1);
        }
 
-       if (!do_drop_slot && dbname == NULL)
+       if (dbname == NULL)
```

Thought?

[1]:
https://www.postgresql.org/message-id/OSCPR01MB14966C6BE304B5BB2E58D4009F5DE2%40OSCPR01MB14966.jpnprd01.prod.outlook.com

Best regards,
Hayato Kuroda
FUJITSU LIMITED