lp:~mattyw/juju-core/ensure-availability-more-output
- Get this branch:
- bzr branch lp:~mattyw/juju-core/ensure-availability-more-output
Branch merges
Branch information
Recent revisions
- 2752. By Andrew Wilkins
-
[r=axwalk],[bug=1319225] agent/mongo: improve TestEnsureAdminUser
Several improvements in an efford to get
TestEnsureAdminUser to stop failing CI.- Call the BaseSuite's SetUpTest
- Log the error Login
- Mock out the starting of mongod with --noauth.
The existing code actually starts up mongod
again inside the test, with the same port as
the still-running server. This *may* be what
caused issues in CI, but I could never reproduce
it.
- Extend TestEnsureAdminUser to check that mongod
is/isn't started/stopped/ etc. Possibly fixes lp:1319225
(I have been unable to reproduce the failure) - 2751. By Ian Booth
-
[r=wallyworld] Replace LoggingSuite with BaseSuite
A mechanical change - the new BaseSuite is used
in place of LoggingSuite across the codebase. This
provides what LoggingSuite does but also ensures
no outgoing network connections are used and also
ensures the HOME var is cleared.A couple of tests in cmd are left using LoggingSuite
to make it easier to extract this package separate
to juju-core later. - 2750. By Jesse Meek
-
[r=waigani] Update username regex
Update valid username regex to allow '.' and '-' and ensure
username ends in a letter and is at least two characters
long.https:/
/codereview. appspot. com/98400043/ R=menno.smits
- 2749. By Menno Finlay-Smits
-
[r=menno.smits] juju/cmd: introduced "user" command
"juju user" now exists. It is a SuperCommand and the existing
"add-user" command is now wired up as "juju user add" instead.https:/
/codereview. appspot. com/92520043/ R=dave-cheney, waigani
- 2748. By Dave Cheney
-
[r=dave-cheney] Fix build error with Go 1.3
A change that landed in Go 1.3 was an additional restriction that you could not `go install` a package if there was no source to install. Prior to this the results of doing `go install launchpad.
net/juju- core` were in consistent. If you did `go install launchpad. net/juju- core/.. .` then the base package would fail to build, but all the sub packages would build. In Go < 1.3 this was a warning. This failure is not an error and the whole `go install` command will return exit 1. To fix this, as well as have a nice high level description that shows up in godoc.org, add a small file at the top level so it can be built (ie, a .a file is generated) which keeps the tool happy. It is worth noting that our hacked version of the Go 1.2 build tool for ppc64el also suffers from this limitation as it is a mash up of bits of Go 1.2 and Go 1.3's build tool.
nb. Why is there no source to install ? Because the only files in that directory end in _test.go; they are test source, not package source, and test source is never installed.
- 2747. By Andrew Wilkins
-
[r=axwalk],[bug=1319652] worker/
instancepoller: TestBatching less race-y The existing TestBatching test was very prone to
failure due to the nature of the test. The test
ensures that requests to Instances are batched
using a token bucket.Whereas the existing test streamed requests, the
new implementation compresses the addition of
requests into the point where we service an
Instances call. We can thus reliably determine
the number of batches that will be made, as calls
to Instances must complete before additional
requests are gathered.Fixes lp:1319652
- 2746. By Ian Booth
-
[r=wallyworld],[bug=1233601],[bug=1312463] Disallow network access for provider tests
Outgoing network access is disabled for provider
tests. This lead to test failures for the maas
provider which were fixed. The "live" Joyent
provider tests were deleted as they never worked
properly anyway and other issues surfaced.Part of the implementation removed the contained
LoggingSuite struct from the jujutests.LiveTests
and jujutests.Tests structs. The various provider
tests which embed these now contain BaseSuite. - 2745. By Vladislav Klyachin
-
[r=klyachin],[bug=1309444] HA: peergrouper spins in local environment
The peergrouper is intended to monitor the MongoDB replica health and make sure everything is functioning well.
However, when run in the local provider, or in an upgraded environment (while we haven't finished all the upgrade steps), it spams the log file with:INFO juju.worker runner.go:260 start "peergrouper"
ERROR juju.worker.peergrouper worker.go:137 peergrouper loop terminated: cannot get replica set status: cannot get replica set status: not running with --replSet This fix ensure NOT running of peergrouper worker in local environment.
https:/
/codereview. appspot. com/97880045/ R=axwalk, fwereade, jameinel, klyachin
R=axwalk, fwereade, jameinel, klyachin
Branch metadata
- Branch format:
- Branch format 7
- Repository format:
- Bazaar repository format 2a (needs bzr 1.16 or later)
- Stacked on:
- lp:~go-bot/juju-core/trunk