| From: | Petr Jelinek <petr(dot)jelinek(at)2ndquadrant(dot)com> |
|---|---|
| To: | Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>, Jeff Janes <jeff(dot)janes(at)gmail(dot)com> |
| Subject: | Re: logical replication busy-waiting on a lock |
| Date: | 2017-06-09 20:28:00 |
| Message-ID: | f598b4b8-8cd7-0d54-0939-adda763d8c34@2ndquadrant.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 09/06/17 03:20, Petr Jelinek wrote:
> On 09/06/17 01:06, Andres Freund wrote:
>> Hi,
>>
>> On 2017-06-03 04:50:00 +0200, Petr Jelinek wrote:
>>> One thing I don't like is the GetLastLocalTransactionId() that I had to
>>> add because we clear the proc->lxid before we get to AtEOXact_Snapshot()
>>> but I don't have better solution there.
>>
>> I dislike that quite a bit - how about we instead just change the
>> information we keep in exportedSnapshots? I.e. store a pointer to an
>> struct ExportedSnapshot
>> {
>> char *exportedAs;
>> Snapshot snapshot;
>> }
>>
>> Then we can get rid of that relatively ugly list_length() business too.
>>
>
> That sounds reasonable, I'll do that.
>
And here it is, seems better (the 0002 is same as before).
--
Petr Jelinek http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
| Attachment | Content-Type | Size |
|---|---|---|
| v2-0002-Don-t-assign-xid-to-logical-decoding-snapshots.patch | text/x-patch | 3.2 KB |
| v2-0001-Use-virtual-transaction-instead-of-normal-ones-in-ex.patch | text/x-patch | 17.5 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andres Freund | 2017-06-09 20:31:55 | Re: logical replication busy-waiting on a lock |
| Previous Message | Petr Jelinek | 2017-06-09 19:52:50 | Re: walsender & parallelism |