Sun Jul 6 12:48:59 PDT 2014
- Previous message: [Slony1-hackers] initial setsync value
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Wasn't that precisely one of our TODO items? I actually thought we did this for 2.2 already, but the code shows otherwise. Jan -- Jan Wieck Senior Software Engineer http://slony.info On Jul 6, 2014 3:05 PM, "Steve Singer" <ssinger at ca.afilias.info> wrote: > In the copy_set function in the slon remote_worker when we copy the set > from the origin we construct values for the initial snapshot (for > sl_setsync) by the following method > > * Get the max SYNC event on the origin and snapshot information for that > sync > * Query all sl_log_1 and sl_log_2 entires that came in after that sync. > This information populates the action list. > > Why don't we just get the snapshot visibility information of what is and > isn't visible to the copy_set transaction directly with > txid_current_snapshot() ? > > It seems simpler? (I would guess that this is a left-over from the > pre-8.3 days when txid_current_snapshot wasn't available?) > > I ask because the current method work with logical replication because > sl_log will be empty (but we still need the snapshot of the copy_set to > filter stuff out later). I'd like to modify copy_set to just use > txid_current_snapshot for both logical and trigger based cases. > > If there some reason why this won't work then please let me know. > > Thanks > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.slony.info/pipermail/slony1-hackers/attachments/20140706/1ebbbf8d/attachment.htm
- Previous message: [Slony1-hackers] initial setsync value
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-hackers mailing list