Miscellaneous fixes for --disable-monitor mode #486
Conversation
--run will be ignored anyway, so prohibit it to prevent confusion.
There was a problem hiding this comment.
Looks good to me, thanks! We might want to act on the comment I've left, and maybe you had other edits you wanted to add to this PR before we merge it?
About the Travis failure, I have another context in which I've seen it and I will prepare a separate PR that fixes the problem, I believe. Also we need more tests around synchronous_standby_names on the monitor and on the primary in the vein of what you've contributed already.
See #488 |
In this case, there's no monitor to talk to.
goalState is argv[0]. Also fix the usage documentation to include `id` for the four-argument variant of the call.
jchampio
force-pushed
the
dev/misc-disable-monitor
branch
from
b16e44f to
6213e59
Compare
Nope, this is it for now.
The Travis failure is #481, I think. (The issue you've filed may also be contributing, but the test should probably be corrected as well.) |
Okay I'll merge it as soon as #488 is merged then!
Well from a quick reading it seems to me it's the same thing, the issue #481 should be fixed with the PR #488 ; can you have a look and confirm I'm not overlooking at some things? |
|
See also #494. |
|
@DimCitus FYI, the squash commit that was pushed to |
Here are the first few things I ran into while investigating #483. See commit messages for details.