public inbox for [email protected]  

help / color / mirror / Atom feed

From: Ramakrishna Reddy Nandyala <[email protected]>
To: Pgsql-admin <[email protected]>
Subject: Any known bugs with respect to data corruption in pg12/pg 14
Date: Wed, 3 Jun 2026 17:44:39 +0530
Message-ID: <CA++_dZP4ygDekyvxmYbM8=egGshB4CRAd1mBHrdusO3CXzYwuw@mail.gmail.com> (raw)
In-Reply-To: <CANzqJaAb149k3pYmbnZ8-Mcc7Ec_DwJMkKiZK=tPA34Rsq_kDw@mail.gmail.com>
References: <CANzqJaCNaqhQ9duPP+eKU2OR1wjGYW-FBQ2FZXqnBRb6XAnKQA@mail.gmail.com>
	<[email protected]>
	<CANzqJaCJGS9S=ibJ_rOmP=acA7n0XCA5R4FbMH+AH3WURhey8Q@mail.gmail.com>
	<[email protected]>
	<CANzqJaAb149k3pYmbnZ8-Mcc7Ec_DwJMkKiZK=tPA34Rsq_kDw@mail.gmail.com>

Hi Gurus

We are facing a peculiar issue with respect to pg 12 and pg14 on
RHEL8u10,post os patch objects are getting corrupted. The details are
unknown until we access the object's.
System Stack:

 Db: pg12.22,pg14.19
Os : Rhel 8u10
THALES CTM FOR encryption solution
Stirage : Pure storage and VSAN DISK

We are not seeing any errors reported with respect to the corruption,unless
until we select The problematic object that to the problematic block only

Below corruption scenarios we observed
1) if the corruption happen on standby node no impact to the production
2) if corruption happen on production we could see same object we are
unable to select from standby node,means corruption is transfer to standby
node
3) if the corruption happen on block 0 it is not allowing to perform any
DML activity,if it is on other blocks it is allowing to
insert/update/delete in other good blocks,if system touches the problematic
block only it is failing


Need your guide lines how to identify what is going wrong,and where is
going wrong?

Is there any known bugs in these version with the given stack pg12/14 with
RHEL8u10

Is there any way we can stop not to replicate the problematic block to
standby node?

What will be the best practice to identify the problematic objects or blocks

We tried amcheck on pg14 it is not giving any results

Your expertise advise will be help us on coming out of the critical issue

view thread (7+ messages)  latest in thread

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: [email protected]
  Cc: [email protected], [email protected]
  Subject: Re: Any known bugs with respect to data corruption in pg12/pg 14
  In-Reply-To: <CA++_dZP4ygDekyvxmYbM8=egGshB4CRAd1mBHrdusO3CXzYwuw@mail.gmail.com>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox