summaryrefslogtreecommitdiffstats
path: root/node-repository
Commit message (Collapse)AuthorAgeFilesLines
* Avoid unnecessary fixation of manual clockHåkon Hallingstad2020-07-301-2/+1
|
* Fix flaky NodeMetricsDbTestHåkon Hallingstad2020-07-301-1/+5
| | | | | | | | | | | ManualClock is initialized internally to Instant.now(). If this instant has a zero subsecond part db.getWindow(clock.instant().minus(Duration.ofHours(30)), Resource.cpu, List.of("host0")).measurementCount() returns 30, while otherwise returns 29. It's hard to say what is correct when falling exactly on this boundary, so this PR avoids the whole problem by ensuring the boundary is not hit (by adding an arbitrary 1 minute).
* Log load balancer state transitionMartin Polden2020-07-242-9/+18
|
* Take config server lock in node-repositoryMartin Polden2020-07-223-12/+63
|
* Inline unnecessary methodsMartin Polden2020-07-2111-29/+19
|
* Measure consecutive maintenance failuresMartin Polden2020-07-211-7/+5
| | | | | | Measuring time since last success results in a wide range of acceptable values, due to maintenance intervals varying from seconds to as long as half a day. Measure consecutive failures instead, to simplify alerting thresholds.
* Always fail out hosts when exceeding acceptable fail countMartin Polden2020-07-171-16/+11
|
* Emit QoS metric for all maintainersMartin Polden2020-07-1632-102/+154
|
* Control maintenance jobs with feature flagMartin Polden2020-07-0919-69/+67
|
* Merge pull request #13779 from vespa-engine/bratseth/healt-check-improvementsJon Bratseth2020-07-021-1/+0
|\ | | | | Bratseth/healt check improvements
| * Improvements to handling of cluster removalJon Bratseth2020-07-021-1/+0
| | | | | | | | | | | | | | | | | | - Don't change health status to "initializing" when creating a new VipStatus, as 'initializing' now requires all clusters to be up to transition to 'up', which means that if we're already up but are missing a cluster we'll go from 'up' to 'initializing' and stay there. - Forget up/down status for removed clusters. - Nicer logging on ignorable reconfiguration errors.
* | isDockerHost -> isHostJon Bratseth2020-07-0113-23/+26
|/
* exclude stream-ranking-rhel7 from periodic maintkkraune2020-06-301-0/+1
|
* Continue on exceptionJon Bratseth2020-06-291-0/+4
|
* Merge pull request #13682 from vespa-engine/mpolden/remove-flagMartin Polden2020-06-252-5/+1
|\ | | | | Remove configserver-provision-lb flag
| * Remove configserver-provision-lb flagMartin Polden2020-06-242-5/+1
| |
* | make tenant node quota adjustable by feature flag (#13697)Andreas Eriksen2020-06-251-1/+7
| |
* | Merge pull request #13689 from ↵Jon Bratseth2020-06-242-11/+7
|\ \ | | | | | | | | | | | | vespa-engine/bratseth/use-the-right-cloud-condition Bratseth/use the right cloud condition
| * | Nonfunctional, changes onlyJon Bratseth2020-06-241-1/+1
| | |
| * | Use smaller defaults when sharing hostsJon Bratseth2020-06-241-10/+6
| |/
* / Set spare count 0 where we can provision on demandValerij Fredriksen2020-06-241-2/+1
|/
* Merge pull request #13679 from vespa-engine/bratseth/prefer-larger-hostsJon Bratseth2020-06-242-1/+38
|\ | | | | Prefer hosts at least twice the size of the node
| * Prefer hosts at least twice the size of the nodeJon Bratseth2020-06-242-1/+38
| |
* | Revert "Revert "From 0 to 1 spare""Jon Bratseth2020-06-2317-30/+76
|/
* Revert "From 0 to 1 spare"Håkon Hallingstad2020-06-2317-76/+30
|
* Merge pull request #13676 from vespa-engine/bratseth/log-on-mitigationValerij Fredriksen2020-06-231-10/+16
|\ | | | | Log on mitigation
| * Log on mitigationJon Bratseth2020-06-231-10/+16
| |
* | From 0 to 1 spareJon Bratseth2020-06-2317-30/+76
|/
* Skip empty clustersJon Bratseth2020-06-231-2/+3
|
* Enable configserver-provision-lb by defaultMartin Polden2020-06-221-2/+2
|
* Provision load balancer for controller clusterMartin Polden2020-06-193-4/+25
|
* Merge pull request #13619 from vespa-engine/bratseth/spare-capacity-maintainerJon Bratseth2020-06-1839-301/+1057
|\ | | | | Bratseth/spare capacity maintainer
| * Update ↵Jon Bratseth2020-06-181-1/+1
| | | | | | | | | | node-repository/src/main/java/com/yahoo/vespa/hosted/provision/provisioning/HostCapacity.java Co-authored-by: Valerij Fredriksen <freva@users.noreply.github.com>
| * No easy finding active nodesJon Bratseth2020-06-173-7/+2
| |
| * Use bitCountJon Bratseth2020-06-171-10/+1
| |
| * Update ↵Jon Bratseth2020-06-171-1/+1
| | | | | | | | | | node-repository/src/main/java/com/yahoo/vespa/hosted/provision/maintenance/MaintenanceDeployment.java Co-authored-by: Valerij Fredriksen <freva@users.noreply.github.com>
| * Less max iterations as each do more exporationJon Bratseth2020-06-172-7/+4
| |
| * Limit by iterations instead of depthJon Bratseth2020-06-172-54/+68
| |
| * MemoizeJon Bratseth2020-06-173-3/+120
| |
| * CleanupJon Bratseth2020-06-171-5/+1
| |
| * More testsJon Bratseth2020-06-172-7/+20
| |
| * More testsJon Bratseth2020-06-173-5/+87
| |
| * Test SpareCapacityMaintainerJon Bratseth2020-06-1711-68/+368
| |
| * SpareCapacityMaintainer sketchJon Bratseth2020-06-1231-122/+300
| |
| * Refactor - no funcntional changesJon Bratseth2020-06-104-73/+116
| |
| * Nonfunctional changes onlyJon Bratseth2020-06-102-3/+4
| |
| * Spare capacity maintainenance skeletonJon Bratseth2020-06-099-93/+122
| |
* | Provision load balancer for config server clusterMartin Polden2020-06-116-31/+79
| |
* | Extract LoadBalancerSpecMartin Polden2020-06-119-25/+64
| |
* | Remove NodeRepository from LoadBalancerService interfaceMartin Polden2020-06-1113-28/+24
| |