Re: Fix pg_waldump to exit cleanly at end of WAL

On Tue, Sep 2, 2025 at 5:42 PM Chao Li <li.evan.chao@gmail.com> wrote:
>
> Hi Hackers,
>
> I noticed a problem with pg_waldump. When "-e" is not specified, or given a bigger number beyond the current LSN,
pg_dumpalways print an error message like: 
>
> ```
> chaol@ChaodeMacBook-Air cndb % pg_waldump -p pg_wal -s 0/2980DE10 -e 0/2980E680
> <...omit some outputs...>
> rmgr: Standby     len (rec/tot):     50/    50, tx:          0, lsn: 0/2980E468, prev 0/2980E3F0, desc: RUNNING_XACTS
nextXid866 latestCompletedXid 865 oldestRunningXid 866 
> pg_waldump: error: error in WAL record at 0/2980E468: invalid record length at 0/2980E4A0: expected at least 24, got
0
> ```
>
> The error message is kinda misleading and unnecessary. So I created a fix.

Can pg_waldump really distinguish between the end of WAL and corruption?

Looking back at past discussions, there seem similar patches and
conversations [1][2].
These might be useful references when considering this issue.

Regards,

[1] https://postgr.es/m/20200228.160100.2210969269596489579.horikyota.ntt@gmail.com
[2] https://postgr.es/m/674af9dc.050a0220.268e1c.4048@mx.google.com

--
Fujii Masao