| From: | Michael Paquier <michael(at)paquier(dot)xyz> |
|---|---|
| To: | Robert Haas <robertmhaas(at)gmail(dot)com> |
| Cc: | Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>, Andres Freund <andres(at)anarazel(dot)de>, amul sul <sulamul(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: ThisTimeLineID can be used uninitialized |
| Date: | 2021-10-21 05:41:16 |
| Message-ID: | YXD9fDgWjRgIQVbw@paquier.xyz |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, Oct 20, 2021 at 09:08:57AM -0400, Robert Haas wrote:
> On Tue, Oct 19, 2021 at 7:30 PM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
>>> Istm we should introduce an InvalidTimeLineID, and explicitly initialize
>>> sendTimeLine to that, and assert that it's valid / invalid in a bunch of
>>> places?
>>
>> That's not a bad idea; it'll help discover bogus code. Obviously, some
>> additional tests wouldn't harm -- we have a lot more coverage now than
>> in embarrasingly recent past, but it can still be improved.
>
> +1.
There is already an assumption in walsender.c where an invalid
timeline is 0, by the way? See sendTimeLineNextTLI and sendTimeLine.
Asserting here and there looks like a good thing to do for code paths
where the timeline should, or should not, be set.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Masahiko Sawada | 2021-10-21 05:45:49 | Re: [Bug] Logical Replication failing if the DateStyle is different in Publisher & Subscriber |
| Previous Message | Michael Paquier | 2021-10-21 05:35:08 | Re: pg_receivewal starting position |