|From:||Victor Wagner <vitus(at)wagner(dot)pp(dot)ru>|
|Subject:||Re: Patch: Implement failover on libpq connect level.|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
On Tue, 6 Sep 2016 07:58:28 +0530
Mithun Cy <mithun(dot)cy(at)enterprisedb(dot)com> wrote:
> Now if only one host is in connection string and it ask for read_write
> connection(connect to master) I mean read_only is set 0 explicitly.
> With above logic we will allow it to connect to standby?. I still
> think psql connection to standby should be handled by changing the
> default value of readonly to 1 (which means connect to any).
It would definitely be more logical, but I haven't found easy way to do
it. May be I should return to this place and rethink. I don't like and
idea to explicitely ignore connection string parameter due to some
Really, I think, test for replication connection can be either
removed from this part of code now.
> further probably replacing readonly parameter with
> targetServerType=any|master (with default being any) should clear
> some confusions and bring consistency since same is used in JDBC
> multi host connection string.
It seems that in this context change readonly=0|1 to
targetServerType=any|master makes sense.
> @@ -1460,33 +1538,80 @@ connectDBStart(PGconn *conn)
> (int) (UNIXSOCK_PATH_BUFLEN - 1));
> conn->options_valid = false;
> + free(nodes->port);
> nodes->port was not allocated at this point.
I'll recheck it.
Victor Wagner <vitus(at)wagner(dot)pp(dot)ru>
|Next Message||Vladimir Sitnikov||2016-09-07 09:09:38||Re: 9.4.1210 release|
|Previous Message||Mithun Cy||2016-09-06 02:28:28||Re: Patch: Implement failover on libpq connect level.|
|Next Message||Abhijit Menon-Sen||2016-09-06 05:18:15||Re: Proposal for changes to recovery.conf API|
|Previous Message||Craig Ringer||2016-09-06 04:54:31||Re: patch: function xmltable|