Re: Advice Needed: Simultaneous Upgrade of Two-Node PostgreSQL 11 Cluster - Mailing list pgsql-admin

From kasem adel
Subject Re: Advice Needed: Simultaneous Upgrade of Two-Node PostgreSQL 11 Cluster
Date
Msg-id CAP-pBm-Uk9VtTppHu=mrRO=DqM-=asdp_VyQRkFHDosY2tS8vA@mail.gmail.com
Whole thread Raw
In response to Advice Needed: Simultaneous Upgrade of Two-Node PostgreSQL 11 Cluster  (kasem adel <kasemadel8@gmail.com>)
List pgsql-admin

Dears,

I hope this message finds you well.

Thank you for your support. After completing the upgrade process, we encountered high disk IO due to queries selecting incorrect plans for specific table. When we ran ANALYZE, it did not choose the correct plan, likely because we are using the default ANALYZE settings, and our data size is approximately 1.8 TB. Could you kindly provide a solution to address this issue?

Thank you in advance for your assistance.



في الجمعة، ١٧ يناير ٢٠٢٥ ١٠:٥٦ م kasem adel <kasemadel8@gmail.com> كتب:
Dear Scott Ribe,

Thanks for your support after upgrade process we faced high disk IO because query chose wrong  plann but when run analyze not chose the correct plan because we are used default analyze and data size is 1.8 TB can you provide us with solution to solve this issue.

Thanks 



في الخميس، ٩ يناير ٢٠٢٥ ٣:٤٤ م Scott Ribe <scott_ribe@elevated-dev.com> كتب:
> On Jan 7, 2025, at 6:50 AM, kasem adel <kasemadel8@gmail.com> wrote:
>
> Kindly be informed that database size is 1.8 TB and we have an low network bandwidth between two node 10 mbps and to transfer the data from scratch that will take from 21 day to 30 day  for this we need solution in upgrade to prevent load data from scratch .

Yes, and I gave you alternatives taking that into account. I think you have all the answers you're going to get, and now have to study them and choose an approach.

pgsql-admin by date:

Previous
From: Laurenz Albe
Date:
Subject: Re: vacuuming taking long time for pg_attribute
Next
From: jayesh thakare
Date:
Subject: Error in production postgresql database