* Dont run most pipelines on nightly cron
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Don't run skipfiles on push to master for arch pipelines
Signed-off-by: Derek Nola <derek.nola@suse.com>
---------
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Include note on service keys
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Fix rotate cert ca test
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Remove periods
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add new test to nightly script
Signed-off-by: Derek Nola <derek.nola@suse.com>
---------
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Remove deprecated nodeSelector label beta.kubernetes.io/os
Problem:
The nodeSelector label beta.kubernetes.io/os in the CoreDNS deployment was deprecated in 1.14 and will likely be removed soon
Solution:
Change the nodeSelector to remove the beta
Signed-off-by: Dan Mills <evilhamsterman@gmail.com>
Add additional kernel config checks for NETFILTER_XT_MATCH_COMMENT and
NETFILTER_XT_MATCH_MULTIPORT as they are both required to run k3s.
Signed-off-by: Richard Steinmetz <richard@steinmetz.cloud>
* add droneignore, make trivial change to README for testing, updating drone config to use droneignore to skip CI when files are all matched
Signed-off-by: matttrach <matttrach@gmail.com>
We need to send the full chain in order for cross-signing to work
properly during switchover to a new root.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
* Wait for kubelet port to be ready before setting
* Wait for kubelet to update the Ready status before reading port
Signed-off-by: Daishan Peng <daishan@acorn.io>
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Co-authored-by: Brad Davidson <brad.davidson@rancher.com>
* Bump wrangler to 1.1.1
* Match golang.org/x/net with flannel version
* Match golang.org/x/sys with containerd version
* Update gax-go to 2.1.1
* Isolate terraform e2e test with seperate go.mod/go.sum
* Bump containerd
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Initial drone vagrant pipeline
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Build e2e test image
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add docker registry to E2E pipeline
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Bump libvirt image
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add ci flag to secretsencryption
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Fix vagrant log on secretsencryption
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Remove DB parallel tests
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Reduce sonobuoy tests even further
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add local build
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add cron conformance pipeline
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Add string output for nodes
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Switch snapshot restore for upgrade cluster
Signed-off-by: Derek Nola <derek.nola@suse.com>
* Fix cp
Signed-off-by: Derek Nola <derek.nola@suse.com>
---------
Signed-off-by: Derek Nola <derek.nola@suse.com>
Turns out etcd-only nodes were never running **any** of the controllers,
so allowing multiple controllers didn't really fix things.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Prevents errors when starting with fail-closed webhooks
Also, use panic instead of Fatalf so that the CloudControllerManager rescue can handle the error
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
