Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | | Update log message | Tor Brede Vekterli | 2018-04-25 | 1 | -1/+1 | |
| | | ||||||
* | | ZooKeeper-persist and load published cluster state bundles | Tor Brede Vekterli | 2018-04-24 | 23 | -125/+716 | |
|/ | | | | | | | | | | | | | | | | Store synchronously upon each new versioned state, load whenever controller is elected master. Effectively carries over visible node states from one controller's lifetime to the next. This removes the edge case where default bucket space content nodes are marked as in Maintainence until their global merge status is known. To avoid controller tripping over its own feet, state bundles are now _not_ versioned at all until the initial send time period has passed. This prevents overwriting the state persisted from a previous controller with a transient state where all nodes are down due to not having Slobrok contact yet. A new cluster state recompute+send edge has been added when the master passes its initial state send time period. | |||||
* | Add metric for node availbility ratio | Tor Brede Vekterli | 2018-04-12 | 3 | -6/+62 | |
| | | | | | Separate node type dimensions are used for distributors and storage nodes | |||||
* | Add missing test for existing node state metrics | Tor Brede Vekterli | 2018-04-12 | 2 | -0/+131 | |
| | ||||||
* | Fix grammar in Event message | Tor Brede Vekterli | 2018-04-06 | 2 | -2/+2 | |
| | ||||||
* | Add information about derived bucket space state in cluster state history. | Geir Storli | 2018-03-23 | 4 | -21/+71 | |
| | ||||||
* | Add buckets total to per bucket space stats on html status page. | Geir Storli | 2018-03-20 | 3 | -13/+33 | |
| | ||||||
* | Add config to clustercontroller for the min merge completion ratio of ↵ | Geir Storli | 2018-03-20 | 11 | -38/+93 | |
| | | | | buckets in a bucket space before it is considered complete. | |||||
* | Fix ClusterStatsChangeTracker to handle case where current node is not found ↵ | Geir Storli | 2018-03-19 | 3 | -6/+15 | |
| | | | | | | in previous state. Cluster stats may have changed in this case. | |||||
* | Extend cluster v2 to output the published distribution state. | Geir Storli | 2018-03-15 | 13 | -34/+92 | |
| | | | | This is the baseline cluster state + per bucket space states. | |||||
* | Remove never used per storage node ops latencies in host info. | Geir Storli | 2018-03-13 | 3 | -69/+4 | |
| | ||||||
* | Merge pull request #5269 from ↵ | Geir Storli | 2018-03-09 | 5 | -70/+104 | |
|\ | | | | | | | | | vespa-engine/geirst/fix-cluster-stats-change-tracker-in-clustercontroller Fix ClusterStatsChangeTracker to keep previous 'may have merges pendi… | |||||
| * | Fix ClusterStatsChangeTracker to keep previous 'may have merges pending' ↵ | Geir Storli | 2018-03-08 | 5 | -70/+104 | |
| | | | | | | | | state per content node. | |||||
* | | More descriptive test names | Tor Brede Vekterli | 2018-03-08 | 1 | -5/+5 | |
| | | ||||||
* | | Update constraint mock variable naming | Tor Brede Vekterli | 2018-03-08 | 1 | -4/+4 | |
| | | ||||||
* | | Must use published state from default bucket space, not baseline | Tor Brede Vekterli | 2018-03-08 | 1 | -1/+7 | |
| | | ||||||
* | | Unify style | Tor Brede Vekterli | 2018-03-08 | 1 | -6/+9 | |
| | | ||||||
* | | Add a simple maintenance state transition constraint for nodes in default space | Tor Brede Vekterli | 2018-03-08 | 6 | -14/+133 | |
|/ | | | | | | | | Avoids transitioning an already up/init/retired node to maintenance mode when global merges are pending, but allows transitions when the node is considered down/maintenance in the already published state. Not stateful, so triggers false positives on cluster controller restart edges. | |||||
* | Simplifications. | Geir Storli | 2018-03-07 | 2 | -7/+3 | |
| | ||||||
* | Fix AggregatedStatsMergePendingChecker to take state of aggregated cluster ↵ | Geir Storli | 2018-03-07 | 9 | -85/+134 | |
| | | | | | | | | | | stats into account. We may have merges pending if: - we don't yet have updates from all distributors - we don't have stats for the content node in question Also move mayHaveMergesPendingInGlobalSpace() to AggregatedStatsMergePendingChecker. | |||||
* | Also list derived bucket space states in html state. | Geir Storli | 2018-03-07 | 2 | -3/+13 | |
| | ||||||
* | Add bucket space column to html state. | Geir Storli | 2018-03-07 | 1 | -6/+20 | |
| | ||||||
* | Use computeIfAbsent() instead of explicit get() and put(). | Geir Storli | 2018-03-06 | 1 | -6/+1 | |
| | ||||||
* | Add factory functions for NodeEvent and improve some function names. | Geir Storli | 2018-03-06 | 8 | -35/+42 | |
| | ||||||
* | Extend EventDiffCalculator to emit events for changes in derived bucket ↵ | Geir Storli | 2018-03-02 | 8 | -80/+324 | |
| | | | | space cluster states. | |||||
* | Add node state reason for nodes that may have buckets pending. | Geir Storli | 2018-03-02 | 3 | -13/+65 | |
| | ||||||
* | Also use AnnotatedClusterState for derived bucket space states in ↵ | Geir Storli | 2018-03-02 | 9 | -53/+71 | |
| | | | | ClusterStateBundle. | |||||
* | Merge pull request #5177 from ↵ | Tor Brede Vekterli | 2018-03-01 | 10 | -55/+187 | |
|\ | | | | | | | | | vespa-engine/geirst/add-buckets-pending-stats-on-clustercontroller-status-page Geirst/add buckets pending stats on clustercontroller status page | |||||
| * | Simplifications. | Geir Storli | 2018-03-01 | 4 | -6/+10 | |
| | | ||||||
| * | Add per node buckets pending stats on html status page. | Geir Storli | 2018-02-28 | 6 | -34/+93 | |
| | | ||||||
| * | Add function to get aggregated node stats per distributor. | Geir Storli | 2018-02-28 | 4 | -21/+90 | |
| | | ||||||
* | | Merge pull request #5164 from ↵ | Geir Storli | 2018-02-28 | 21 | -92/+490 | |
|\ \ | |/ |/| | | | | | vespa-engine/vekterli/add-v3-cluster-state-propagation-rpc Add v3 RPC for sending cluster states to content nodes | |||||
| * | Add class comments | Tor Brede Vekterli | 2018-02-28 | 5 | -0/+32 | |
| | | ||||||
| * | Factor out common RPC version downgrade code | Tor Brede Vekterli | 2018-02-27 | 1 | -12/+11 | |
| | | ||||||
| * | Add v3 RPC for sending cluster states to content nodes | Tor Brede Vekterli | 2018-02-27 | 21 | -84/+451 | |
| | | | | | | | | | | | | | | | | | | Supports sending states for multiple bucket spaces at once, with possibilities for extending protocol later for sending distribution config etc. Payload is LZ4-compressed when possible. If a receiver node does not understand the v3 protocol, the protocol version is transparently downgraded to legacy v2 RPC. | |||||
* | | Add switch to enable multiple bucket spaces in cluster controller. | Geir Storli | 2018-02-27 | 4 | -2/+11 | |
|/ | ||||||
* | Rename ClusterStateChangeTracker -> ClusterStatsChangeTracker. | Geir Storli | 2018-02-27 | 3 | -22/+22 | |
| | ||||||
* | Rename hasMergesPending() -> mayHaveMergesPending() followup. | Geir Storli | 2018-02-27 | 1 | -6/+6 | |
| | ||||||
* | Rename hasMergesPending() -> mayHaveMergesPending(). | Geir Storli | 2018-02-27 | 6 | -10/+10 | |
| | ||||||
* | Rename ClusterStatsBucketsPendingState -> ClusterStateChangeTracker. | Geir Storli | 2018-02-27 | 3 | -13/+12 | |
| | ||||||
* | Wire buckets pending state into state version tracker. | Geir Storli | 2018-02-27 | 4 | -2/+65 | |
| | ||||||
* | Remove unused function parameter. | Geir Storli | 2018-02-27 | 4 | -11/+8 | |
| | ||||||
* | Add class tracking whether we have changes in buckets pending state in the ↵ | Geir Storli | 2018-02-27 | 4 | -3/+171 | |
| | | | | 'global' bucket space. | |||||
* | Let BucketSpaceStats track invalid count to allow transition from invalid -> ↵ | Geir Storli | 2018-02-27 | 4 | -19/+52 | |
| | | | | valid. | |||||
* | Add class checking whether a bucket space on a content node might have ↵ | Geir Storli | 2018-02-27 | 5 | -51/+161 | |
| | | | | buckets pending. | |||||
* | Track whether BucketSpaceStats is valid or not. | Geir Storli | 2018-02-27 | 3 | -3/+64 | |
| | ||||||
* | Nonfunctional changes only | Jon Bratseth | 2018-02-26 | 3 | -165/+24 | |
| | ||||||
* | Remove never used distributor-put-latency metrics from restapiv2. | Geir Storli | 2018-02-22 | 13 | -327/+0 | |
| | ||||||
* | Initial support for per bucket space cluster states in cluster controller | Tor Brede Vekterli | 2018-02-22 | 21 | -102/+757 | |
| | | | | | Multiple spaces are only enabled in tests, so the controller still only generates the legacy baseline cluster state, maintaining today's behavior. | |||||
* | Cleanup import statements. | Geir Storli | 2018-02-20 | 1 | -1/+0 | |
| |