Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Improve status messages | Jon Bratseth | 2020-12-07 | 1 | -2/+2 |
| | |||||
* | Merge pull request #15682 from vespa-engine/mpolden/detect-clashing-names | Jon Marius Venstad | 2020-12-07 | 2 | -3/+70 |
|\ | | | | | Reject endpoints with clashing non-compactable IDs | ||||
| * | Reject load balancers with clashing non-compactable IDs | Martin Polden | 2020-12-04 | 2 | -3/+70 |
| | | |||||
* | | Start a little more optimistic | Jon Bratseth | 2020-12-04 | 1 | -1/+1 |
| | | |||||
* | | Adaptive scaling window | Jon Bratseth | 2020-12-04 | 10 | -61/+145 |
| | | |||||
* | | Merge pull request #15624 from vespa-engine/mpolden/cleanup | Martin Polden | 2020-12-04 | 2 | -7/+7 |
|\ \ | |/ |/| | Remove redundant compatibility with old format | ||||
| * | Update node-repository tests | Martin Polden | 2020-12-03 | 2 | -7/+7 |
| | | |||||
* | | Merge pull request #15642 from vespa-engine/bratseth/scaling-completion | Jon Bratseth | 2020-12-03 | 17 | -51/+168 |
|\ \ | | | | | | | Record scaling event completion | ||||
| * | | Record scaling event completion | Jon Bratseth | 2020-12-03 | 17 | -51/+168 |
| | | | |||||
* | | | Move definition of permanent flags to separate class | Bjørn Christian Seime | 2020-12-03 | 6 | -25/+25 |
| | | | |||||
* | | | Merge pull request #15638 from ↵ | Håkon Hallingstad | 2020-12-03 | 2 | -100/+2 |
|\ \ \ | | | | | | | | | | | | | | | | | vespa-engine/hakonhall/remove-compact-preprovision-capacity-flag Remove compact-preprovision-capacity flag | ||||
| * | | | Remove compact-preprovision-capacity flag | Håkon Hallingstad | 2020-12-03 | 2 | -100/+2 |
| | |/ | |/| | |||||
* | | | Limit switch rebalancing to production zones | Martin Polden | 2020-12-03 | 1 | -6/+5 |
| |/ |/| | |||||
* | | Simplify | Jon Bratseth | 2020-12-03 | 2 | -31/+3 |
| | | |||||
* | | Explicitly allow unknown generations | Jon Bratseth | 2020-12-03 | 2 | -1/+3 |
| | | |||||
* | | Simplify | Jon Bratseth | 2020-12-03 | 1 | -8/+2 |
| | | |||||
* | | Sort snapshots explicitly | Jon Bratseth | 2020-12-03 | 3 | -30/+32 |
|/ | |||||
* | Allow no status | Jon Bratseth | 2020-12-02 | 1 | -2/+0 |
| | |||||
* | Revert "Merge pull request #15619 from ↵ | Jon Bratseth | 2020-12-02 | 6 | -24/+72 |
| | | | | | | | vespa-engine/revert-15614-bratseth/record-scaling-event-completion" This reverts commit 49ecd29903215b133505f316773631ec9161ff44, reversing changes made to ed58cd5826de9da0ed6a963a35c1246abebac1e4. | ||||
* | Revert "Store the 15 last autoscaling events" | Jon Bratseth | 2020-12-02 | 6 | -72/+24 |
| | |||||
* | Merge pull request #15614 from ↵ | Jon Bratseth | 2020-12-02 | 6 | -24/+72 |
|\ | | | | | | | | | vespa-engine/bratseth/record-scaling-event-completion Store the 15 last autoscaling events | ||||
| * | Store the 15 last autoscaling events | Jon Bratseth | 2020-12-02 | 6 | -24/+72 |
| | | |||||
* | | Merge pull request #15595 from vespa-engine/freva/grace-after-activate | Jon Bratseth | 2020-12-02 | 2 | -2/+35 |
|\ \ | | | | | | | Allow a grace period after node re-activation | ||||
| * | | Allow a grace period after node re-activation | Valerij Fredriksen | 2020-12-02 | 2 | -2/+35 |
| | | | |||||
* | | | Merge pull request #15606 from vespa-engine/bratseth/more-questdb-robustness | Valerij Fredriksen | 2020-12-02 | 1 | -11/+23 |
|\ \ \ | | | | | | | | | Start over if we cannot alter the db | ||||
| * | | | Start over if we cannot alter the db | Jon Bratseth | 2020-12-02 | 1 | -11/+23 |
| | |/ | |/| | |||||
* | | | Merge pull request #15601 from ↵ | Harald Musum | 2020-12-02 | 1 | -1/+1 |
|\ \ \ | | | | | | | | | | | | | | | | | vespa-engine/hmusum/increase-default-provision-lock-timeout Increase default provision lock timeout to 6 minutes | ||||
| * | | | Increase default provision lock timeout to 6 minutes | Harald Musum | 2020-12-02 | 1 | -1/+1 |
| |/ / | |||||
* / / | Report 0 metrics once | Håkon Hallingstad | 2020-12-02 | 1 | -0/+10 |
|/ / | | | | | | | | | | | | | | | | | | | A metric like configserver.lockAttempt.acquireTimedOut.average may be reported as non-zero once, and then not reported for a while. In the meantime, the value reported for this metrics becomes 'last', carried over from earlier periods. This PR ensures that the value is set to 0 the first period after the value becomes 0, but not thereafter (as carrying 0 is fine). | ||||
* / | Alter table when necessary | Jon Bratseth | 2020-12-02 | 3 | -26/+57 |
|/ | |||||
* | Merge pull request #15567 from ↵ | Håkon Hallingstad | 2020-12-01 | 1 | -27/+34 |
|\ | | | | | | | | | vespa-engine/hakonhall/do-not-print-0-fields-in-the-locks-response Do not print 0 fields in the locks response | ||||
| * | Do not print 0 fields in the locks response | Håkon Hallingstad | 2020-12-01 | 1 | -27/+34 |
| | | |||||
* | | Merge pull request #15569 from ↵ | Håkon Hallingstad | 2020-12-01 | 1 | -8/+14 |
|\ \ | | | | | | | | | | | | | vespa-engine/hakonhall/only-report-non-zero-lock-metrics Only report non-zero lock metrics | ||||
| * | | Only report non-zero lock metrics | Håkon Hallingstad | 2020-12-01 | 1 | -8/+14 |
| |/ | | | | | | | This should reduce the number of reported lock metrics to 3%. | ||||
* | | Merge pull request #15525 from vespa-engine/mpolden/stateful-property | Martin Polden | 2020-12-01 | 8 | -38/+35 |
|\ \ | |/ |/| | Set stateful property for relevant clusters | ||||
| * | Use stateful property | Martin Polden | 2020-12-01 | 6 | -35/+32 |
| | | |||||
| * | Write stateful tag | Martin Polden | 2020-12-01 | 2 | -3/+3 |
| | | |||||
* | | Merge pull request #15564 from ↵ | Håkon Hallingstad | 2020-12-01 | 2 | -32/+20 |
|\ \ | | | | | | | | | | | | | vespa-engine/hakonhall/remove-lock-metrics-not-used-in-dashboard Remove lock metrics not used in dashboard | ||||
| * | | Remove lock metrics not used in dashboard | Håkon Hallingstad | 2020-12-01 | 2 | -32/+20 |
| |/ | |||||
* | | Auto-repair on unrecoverable failure | Jon Bratseth | 2020-12-01 | 1 | -15/+43 |
| | | |||||
* | | Close db on deconstruct | Jon Bratseth | 2020-12-01 | 1 | -1/+5 |
|/ | |||||
* | Merge pull request #15546 from ↵ | Håkon Hallingstad | 2020-12-01 | 2 | -16/+107 |
|\ | | | | | | | | | vespa-engine/hakonhall/support-lower-bound-on-number-of-shared-hosts Support lower bound on number of shared hosts | ||||
| * | Support lower bound on number of shared hosts | Håkon Hallingstad | 2020-11-30 | 2 | -16/+107 |
| | | | | | | | | | | | | | | | | | | | | | | | | Adds a 'minCount' field to the shared host jackson flag, denoting the minimum number of "shared hosts" that must exist, otherwise the deficit will be provisioned by DynamicProvisioningMaintainer. A "shared host" is one that is considered for allocation if current tenant node allocations were removed: It must be a tenant host, cannot be an exclusiveTo host, etc. minCount requires the setting of (at least one) shared host. | ||||
* | | Log exception | Jon Bratseth | 2020-11-30 | 1 | -7/+2 |
|/ | |||||
* | Remove dead code | Jon Bratseth | 2020-11-30 | 1 | -8/+0 |
| | |||||
* | Add temporary additional information | Jon Bratseth | 2020-11-30 | 2 | -1/+19 |
| | |||||
* | Add tests | Jon Bratseth | 2020-11-30 | 2 | -3/+42 |
| | |||||
* | Filter at read time | Jon Bratseth | 2020-11-30 | 6 | -12/+25 |
| | |||||
* | Store metrics with inService or stable false | Jon Bratseth | 2020-11-30 | 8 | -23/+74 |
| | |||||
* | Filter unstable at db insert | Jon Bratseth | 2020-11-30 | 12 | -29/+56 |
| |