Re: Minimal logical decoding on standbys

Hi,

On 2/25/22 10:34 AM, Drouvot, Bertrand wrote:
> Hi,
>
> On 10/28/21 11:07 PM, Andres Freund wrote:
>> Hi,
>>
>> On 2021-10-28 16:24:22 -0400, Robert Haas wrote:
>>> On Wed, Oct 27, 2021 at 2:56 AM Drouvot, Bertrand
>>> <bdrouvot(at)amazon(dot)com> wrote:
>>>> So you have in mind to check for XLogLogicalInfoActive() first, and
>>>> if true, then open the relation and call
>>>> RelationIsAccessibleInLogicalDecoding()?
>>> I think 0001 is utterly unacceptable. We cannot add calls to
>>> table_open() in low-level functions like this. Suppose for example
>>> that _bt_getbuf() calls _bt_log_reuse_page() which with 0001 applied
>>> would call get_rel_logical_catalog(). _bt_getbuf() will have acquired
>>> a buffer lock on the page. The idea that it's safe to call
>>> table_open() while holding a buffer lock cannot be taken seriously.
>> Yes - that's pretty clearly a deadlock hazard. It shouldn't too hard
>> to fix, I
>> think. Possibly a bit more verbose than nice, but...
>>
>> Alternatively we could propagate the information whether a relcache
>> entry is
>> for a catalog from the table to the index. Then we'd not need to
>> change the
>> btree code to pass the table down.
>
> +1 for the idea of propagating to the index. If that sounds good to
> you too, I can try to have a look at it.
>
> Thanks Robert and Andres for the feedbacks you have done on the
> various sub-patches.
>
> I've now in mind to work sub patch by sub patch (starting with 0001
> then) and move to the next one once we agree that the current one is
> "ready".
>
> I think that could help us to get this new feature moving forward more
> "easily", what do you think?
>
> Thanks
>
> Bertrand
>
I'm going to re-create a CF entry for it, as:

- It seems there is a clear interest for the feature (given the time
already spend on it and the number of people that worked on)

- I've in mind to resume working on it

- It would give more visibility in case others want to jump in

Hope that makes sense,

Thanks,

Bertrand