43f0407178
Ensure a node can join an existing single node raft cluster
2015-07-23 10:21:23 -06:00
cb718b9832
Only allow adding non-raft nodes for now
...
Adding raft nodes is more complicated and can break a cluster when
moving from 1 to 2 nodes. This will be fixed in later PR.
2015-07-23 10:21:23 -06:00
9db3a779e9
Assign and store a node ID when joining an existing cluster
2015-07-23 10:21:23 -06:00
3f90891b82
Proxy raft and rpc connections to leader transparently
...
Nodes that are not part of the raft cluster will not reliably know who the
current raft cluster leader is. To make communication simpler, proxy all
rpc and raft calls to the current raft leader if a non-leader receives one.
2015-07-23 10:21:23 -06:00
e06f6f421b
Add ability to join an existing cluster
...
This adds some basic ability to join a node to an existing cluster. It
uses a rpc layer to initiate a join request to an existing memeber. The
response indicates whether the joining node should take part in the raft
cluster and who it's peers should be. If raft should not be started, the
peers are the addresses of the current raft members that it should delegate
consensus operations.
To keep the meta store implementation agnostic of whether it's running
a local raft or not, a consensusStrategy type was also added.
2015-07-23 10:21:22 -06:00
abfd438a11
Implement remote meta data fetch
...
This adds some basic plumbing to make remote procedure calls to other cluster
members. This first implementation allows a node to contact the raft leader
and fetch a copy of the meta data. This will be used by non-raft members to
pull down the latest metadata.
2015-07-23 10:21:22 -06:00
641b38f394
Merge pull request #3256 from cannium/remove-unnecessary-timeout
...
Remove unnecessary timeout in WaitForLeader()
2015-07-19 12:01:22 -07:00
f549910a18
Merge pull request #3279 from LK4D4/fix_style_else
...
Fix style issues with else
2015-07-17 11:53:42 -07:00
9ba37325f6
Fixes authorization.
...
Adds GRANT and REVOKE statements for admin privilege. Adds authorization to the query endpoint.
2015-07-17 11:33:06 -07:00
675eacbf2c
Fix style issues with else
...
In go it's better to just continue flow without "else", if it is return in
"if" statement.
Signed-off-by: Alexander Morozov <lk4d4@docker.com>
2015-07-17 11:10:23 -07:00
ee8ba11c4f
Add ON token to SHOW RETENTION POLICIES parser with tests
2015-07-16 13:45:50 -07:00
74cb96646c
Refactor query engine for distributed query support
...
With this change, the query engine code gathers information about
shards and tagsets by working with individual shards, collating the
information, and returning that to the client. It does not assume that any
particular shard is local, and accesses all shards through abstracted
Mappers, of which there are two types -- a Mapper type for Raw queries
and a second type for Aggregate queries. There are corresponding
Executors for each type of Mapper, but both types of Executors share the
same interface.
2015-07-15 12:54:55 -07:00
c41ef86c3c
Remove unnecessary timeout in WaitForLeader()
...
If the timer expires and a leader is still not selected, the local data node will never be created.
2015-07-09 15:00:24 +08:00
bc15a6ea27
Set log prefix for Metastore
...
Log messages from this package do not look the same as all other log
messages.
2015-07-06 12:24:00 -04:00
30b15d3c55
Use salted hashes for in memory password cache
2015-06-30 16:55:30 -06:00
153d18f38c
Add synchronization around authcache accesses
2015-06-30 18:29:36 -04:00
ba6bddc5f8
fix #3102 : fix race in test
2015-06-30 18:29:36 -04:00
3463d906e9
fix #3102 : add unit test for authentication
2015-06-30 18:29:36 -04:00
debc3cc11c
fix #3102 : add authentication cache
2015-06-30 18:29:11 -04:00
b9b80b3d55
Make show stats return error instead of panic
2015-06-29 22:53:02 -06:00
05d660a24d
Merge pull request #3169 from jbcrail/fix-spelling-errors
...
Fix spelling errors in comments and strings.
2015-06-29 12:47:35 -04:00
88e8babc91
Shard supports returned OwnedBy node ID
2015-06-29 12:33:10 -04:00
5fccee3d16
Fix spelling errors in comments and strings.
2015-06-28 02:54:34 -04:00
64ace983e8
Support return of shard groups for timerange
2015-06-26 17:28:52 -04:00
4c194f2c8d
fix #3014 : make code review changes
2015-06-25 22:26:53 -04:00
e2f0814904
fix #3014 : make code review changes
2015-06-25 20:10:17 -04:00
3c7a82d91c
fix #3014 : implement snapshots
2015-06-25 19:54:00 -04:00
90520beb70
fix #2944 : update unit test
2015-06-22 15:40:31 -04:00
0467aa8fec
fix #2944 : don't require WHERE time on create CQ
2015-06-22 15:39:22 -04:00
02b8c6fbf1
Don't use deleted shard groups on pre-creation
...
The way it was, shard groups that were deleted by retention policy
enforcement were being recreated again, just to be deleted in the
next enforcement run. This change will help keep raft log free from
this unnecessary creation and deletion.
2015-06-18 15:20:23 -03:00
01720aa246
Fix excessive shard group creation logging
...
Fixes issue #3029 .
2015-06-17 14:45:47 -07:00
2428edb8a1
Correct skip of test
2015-06-17 11:58:16 -07:00
c73ae8b845
Skip test TestStatementExecutor_ExecuteStatement_ShowGrantsFor
2015-06-17 11:56:48 -07:00
4b332b7348
Enforce minimum retention duration of 1 hour
...
Fixes issue #2991 .
2015-06-17 11:02:07 -07:00
000d6b8b0b
added tests for SHOW GRANTS FOR statements
2015-06-17 01:24:09 +10:00
f133ceb350
Added support for SHOW GRANTS FOR USER syntax
2015-06-17 01:00:26 +10:00
a966482a8e
Ensure default retention policies are fully replicated
...
This commit sets the replication factor to the number of nodes
currently in the cluster.
2015-06-11 14:43:44 -06:00
eaae2a1afb
Don't bother polling if leader already exists
2015-06-11 11:34:52 -07:00
4dff5f48aa
Export WaitForLeader on MetaStore
...
This exported function can then be used by Services and
server-reporting, so those components don't make progress until the
cluster is ready.
2015-06-11 11:10:07 -07:00
431034c8f4
Wait for Raft leader even when cluster exists
2015-06-11 01:23:30 -07:00
b15a0df6d4
Wait for leader before reporting
2015-06-11 00:21:15 -07:00
405ec78f42
Limit cluster to 3 nodes
...
This commit restricts the maximum number of nodes in a cluster to 3.
Fixes #2750
2015-06-11 00:11:47 -06:00
52696ae3a1
Require replication factor to equal cluster size
...
This commit adds a requirement that retention policies must
be fully committed. This means that the replication factor
must be set to whatever the node count is in the cluster.
Fixes #2738
2015-06-11 00:11:03 -06:00
9f91cd7461
Fix missing protobuf data fields
...
This commit adds the following missing fields to the internal.Data:
- Term
- Index
- ClusterID
- MaxNodeID
- MaxShardGroupID
- MaxShardID
2015-06-10 22:48:44 -06:00
586da7f558
Generate cluster id on first node
...
This commit sets a cluster ID when the first node is initialized.
The ID is generated on every CreateNodeCommand so that it can be
applied consistently in the state machine of every server.
2015-06-10 22:33:58 -06:00
bc31783a00
Refactor backup and restore
...
This commit updates the snapshot code as well as the "backup" and
"restore" command to work with the new architecture.
2015-06-10 22:07:01 -06:00
5798d99692
Move shard precreation logic to meta/store
2015-06-10 15:37:00 -07:00
481e1860a0
Remove obsolete shard precreation source
2015-06-10 15:27:38 -07:00
6ce7c6e4ad
Merge pull request #2852 from influxdb/panic_change_rp
...
Don't panic when altering retention policy
2015-06-09 14:16:11 -07:00
c455c374d1
Increase Raft timeouts during test
...
Previous timeouts may be too short for CI testing resources.
2015-06-09 12:32:16 -07:00
21343e4e0f
Don't panic when altering retention policy
...
Fixes issue #2846
2015-06-09 11:39:09 -07:00
4b94a00644
Set auto-created retetion period to infinity
...
Fix issue #2817
2015-06-08 19:47:37 -07:00
cc03030b5e
Add missing call to WaitGroup.Done()
...
Fix issue #2806
2015-06-08 11:10:02 -07:00
38f38a5bd4
Improve AutoCreate camelcase
2015-06-06 17:03:52 -07:00
5fcae1eb04
Improve Go style of retention auto-create
2015-06-06 17:02:32 -07:00
37c11e4c5d
Remove retention auto-create magic values
2015-06-06 17:02:32 -07:00
a64f7edc33
Add config support for retention autocreate
2015-06-06 17:02:32 -07:00
8d9a28a583
Store supports auto-creating retention policy
2015-06-06 17:02:32 -07:00
d3d43246e4
Add raft check to meta store.
2015-06-06 14:24:41 -06:00
5ffb032c43
Invalidate cache on shard owner lookup.
2015-06-05 23:23:10 -06:00
1024965db7
Create shard received from cluster writer
2015-06-05 22:16:51 -06:00
413da5cdc4
add mux.Serve()
2015-06-05 22:13:49 -06:00
4bf024109f
Merge pull request #2779 from influxdb/hook-CQs-back-2733
...
fix #2733 : hook CQs back in
2015-06-05 20:40:31 -07:00
5f49934a23
fix #2733 : trying to fix CI race
2015-06-05 20:49:12 -04:00
6cf78206f0
refactor run.Server to use tcp.Mux
2015-06-05 16:31:58 -06:00
db2d94a6a5
refactor store to use external listeners
2015-06-05 16:08:07 -06:00
b925e1c1af
Multi-node clustering.
...
This commit adds the ability to cluster multiple nodes together to share
the same metadata through raft consensus.
2015-06-05 14:41:19 -06:00
c59a234e7b
Deleted shard groups are not counted as "expired"
2015-06-05 13:08:36 -07:00
d1864a4cdc
Unit test expired and deleted shard groups
2015-06-05 12:31:10 -07:00
5e5f2cd37d
Move expired and deleted shard logic to MetaStore
2015-06-04 22:18:52 -07:00
8600e2e036
Add shard group "DeletedAt" timestamps
2015-06-04 21:46:52 -07:00
fab9adb361
Hook in actual retention service
2015-06-04 17:02:22 -07:00
5b77a0f8e4
MetaStore supports returning leader state
2015-06-04 17:02:22 -07:00
f449d8f687
Remove obsolete retention policy enforcer
2015-06-04 12:22:41 -07:00
4fa9fd5fe9
Merge pull request #2725 from influxdb/jw-client
...
Update client to use line protocol
2015-06-01 17:35:26 -06:00
e400e8f2d6
Use default retention policy if not specified during writes
2015-06-01 17:16:44 -06:00
d962283ae6
adding some integration tests
2015-06-01 16:04:20 -06:00
d581254719
fix tests
2015-06-01 11:59:58 -06:00
3597565955
reading and writing yo!
2015-06-01 11:59:58 -06:00
bf823d9887
Integrating cmd/influxd/run.
2015-05-30 14:06:36 -06:00
f3245ab9ac
Make Authenticate return true if no users and creating a root user.
2015-05-30 14:29:16 -04:00
8c8a55a737
Removed 'failed' from test suite.
2015-05-30 08:59:27 -06:00
4872e6db7c
Move services to subpackage.
2015-05-30 08:20:12 -06:00
6677ea074f
more services, more test fixes, getting closer...
2015-05-29 14:59:57 -06:00
9d4527071e
Refactor run command.
2015-05-29 14:59:57 -06:00
1f294ce8de
Add httpd.Handler.serveQuery() tests.
2015-05-29 14:59:57 -06:00
072dacc1ab
Update tx.go to work with tsdb
2015-05-29 14:59:57 -06:00
df1aeee70a
WIP
2015-05-29 14:56:30 -06:00
75ab63b538
Refactor meta statement execution to meta.StatementExecutor.
2015-05-29 14:54:04 -06:00
736875b858
Integrate meta package.
2015-05-29 14:54:04 -06:00
7258a9be3c
Implement meta.Store and meta.Data.
2015-05-25 16:28:58 -06:00
01d7bfb398
Move coordinator to data package
2015-05-21 10:16:54 -06:00
12e4253e15
Add single-node raft back metastore.
...
This commit adds the meta.RaftStore which can start a single node
cluster and create nodes on the meta.Data internal structure.
Conflicts:
meta/meta.go
2015-05-20 16:49:03 -06:00
0361d2d61a
Move point/shard calculation to ShardGroupInfo
2015-05-20 10:01:29 -06:00
19ece442d8
Add meta.Store accessors.
2015-05-08 11:56:08 -06:00
85e285623e
Refactor meta data to separate package.
...
This commit moves meta data from the influxdb.Server and extracts it
out to a new "meta" package. This package handles all strongly
consistent data in the database and maintains a small set of metadata
in-memory.
Conflicts:
meta/store.go
2015-05-08 11:52:05 -06:00
229638cfaf
Write path interfaces
...
This adds some placeholder interfaces for the new write path.
2015-05-07 23:12:12 -06:00
Jason Wilder
Philip O'Toole
gunnaraasen
Alexander Morozov
Philip O'Toole
Can ZHANG
David Norton
Joseph Crail
Vladimir Lopes
Dejan Golja
Ben Johnson
Todd Persen
Cory LaNou
Paul Dix