Skip to content

Latest commit

 

History

History
81 lines (61 loc) · 2.47 KB

CHANGELOG.md

File metadata and controls

81 lines (61 loc) · 2.47 KB
Raw

3.1.1

  • refacto mesh script and config file
  • only override hazelcast.xml, logback.xml, default-distributed-db-config.json and keystore.jceks
  • bump mesh to 1.4.0

3.0.0

  • update healthchecks

3.0.2

  • add admin password management

3.0.0

  • update mesh to 1.3.2

2.3.2

  • add keycloack auth variables

2.3.1

  • update mesh to 1.0.2

2.3.0

  • update mesh to 1.0.1

2.2.2

  • update mesh to 0.41.0

2.2.1

  • update mesh to 0.40.3

2.2.0

  • update mesh to 0.40.2
  • update elasticsearch to 6.8.2

2.1.0

  • add keycloack
  • update ui : better textarea scroll and filter option in create node

2.0.0

  • label refactoring install instructions:

k delete statefulsets.apps RELEASE-NAME-master RELEASE-NAME-replicas --cascade=false k delete deployments RELEASE-NAME-ui RELEASE-NAME-restore RELEASE-NAME-backup-explorer k delete cronjobs.batch RELEASE-NAME-backup-cronjob

  • add metrics : getmesh and elasticsearch (based on justwatch/elasticsearch_exporter:1.0.2)
  • add 2 grafana dashboard: getmesh and getmesh elasticsearch
  • update to getmesh 0.33.1 (getmesh with orientdb 3.0.19)

1.2.0

  • update to getmesh 0.28.3 (latest getmesh with orientdb 2.x)
  • add custom legacy mesh-ui (fix input to textarea for string type edit)
  • more vertx thread to prevent deadlock under certains conditions
  • update backup and restore jobs
  • add orientdb console and various tools (consistency check, elasticsearch reindex, ...)
  • add horizontalPodScaling (optional)
  • add hard affinity between elasticsearch nodes and getmesh nodes
  • add sdd storage to elasticsearch and getmesh nodes
  • schedule getmesh master node under sustainable pool

1.1.1

  • add sleep + file time verification after invoking backup to let time to backup process to be done

1.1.0

  • add backup cronjob, (only when nfs is available), backup are stored to nfs volume
  • add restore pod, start it manually with deployment scaling then kubectl exec -ti into pod and invoke /config/restore.sh
  • add backup explorer eg : https://getmesh-backup-explorer.getmesh.mydomain.com/ so we cant remote dl backup easily

1.0.1

  • update getmesh to 0.28.2
  • update default healthcheck probe url to better one

1.0.0

  • add anygetmesh svc (routing to master + replicas instances)
  • remove uuid in pv / pvc but let an unique name
  • BREAKING CHANGE: remove version from getmesh-master-statefulset
  • add default nginx proxy body size to 8m
  • prepare backup mode

before updating the chart we must manually delete statefulsets k delete statefulsets.apps getmesh-master --cascade=false