Peter Rifel
045bd4932b
Continue attemps to dump artifacts in `toolbox dump`
...
This way if one of the earlier methods fails, we can still get some artifacts from the later methods
2024-02-28 16:59:10 +01:00
justinsb
2a9343a168
Generate revisions of NLB objects, and introduce cleanup phase
...
This lets us safely make changes to otherwise immutable fields, in
particular for adding security groups to NLBs created without them.
We detect the older versions, and create deletion tasks to remove
them. These tasks can be deferred, and we expect them to be
deferred to a "prune" phase that runs after cluster apply.
Co-authored-by: Ciprian Hacman <ciprian@hakman.dev>
2024-02-17 11:41:15 -05:00
Kubernetes Prow Robot
1fbdc1b9b4
Merge pull request #16308 from jrabbit/patch-1
...
create command: remove example docs say is not implemented yet.
2024-02-13 22:35:41 -08:00
Ciprian Hacman
133f33f779
Update dependencies
2024-02-04 10:18:19 +02:00
Jack Laxson
7d0ed220af
remove example docs say is not implemented yet.
2024-02-01 14:50:48 -05:00
upodroid
4717450b1c
fix dumpling logs for GCE scale tests
2024-01-20 13:33:04 +00:00
Ciprian Hacman
9bef6ef55a
test: Improve cluster deletion defaults
2024-01-08 22:10:01 +02:00
Ciprian Hacman
03c109b74b
Make cluster deletion configurable
2024-01-07 07:15:25 +02:00
Ciprian Hacman
71158230bd
Check if kubeconfig exists before dumping resources
2024-01-03 07:53:51 +02:00
Peter Rifel
8ad0661975
Dump pod logs in parallel
2024-01-01 13:11:25 -06:00
Peter Rifel
db32f982d1
Remove unused cluster name param
...
Signed-off-by: Peter Rifel <pgrifel@gmail.com>
2024-01-01 13:10:37 -06:00
Peter Rifel
8524175f86
Dump k8s resources with kops toolbox dump
...
Signed-off-by: Peter Rifel <pgrifel@gmail.com>
2023-12-30 19:22:53 -08:00
upodroid
3f17147571
add validation interval, cleanup 0 byte log files
2023-12-21 19:48:31 +00:00
Peter Rifel
2956f37e9b
Add --max-nodes flag to toolbox dump, default to 500
2023-12-07 06:30:35 -06:00
upodroid
1a593e4f6f
fix the gce info shown in kops get cmd and set node-tag ginkgo flag
2023-12-03 11:41:45 +00:00
justinsb
d5e55e4f8e
feat: kops toolbox enroll command
...
Implement the enroll command, to add a node to an existing cluster.
This is intended to be used with (experimental) metal support.
2023-12-02 17:48:34 -05:00
justinsb
010a0d5e4c
feat: Support PKI bootstrap
...
Similar to the TPM bootstrapping on GCE (indeed, a lot of the code is
modified from there), but we verify the PKI signature against a public
key in a Host CRD object.
2023-11-30 18:35:58 -05:00
justinsb
592b575412
feat: Support multiple token verifiers in kops-controller
...
This will allow us to support nodes running in multiple clouds. If we
don't configure multiple verifiers, this should be a no-op.
2023-11-30 10:44:53 -05:00
Kubernetes Prow Robot
f73595fe05
Merge pull request #16128 from justinsb/quotes_in_set_2
...
kops: Support quotes in --set flags
2023-11-26 18:59:18 +01:00
justinsb
3ea7af23da
kops: Support quotes in --set flags
...
The pflag library tries to split comma separated values using strict
CSV semantics, which requires quoting of the full value if the value
contains a quote. We switch to LazyQuotes which is much more
tolerant.
2023-11-23 23:16:11 -05:00
Kubernetes Prow Robot
b4d787f390
Merge pull request #16109 from prashantrewar/improve-kops-edit-cluster
...
Improve kops edit command
2023-11-23 19:12:57 +01:00
Ciprian Hacman
464975c53d
Apply suggestions from code review
2023-11-23 19:18:10 +02:00
Prashant Rewar
a39e3f9043
Improve kops edit cluster
...
Signed-off-by: Prashant Rewar <108176843+prashantrewar@users.noreply.github.com>
2023-11-23 21:12:18 +05:30
justinsb
e0622ba3a2
Support writing traces to a directory, wire it up in kubetest2
...
If given a directory, we can construct a reasonable name based on the
executable name, pid and timestamp. Then this is relatively easy to
wire up from kubetest2, if we have an artifacts directory.
2023-11-18 09:12:19 -05:00
justinsb
d8c449a4f8
Basic OpenTelemetry support
...
We initially support capturing to a file (in our own format, as it
doesn't appear a suitable format exists). This means we don't need a
server to capture the traces, and can start capturing through prow
without a lot of infrastructure changes.
Co-authored-by: Peter Rifel <rifelpet@users.noreply.github.com>
2023-11-18 09:12:19 -05:00
Ciprian Hacman
eb5faec456
Keep validating after timeout when cluster is healthy
2023-10-26 04:48:01 +03:00
Ciprian Hacman
ef85e29286
Use dns=none for newly created clusters except for AWS and GCE
2023-10-07 15:46:27 +03:00
Mia-Cross
e12f3dc07b
scaleway: terraform support ( #15892 )
...
* tasks dependencies
* FindAddresses should proceed even if LBID is empty
* refacto instance: handle several instances in IG + volumeSize in model
* refacto lb_backend: set servers' IPs in backend task rather than instance
* dns changes for tf support
* render Scaleway s3 objects + add zone to outputs
* add terraform support to documentation
* new instance unique name should be given according to actual instances names
* avoid untimely changes in tf + tag instance IPs to avoid orphaned IPs at cluster deletion
* update integration tests
* review comments
* add IPs to the resources to be deleted with the cluster
* add task + model for DNS record
2023-09-29 04:22:42 -07:00
Ciprian Hacman
1aecb5304d
aws: Filter ENI list by VPC
2023-09-23 05:25:56 +03:00
justinsb
0096451a83
Cleanup a few places where we used Warningf with %w
...
Fixing up a few places where we used the wrong marker.
2023-09-18 11:02:18 -04:00
Dan Ports
f5871271e6
Make --cloudonly sound less innocuous.
2023-09-14 23:39:07 -04:00
Ciprian Hacman
c754cdf017
kops-controller: Increase client-side throttling limits
2023-09-13 11:25:41 +03:00
upodroid
20e7a5aad4
disable metadata-proxy on k8s version older than 1.29
2023-09-09 16:18:51 +01:00
Kubernetes Prow Robot
09bc180686
Merge pull request #15871 from hakman/remove-docker
...
Remove Docker config option
2023-09-05 15:00:51 -07:00
Kubernetes Prow Robot
14c0b647db
Merge pull request #15874 from johngmyers/no-master-label
...
Legacy control-plane node labels are no longer used
2023-09-04 23:05:49 -07:00
Ciprian Hacman
c43b48a8d8
Remove Docker config option
2023-09-05 07:22:33 +03:00
John Gardiner Myers
76cb3924c4
Legacy control-plane node labels are no longer used
2023-09-04 20:23:08 -07:00
John Gardiner Myers
1ea0fd3004
AWS always uses resource-based names
2023-09-04 16:08:48 -07:00
John Gardiner Myers
b3908e592c
Remove support for Kubernetes 1.23
2023-09-03 16:22:18 -07:00
Ciprian Hacman
6fbc9e58a2
aws: Add option to provide multiple instance types on cluster creation
2023-09-02 17:01:02 +03:00
Kubernetes Prow Robot
62e7d71ae8
Merge pull request #15842 from hakman/update-deps
...
Update dependencies
2023-08-31 09:18:49 -07:00
Ciprian Hacman
5d43d5fe88
Update dependencies
2023-08-31 13:53:46 +03:00
justinsb
cc5d667eb5
Avoid depending on kubectl from kops update-cluster
...
This is helpful if kubectl is not installed.
2023-08-26 07:39:01 -04:00
Ciprian Hacman
35e7bbad85
Apply suggestions from code review
2023-08-26 10:38:36 +03:00
Ciprian Hacman
4e0e28a32b
aws: Create subnets for additional network CIDRs
2023-08-26 10:18:44 +03:00
Kubernetes Prow Robot
99e6909693
Merge pull request #15743 from hakman/mark_deprecated_instead_of_normalizing
...
Mark flags as deprecated instead of normalizing
2023-08-16 06:28:26 -07:00
Kubernetes Prow Robot
ca1dd07fc9
Merge pull request #15726 from johngmyers/asset-acl
...
Don't set object-level public ACL in S3 FileRepository
2023-08-12 20:53:19 -07:00
John Gardiner Myers
0d9c130b07
Remove use of ClusterSpec in nodeup
2023-08-09 18:12:37 -07:00
Ciprian Hacman
01684ac206
Mark flags as deprecated instead of normalizing
2023-08-05 09:39:26 +03:00
John Gardiner Myers
054c3960a5
Don't set object-level public ACL in S3 FileRepository
2023-08-01 04:26:41 -07:00