Re: AtEOXact_ApplyLauncher() and subtransactions - Mailing list pgsql-hackers

From Amit Khandekar
Subject Re: AtEOXact_ApplyLauncher() and subtransactions
Date
Msg-id CAJ3gD9dKN1quTH5Exnfk6kKE=55UQ1Rm1fdEaH6rs+nUbShxtg@mail.gmail.com
Whole thread Raw
In response to Re: AtEOXact_ApplyLauncher() and subtransactions  (Amit Khandekar <amitdkhan.pg@gmail.com>)
Responses Re: AtEOXact_ApplyLauncher() and subtransactions
List pgsql-hackers
On 18 June 2018 at 15:02, Amit Khandekar <amitdkhan.pg@gmail.com> wrote:
> On 16 June 2018 at 00:03, Amit Khandekar <amitdkhan.pg@gmail.com> wrote:
>> The way I am implementing this can be seen in attached
>> apply_launcher_subtrans_WIP.patch. (check launcher.c changes). I
>> haven't started testing it yet though.
>
> Attached patch passes some basic testing I did. Will do some more
> testing, and some self-review and code organising, if required.

Done. Attached is v2 version of the patch. Comments welcome.

Changed GetSubscriptionRelations() to GetSubscriptionRelids(), which
now returns only the oids, not the subrel states. This was convenient
for storing the exact returned list into the committed subscription
rels. And anyways the subrel states were not used anywhere.

> I will also split the patch into two : one containing the main issue
> regarding subtransaction, and the other containing the other issue I
> mentioned earlier that shows up without subtransaction as well.

Did not split the patch. The changes for the other issue that shows up
without subtransaction are all in subscriptioncmds.c , and that file
contains mostly the changes for this issue. So kept it as a single
patch. But if it gets inconvenient for someone while reviewing, I will
be happy to split it.

-- 
Thanks,
-Amit Khandekar
EnterpriseDB Corporation
The Postgres Database Company

Attachment

pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: Fast default stuff versus pg_upgrade
Next
From: Andres Freund
Date:
Subject: Re: Fast default stuff versus pg_upgrade