Commit Graph

7484 Commits

Author SHA1 Message Date
Pierre Souchay 360dc1dd8d Simplified error handling for maxIndexForService
* added unit tests to ensure service index is properly garbage collected
* added Upgrade from Version 1.0.6 to higher section in documentation
2018-03-01 14:09:36 +01:00
Paul Banks 4605d4ed0b
Update CHANGELOG.md 2018-02-28 13:26:08 +00:00
Jack Pearkes 732e0c3bb4
Merge pull request #3922 from hashicorp/docs-fix-two-dc-links
website: override automatic linking of list items for softlayer dc
2018-02-27 12:09:34 -08:00
Jack Pearkes 10f3eb623a website: override automatic linking of list items for softlayer dc
This avoids a conflict with #datacenter later on the page. We're mixing
histroic manually specified anchors with generated anchors (via
redcarpet / middleman-hashicorp) so we have to manually override the
automatic generation here.

I was tempted to rewrite the old manual anchors to use the automatic
generation, but there is no way to maintain backwards compatibility,
so will leave that for a time when it is appropriate for us to break
links (or redirect them, etc).

Fixes #3916
2018-02-27 10:53:12 -08:00
Preetha acfc60523f
Merge pull request #3914 from alvin-huang/fix_vendor
remove old pkgs and put deps of missing packages in vendor.json
2018-02-24 10:01:12 -06:00
Alvin Huang 85c9cfea05 remove old pkgs and put deps of missing packages in vendor.json 2018-02-23 17:08:24 -05:00
Paul Banks 9b97bebc95
Merge pull request #3903 from hashicorp/build-fixes
[WIP] Attempt to find some low-hanging fruit for CI failures
2018-02-23 13:12:45 +00:00
Paul Banks 64fa1ebb59
Merge pull request #3910 from hashicorp/fix-shell
Fix test running in non-bash shells
2018-02-23 13:12:18 +00:00
Kyle Havlovitz 5956e778db
Add a link to the leader election guide in the lock API docs 2018-02-22 15:57:46 -08:00
Kyle Havlovitz 1e11082709
Use GOTAGS in the vet make goal 2018-02-22 15:57:09 -08:00
Paul Banks dbaabb1dbc
Fix test running in non-bash shells 2018-02-22 14:06:06 +00:00
Pierre Souchay 7ff20fc4be Revert "Change .travis.yml, set parallel to 1 to pass tests"
This reverts commit e7d1668347.
2018-02-22 14:16:24 +01:00
Pierre Souchay e7d1668347 Change .travis.yml, set parallel to 1 to pass tests 2018-02-22 09:27:55 +01:00
Pierre Souchay 048db1d033 [Revert] travis tunning as requested by @banks 2018-02-22 08:38:42 +01:00
Paul Banks 6da6e086ef
Merge pull request #3900 from hashicorp/fix-monitor-sigint-3891
Fixes #3891: agent monitor no longer unresponsive before logs stream.
2018-02-21 21:28:33 +00:00
Preetha Appan c99e393b4f
Update CHANGELOG 2018-02-21 13:28:17 -06:00
Preetha 68dc2eee0a
Merge pull request #3909 from hashicorp/b-leaderloop-revokeonerror
Make sure revokeLeadership is called if establishLeadership errors
2018-02-21 13:23:31 -06:00
Preetha Appan 80791d5b21
Remove extra newline 2018-02-21 13:21:47 -06:00
Preetha Appan 907b97b7f2
Unit test that calls revokeLeadership twice to make sure its idempotent 2018-02-21 12:48:53 -06:00
Preetha Appan f59abcc394
Make sure revokeLeadership is called if establishLeadership errors 2018-02-21 12:33:22 -06:00
Alex Dadgar d1ede2c93d
Merge pull request #3908 from hashicorp/f-improved-autopilot-shutdown
Improve autopilot shutdown to be idempotent
2018-02-21 10:19:48 -08:00
Alex Dadgar 18bf9647d5 Test autopilots start/stop idempotency 2018-02-21 10:19:30 -08:00
Paul Banks 7b85f09d59
Travis evaluates ENV before cloning git repo and cding so we need to delay gathering packages until the makefile 2018-02-21 12:54:23 +00:00
Paul Banks 0166b34b5d
Use relative paths as Travis doesn't setup GOPATH right 2018-02-21 12:27:30 +00:00
Paul Banks 41ae19c677
Split the heavy test packages out to their own Jobs. 2018-02-21 12:20:33 +00:00
Paul Banks 6272148d1f
Try parallel packages but not tests 2018-02-21 11:58:51 +00:00
Alex Dadgar 33c5afdb31 Improve autopilot shutdown to be idempotent 2018-02-20 15:51:59 -08:00
Pierre Souchay a8d3745104 Fixed comments for function maxIndexForService 2018-02-20 23:57:28 +01:00
Pierre Souchay 09351ba9a6 [Revert] Only update services if tags are different
This patch did give some better results, but break watches on
the services of a node.

It is possible to apply the same optimization for nodes than
to services (one index per instance), but it would complicate
further the patch.

Let's do it in another PR.
2018-02-20 23:34:42 +01:00
Pierre Souchay 60454b570a Only update services if tags are different 2018-02-20 23:08:04 +01:00
Jack Pearkes 390487866f try enabling sudo in Travis to run builds in GCE
https://docs.travis-ci.com/user/reference/overview/#Virtualisation-Environment-vs-Operating-System
2018-02-20 10:52:51 -08:00
Paul Banks 29feb66fbd
Portability! 2018-02-20 15:48:18 +00:00
Paul Banks 6f266a0103
Only output sparse lines to keep Travis happy while logging verbosely to disk 2018-02-20 15:33:36 +00:00
Paul Banks e416f828dc
Output from test as we run to stop travis killing us and try no parallelism to avoid timing issues on their contended CPUs 2018-02-20 13:16:53 +00:00
Pierre Souchay 5dd77e132f Improve travis parameters to avoid flacky tests, use -p 3 2018-02-20 02:14:58 +01:00
Pierre Souchay 0861b8b1f9 Improve travis parameters to avoid flacky tests 2018-02-20 02:07:01 +01:00
Pierre Souchay a05d38737c Enable Raft index optimization per service name on health endpoint
Had to fix unit test in order to check properly indexes.
2018-02-20 01:35:50 +01:00
Pierre Souchay d15f9d9f8c [Revert] Do not run tests in parallel since it breaks randomly
This is causing timeouts on Travis (more than 10min tests)
2018-02-19 23:34:57 +01:00
Pierre Souchay 99c9ea5905 Do not run tests in parallel since it breaks randomly 2018-02-19 23:04:17 +01:00
Paul Banks de58eb1820
Fixes #3891: agent monitor no longer unresponsive before logs stream.
The root cause is actually that the agent's streaming HTTP API didn't flush until the first log line was found which commonly was pretty soon since the default level is INFO. In cases where there were no logs immediately due to level for instance, the client gets stuck in the HTTP code waiting on a response packet from the server before we enter the loop that checks the shutdown channel from the signal handler.

This fix flushes the initial status immediately on the streaming endpoint which lets the client code get into it's expected state where it's listening for shutdown or log lines.
2018-02-19 21:53:10 +00:00
Pierre Souchay 4f10fae3c3 Get only first service to test whether we have to cleanup index of a service 2018-02-19 22:44:49 +01:00
Pierre Souchay bac8fb046f Fixed comment about raftIndex + use test.Helper() 2018-02-19 19:30:25 +01:00
Pierre Souchay 6cb0846d58 Merge remote-tracking branch 'origin/master' into fix_blocking_queries_index 2018-02-19 18:41:22 +01:00
Pierre Souchay 73127ef407 Services Indexes modified per service instead of using a global Index
This patch improves the watches for services on large cluster:
each service has now its own index, such watches on a specific service
are not modified by changes in the global catalog.

It should improve a lot the performance of tools such as consul-template
or libraries performing watches on very large clusters with many
services/watches.
2018-02-19 18:29:22 +01:00
Paul Banks 4c90fbfd90
Merge pull request #3894 from hashicorp/anubhavmishra-patch-1
fix typo
2018-02-19 14:14:45 +00:00
Anubhav Mishra 0813175340
fix typo 2018-02-18 23:13:57 -08:00
Edd Steel d0f0d67b4a
Clarify comments 2018-02-17 17:46:11 -08:00
Edd Steel f770f360e9 Test every endpoint for OPTIONS/MethodNotFound 2018-02-17 17:34:13 -08:00
Edd Steel c5f0bb3711 Allow endpoints to handle OPTIONS/MethodNotFound themselves 2018-02-17 17:34:03 -08:00
Edd Steel f5af8b0f03
Initialise `allowedMethods` in init() 2018-02-17 17:31:24 -08:00