Horton

Re: Wrong statistics for size of XLOG_SWITCH during pg_waldump.

movead.li@highgo.ca <movead.li@highgo.ca>

View thread View on postgr.es
From: "movead.li@highgo.ca" <movead.li@highgo.ca>
To: Horiguchi <horikyota.ntt@gmail.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>, ahsan.hadi <ahsan.hadi@highgo.ca>
Date: 2020-10-10T01:50:02Z
Lists: pgsql-hackers

Attachments

>I think that the length of the XLOG_SWITCH record is no other than 24
>bytes. Just adding the padding? garbage bytes to that length doesn't
>seem the right thing to me.
>
>If we want pg_waldump to show that length somewhere, it could be shown
>at the end of that record explicitly:
> 
>rmgr: XLOG        len (rec/tot):     24/16776848, tx:          0, lsn: 0/02000148, prev 0/02000110, desc: SWITCH, trailing-bytes: 16776944

Thanks, I think it's good idea, and new patch attached.

Here's the lookes:
rmgr: XLOG        len (rec/tot):     24/    24, tx:          0, lsn: 0/030000D8, prev 0/03000060, desc: SWITCH, trailing-bytes: 16776936




Regards,
Highgo Software (Canada/China/Pakistan) 
URL : www.highgo.ca 
EMAIL: mailto:movead(dot)li(at)highgo(dot)ca

Commits

  1. pg_waldump: Fix bug in per-record statistics.