Re: .ready and .done files considered harmful - Mailing list pgsql-hackers

From Dipesh Pandit
Subject Re: .ready and .done files considered harmful
Date
Msg-id CAN1g5_EZB7W=+koas6yvG4huNy-bkWau9YXHQEEFN_6Y9D0igg@mail.gmail.com
Whole thread Raw
In response to Re: .ready and .done files considered harmful  (Kyotaro Horiguchi <horikyota.ntt@gmail.com>)
Responses Re: .ready and .done files considered harmful  ("Bossart, Nathan" <bossartn@amazon.com>)
Re: .ready and .done files considered harmful  ("Bossart, Nathan" <bossartn@amazon.com>)
List pgsql-hackers
Hi,

Thanks for the feedback.

The possible path that archiver can take for each cycle is either a fast
path or a fall-back patch. The fast path involves checking availability of
next anticipated log segment and decide the next target for archival or
a fall-back path which involves full directory scan to get the next log segment.
We need a mechanism that enables the archiver to select the desired path
for each cycle.

This can be achieved by maintaining a shared memory flag. If this flag is set
then archiver should take the fall-back path otherwise it should continue with
the fast path.

This flag can be set by backend in case if an action like timeline switch,
.ready files created out of order,...  requires archiver to perform a full
directory scan.

I have incorporated these changes and updated a new patch. PFA patch v6.

Thanks,
Dipesh
Attachment

pgsql-hackers by date:

Previous
From: Nitin Jadhav
Date:
Subject: Re: when the startup process doesn't (logging startup delays)
Next
From: Greg Nancarrow
Date:
Subject: Re: Skipping logical replication transactions on subscriber side