Closed
Milestone
expired on Dec 22, 2016
Milestone-6
Infrastructure:
-
Kill CephFS completely - Reduce the CI bill
HA prometheusDefault alerting rules
Performance:
- Terminate Git HTTP requests more eagerly
Add Sidekiq supervision tree for scale adjustment-
Flush event caches without updating the database: conclusion: remove caching - Merge request dashboard page takes over a minute to load: improved by https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/7760, but needs more work
-
GitLab (Unicorn) Memory Consumption Issuesclosed in favour of: -
Dashboard labels dropdown times out
- More or less superseded by https://gitlab.com/gitlab-org/gitlab-ce/issues/25504 and https://gitlab.com/gitlab-org/gitlab-ce/issues/25503
- Listing projects takes too much time
Unstarted Issues (open and unassigned)
0
Ongoing Issues (open and assigned)
0
Completed Issues (closed)
18
- Walking Away from CephFS on Azure
- CephFS Turn Down
- Alert for SSL certificate expiration
- Add severity=warn for alerting in prometheus
- Update prometheus to 1.4.0
- Invalid targets in prometheus configuration
- Enable hide-refs in production to reduce git-refs pressure
- Alert for high number of git cat-file processes
- Create alerts if GitLab pages is down
- Create alert if GitLab.com is down
- Build High Available (Federated) prometheus monitoring solution
- Move runner droplets to NYC1 and/or SFO2 data center
- Move LFS storage to its own host
- Stand up small kubernetes cluster
- Creating Default Alerting Rules with Prometheus
- Colocation plan
- Create gitlab-ce/ee shared runners managers
- Prometheus alertmanager - add pagerduty plugin
Loading
Loading
Loading