RE: Unable to start replica after failover - Mailing list pgsql-general

From Lahnov, Igor
Subject RE: Unable to start replica after failover
Date
Msg-id 27cb3d242a9d4b88a6f1e22ba9716441@nexign.com
Whole thread Raw
In response to Re: Unable to start replica after failover  (Alexander Kukushkin <cyberdemn@gmail.com>)
Responses Re: Unable to start replica after failover
List pgsql-general

Hi,

Yes, the *patial* from the *new leader* is restored to *last leader* and renamed to 000000020000054E000000FB, without *partial* postfix.

 

>>Postgres asks for file 000000020000054E000000FB but somehow gets

>>000000020000054E000000FB.partial instead. Why?

 

Yes, Postgres asks for 000000020000054E000000FB and gets renamed 000000020000054E000000FB.partial (without *partial* postfix).

 

And, finally, the *last leader* has a local 000000020000054E000000FB.partial, with the segments needed for recovery, and 000000020000054E000000FB ( renamed 000000020000054E000000FB.partial from new leader). But 000000020000054E000000FB has a higher priority than 000000020000054E000000FB.partial

Also, the *new leader* archives full 000000030000054E000000FB of the new timeline. And this file contains the same as 000000020000054E000000FB data.

 

 

 

 

pgsql-general by date:

Previous
From: Laurenz Albe
Date:
Subject: Re: recovery_command has precedence over phisical slots?
Next
From: Ron
Date:
Subject: Re: Setting up a server with previous day data