Re: How to use pg_waldump?

Hi André,

> On 01. May, 2020, at 12:47, André Hänsel <andre(at)webkr(dot)de> wrote:
>
> I知 using pgBackRest for incremental backups which, as far as I understand,
> use the WAL. These backups are relatively large, so I wanted to take a look
> at my WAL. I understand pg_waldump is the tool for this.
>
> However, I struggle with its usage.
>
> The --help output suggests all command line parameters are optional, but
> running it like that yields "pg_waldump: no arguments specified". So I tried
> "pg_waldump -z" which yields "pg_waldump: FATAL: could not find any WAL
> file". Ok, so apparently it doesn't know the location of the WAL files. I
> then tried "pg_waldump -p /var/lib/postgresql/11/main/pg_wal".
>
> Now I get "pg_waldump: no start WAL location given". And this is where I'm
> stuck. I don't know any WAL location. I don't even know how far back my WAL
> goes, that's one thing I want to find out, among other things.
>
> On https://www.postgresql.org/docs/11/pgwaldump.html it says about
> "--start":
>> WAL location at which to start reading. The default is to start reading
> the first valid log record found in the earliest file found.
>
> If that's the default, why does it ask me for a WAL location?
>
> What do I need to do?

try:

pg_waldump -p /var/lib/postgresql/11/main/pg_wal <start-wal> [<end-wal>]

where <start-wal> is the name of the WAL file to start and (optionally) <end-wal> is the WAL file to stop. It reads and shows all information of the WAL files in this range.

Hope, this helps.

Cheers,
Paul