Input for this executor could be empty, so we have to handle
ErrDocumentEntrypointNotDefined error more intelligently.
Change-Id: I73e0f770e64cead7dbbbce2af2b7a2ebd323b83b
Signed-off-by: Ruslan Aliev <raliev@mirantis.com>
Relates-To: #485
Closes: #485
* phase related errors were moved to separate package to prevent
cycling import error;
* error duplicates were removed;
* non-related to package errors were moved to appropriate ones.
Change-Id: If6cfb08c48d356c1517a13157a5fe740dc1635a4
Signed-off-by: Ruslan Aliev <raliev@mirantis.com>
Relates-To: #485
Additionally aligned between Treasuremap and Airchipctl to use the same
environemnt variable "SITE" for a site name.
Change-Id: I50494b50691a40efe68996cd11ccede6517b23d6
Signed-off-by: James Gu <james.gu@att.com>
This change fixes a bug where the default file and directory
permissions weren't getting added (if missing) to the Config
object at runtime.
Change-Id: I87ed2f40af9fe079baade67c506a19de427c4e64
Added 2 phases:
1. secret-reencrypt - This phase can be used to
reecnrypt the existing secrets with new key.
To do so SOPS_IMPORT_PGP must contain
old public key and new private key (but
may also include other data).
SOPS_PGP_FP must contain fingerprint of
new private key.
2. secret-show - This phase may be useful for some users
that need to see what generated by secret-generate phase.
Disabled SOPS debug by default.
To enable it back run commands with
env variable DEBUG_SOPS_GPG=true
Change-Id: Id7fe13d6943d386577df25dba4aaa83e62e58980
This change fixes a bug where the default values that are
populated during an `airshipctl config init` were getting
mixed in with user specified values at runtime. For example,
a user with two contexts defined in their airship config
file would actually see a third phantom context with default
values in the output of `airshipctl config get-context`.
Also fixes a number of the config cmd and pkg unit tests that
were affected by this bug.
Change-Id: Iac591d4c2a616090028eb730576478edefb82544
Closes: #447
Commit fixes CAPD deployment and removes redundant scripts
that check expiration for CAPD site.
They must be tested separately outside CAPD pipeline
Related-To: #482
Closes: #482
Change-Id: I60ffd76a4f3f08bd7bd198a0c2b15483dfbdd6a6
Since we switched to tigera operator to deploy calico, the directory
manifests/composite/infra/replacements became obsolete since it contains
link to cni/calico, which is no longer exists. Also, this directory
isn't used anywhere in manifests, so it can be safely removed.
Signed-off-by: Ruslan Aliev <raliev@mirantis.com>
Relates-To: #492
Closes: #492
Change-Id: I14d9b9ba192aa9546e198300f81045cc035fe13a
The existing recommendation of 60GB disk space is not enough to configure
all-in-one airshitpctl. It is not easy to expand disk space in virtual box
and other envivonrments and developer needs to recreate the environment from
scratch again.
Change-Id: Ifdb260e443647eb4dc942674ee495c9790e28fe6
Now if we run `airshipctl cluster get-kubeconfig` without args
it will return kubeconfig for the entire site, which will have
contexts for every cluster defined in cluster map.
Relates-To: #460
Closes: #460
Change-Id: Icf1f09724a5c60ac520b1dbcbda69c3280ac4c7e
This adds some context e.g. `v1` instead of just `1`, and matches
what some nearby projects use, such as the kpt container functions [0].
[0]: https://github.com/GoogleContainerTools/kpt/issues/1358
Change-Id: I2026fba873a0f074233e7628de35e53f1f880027
Signed-off-by: Sean Eagan <seaneagan1@gmail.com>
Instead of a static list, read list of krm-functions from containing directory.
This mirrors what we do for git SHA tag publishing for these images upon merge.
Before this, the cloud-init krm function was missing from the list.
Change-Id: Iacb5bbcc84c5eeb9a6bb6cebbf674c3068f103b0
Signed-off-by: Sean Eagan <seaneagan1@gmail.com>
Cluster-Api kubeconfig source for target-cluster has higher
priority, airshipctl waits 300 timeout before proceeding
to next source each phase run. This commit will give bundle
higher priority, this will speed up deployments by 5 minutes
per each airshipctl phase run, when ephemeral cluster-api is
unavailable.
Change-Id: I7711d0594a513b7cf9fe3197e90047a9ba96123e
We are hitting docker pull limit on calico images. so switching it from
docker hub to quay
Signed-off-by: Sreejith Punnapuzha <Sreejith.Punnapuzha@outlook.com>
Change-Id: If2ef94885098bd87b2441c6eeb3738076fdaee73
Now if clusterName is specified for kubebuilder, Build() method
will return kubeconfig containing only context for that cluster.
If no clusterName is specified kubeconfig will have contexts of
all clusters that builder can build. This behaviour is similar
to `kubeconfig get` command.
Relates-To: #460
Change-Id: Ib3bebf61bc60430d347ded41b2e31249b11126eb
It is not used anywhere anymore as commands themselves were
deprecated and removed in favor of clusterctl move and init
executors.
Change-Id: I526f75d95b70754034f0846edfc9d3e468648ced
ipam yaml was getting downloaded in customization file. This will cause
problem behind proxy. So adding the ipam yaml to respective folder
Signed-off-by: Sreejith Punnapuzha <Sreejith.Punnapuzha@outlook.com>
Change-Id: Idf0f4b13fb6380363d3cabacfeacc7179f09cff9
Change ClusterMap API object to support multiple kubeconfig sources
for a cluster. If one kubeconfig source fails, kubeconfig builder
will not fail and move on to the next one. This behaviour will allow
to support cases when ephemeral cluster is not accesible anymore or
when target cluster is not yet accessible.
For more information please read issue #460 in airshipctl github
Relates-To: #460
Related-To: #460
Change-Id: I7cd32f78cd7c4ad8814eac357424c24216f40d76
