This repository has been archived by the owner on Feb 9, 2024. It is now read-only.
Releases: gravitational/gravity
Releases · gravitational/gravity
Release 6.3.18
6.3.18 (June 1st, 2020)
Bugfixes
- Update Kubernetes to 1.17.6 (CVE-2020-8555) (#1631)(planet#655)
- Bump etcd to v3.3.22 (#1617)(planet#649)
- Prevent etcd portion of upgrades from taking excessive amount of time (#1617)
- Update CNI plugins to 0.8.6 (CVE-2020-10749) (#1617)(planet#649)
- Fix several issues that prevented expand operations from being resumable (#1594)
Warning
- This release fixes a security vulnerability in kubernetes. Please see Kubernetes Announcement for more information.
- This release fixes a security vulnerability in CNI. Please see Kubernetes Announcement for more information.
Release 6.1.28
6.1.28 (June 1st, 2020)
Improvements
- Add gravity rollback command (#1620)
- Add planet election change events to
gravity status history
(#1355)
Bugfixes
- Update Kubernetes to 1.15.12 (CVE-2020-8555) (#1627)(planet#656)
- Bump etcd to v3.3.22 (#1618)(planet#648)
- Prevent etcd portion of upgrades from taking excessive amount of time (#1618)
- Update CNI plugins to 0.8.6 (CVE-2020-10749) (#1618)(planet#648)
- Fix several issues that prevented expand operations from being resumable (#1605)
- Use consistent naming to prevent issues with re-running phases (#1601)
Warning
- This release fixes a security vulnerability in kubernetes. Please see Kubernetes Announcement for more information.
- This release fixes a security vulnerability in CNI. Please see Kubernetes Announcement for more information.
Release 5.5.47
5.5.47 LTS (May 29th, 2020)
Improvements
- Upgrade etcd to
v3.3.22
(#1619, planet#650). - Upgrade CNI plugins to
v0.8.6
(#1619, planet#650).
Bugfixes
Release 5.5.46
5.5.46 LTS (May 21st, 2020)
Improvements
- Add
gravity rollback
command to support automatic plan rollback of failed upgrade operations (#1558). - Add profiling endpoint to planet-agent (#1579, planet#569, satellite#165).
- Bump Grafana to
5.4.5
and Telegraf to1.11.5
(#1593, monitoring-app#168). - Update monitoring containers to run as non-root and remove unnecessary binaries (#1593, monitoring-app#168).
Bugfixes
- Re-enable nethealth checker as a warning and fix a number of issues related to removing stale data for deleted nodes (#1580, planet#637, monitoring-app#166).
- Increase image pull deadline on kubelet to 15m (#1584, planet#640).
Release 5.5.45
Release 7.0.5
7.0.5 (May 13th, 2020)
Improvements
- Add Teleport nodes status to
gravity status
(#1487). - Make the hook that installs a system DNS application idempotent (#1514).
Bugfixes
- Make upgrade operation more tolerant to Kubernetes version skew policy (#1458).
- Make sure Teleport node has connected successfully when joining a new node (#1487).
- Fix dates displayed in the UI to use standard US format MM/DD/YYYY (#1491).
- Fix an issue with inability to set a
role_map
property when configuring a trusted cluster (#1527). - Loosen preflight check to allow variance in OS patch versions between cluster nodes (#1530).
- Fix an issue with bogus expand operations sometimes appearing after multi-node installs (#1537).
- Fix issues with empty Prometheus metrics that prevented
kubectl top
and HPA from working (#1543, monitoring-app#163). - Update default trusted cluster role mapping to only map remote admin role to local admin role (#1546).
- Fix an issue with kube-controller-manager getting unauthorized errors after changing the node advertise address (#1548).
- Fix an issue with Teleport node not being able to join after changing the node advertise address (#1548).
Release 6.3.17
Release 6.1.27
Release 6.1.26
Release 5.5.44
5.5.44 LTS (May 13th, 2020)
Improvements
- Add Teleport nodes status to
gravity status
(#1477). - Add ability to display warning probes in
gravity status
(#1498).
Bugfixes
- Make sure Teleport node has connected successfully when joining a new node (#1477).
- Fix a number of issues that may have led to crashes when trying to resume operations (#1479, #1526).
- Fix an issue with inability to set a
role_map
property when configuring a trusted cluster (#1556). - Update default trusted cluster role mapping to only map remote admin role to local admin role (#1556).