diff options
| author | Tom Lane <tgl@sss.pgh.pa.us> | 2016-06-19 13:45:03 -0400 |
|---|---|---|
| committer | Tom Lane <tgl@sss.pgh.pa.us> | 2016-06-19 13:45:03 -0400 |
| commit | d56c02f1ac9247295ec1e5143361020a91f708c8 (patch) | |
| tree | de0ee469d408b7c970bff69e7d9e824859c4e878 /src/backend/executor/nodeModifyTable.c | |
| parent | b9b925f83fbbed9ae7dd60d94dcd19eb6f6d2bcd (diff) | |
| download | postgresql-d56c02f1ac9247295ec1e5143361020a91f708c8.tar.gz postgresql-d56c02f1ac9247295ec1e5143361020a91f708c8.zip | |
Revert "Fix "pg_ctl start -w" to test child process status directly."
This reverts commit c869a7d5b44e7164fadfb638786def05d510312a.
As pointed out by Maksym Sobolyev in bug #14199, that approach doesn't
work if the postmaster forks itself an extra time due to silent_mode
being enabled. We removed silent_mode in 9.2, so the pg_ctl change is
fine in 9.2 and later, but it fails when that option is enabled in 9.1.
Seeing that 9.1 is close to end-of-life, let's adopt the most conservative
fix we can, which is to revert the pg_ctl change in the 9.1 branch.
Discussion: <20160618042812.5798.85609@wrigleys.postgresql.org>
Diffstat (limited to 'src/backend/executor/nodeModifyTable.c')
0 files changed, 0 insertions, 0 deletions