On 14 November 2016 at 07:01, Craig Ringer <cr...@2ndquadrant.com> wrote: > Every test I write with the TAP framework for recovery seems to need > to wait for one node to catch up to another or examine replication > slot state. So: attached is a patch to add helper methods for these > tasks. > > Also add a new pg_lsn.pm helper class to deal with PostgreSQL's > awkward representation of LSNs in perl. Made extra fun by our use of > Perl 5.8.8 with no new modules, so we can't rely on having 64-bit > integers. Provides sensible LSN comparisons. I'll be using this in > coming patches that have to make assertions about differences between > LSNs, and I think it's sensible to have anyway. Incorporates tests for > the class. > > Finally, add some coverage of physical replication slots to recovery tests. > > Backpatch to 9.6 desirable, since we seem to be doing that for TAP > infrastructure. > > These three are related enough, and all only touch the TAP framework, > so I've bundled them in a series.
Good to see more tests going in. Bit confused... can't see a caller for wait_for_slot_catchup() and the slot tests don't call it AFAICS. Also can't see anywhere the LSN stuff is used either. No specific problems with the code, just don't want to commit code that isn't used anywhere, yet. I want to commit the extra tests soon, as a stronger test for my recovery.conf changes. -- Simon Riggs http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
