Age | Commit message (Collapse) | Author | Files | Lines |
|
Add cli emcoctl as a client
utility for EMCO
Issue-ID: MULTICLOUD-1065
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: Ie1951910628469b5a7e75550b9daa34ba377d1a4
|
|
Update plugin_fw.sh test script to also check for proper work of
override parameters functionality of instance API.
Also update instance API response to exclude duplicated
override-parameters entry, that's anyway accessible under
'.request.override-values' json path.
Issue-ID: MULTICLOUD-1176
Signed-off-by: Konrad Bańka <k.banka@samsung.com>
Change-Id: Ie1a336ceb7de1a656f77d4c43ee4775c60cb88fb
|
|
Adding input validations for API based
on Json schemas
Issue-ID: MULTICLOUD-1096
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: If6591bdef6305f87cbce7ef56d894376f687c6c1
|
|
|
|
|
|
Add support in the AppContext for managing an AppContext
(composite app level) status value.
Also adds support for tracking rsync status at the resource
level.
A mechanism for tracking history at the controlling resource
level (i.e. DeploymentGroupIntnt or Cluster) is added, in part,
so that all AppContexts associated can be deleted when
the resource is eventually deleted.
Issue-ID: MULTICLOUD-1042
Change-Id: I3d0a9a97ea45ca11f9f873104476e4b67521e56a
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
|
|
dcm_call_api.sh test script for DCM API is now enough to get /apply
to work correctly out-of-the-box (i.e., all resource quotas are now
in the correct format and apiGroups has been set to "" by default).
Issue-ID: MULTICLOUD-1143
Signed-off-by: Igor D.C <igor.duarte.cardoso@intel.com>
Change-Id: Ia44ec3006409542ddd9e50d2345d2f9eaaad3ad5
|
|
If error in any resource stop processing
and end all goroutines. Also return gRpc
call after starting the goroutine. Adds
retry checks also
Issue-ID: MULTICLOUD-1005
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: I1189e02f0c0426181fdc995a0c4816ceaa64ec7d
|
|
Part two of fixing code format/syntax with Visual Studio Code.
Furthermore, also switched line endings from CRLF to LF (all test
files seemed to be ending this way, unlike the other Go files).
Issue-ID: MULTICLOUD-1143
Signed-off-by: Igor D.C <igor.duarte.cardoso@intel.com>
Change-Id: Iaae868c780a42ea82ca1208e3cbdc3ec6ffd1f97
|
|
|
|
Add a StateInfo structure to the Cluster and
Deployment-Intent-Group resources to keep track of
the lifecycle state of these resources. Moved the
appcontext id that was being kept into this structure
as well. Enabled the approve state (and API) for
the deployment intent group.
Issue-ID: MULTICLOUD-1042
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I36602d8a0658d9d6d37b8799f9a372a7d1042496
|
|
Developed a script which can be used to test a new release and
confirm that the minimum components like clm, instantiation by
orchestrator and rsync are working fine after the code change.
Issue-ID: MULTICLOUD-1174
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I7d519fa88b71fb34d13b7d61f3f8b36edc9fa5f3
|
|
dcm_call_api.sh exercises some DCM API calls.
This patch adds an optional cleanup phase where it
also exercises the API calls to delete resources.
Plus added two extra GET calls for individual clusters
and minor tweaks like set DCM port to 9077 (prevent
conflict with orchestrator).
Issue-ID: MULTICLOUD-1143
Signed-off-by: Igor D.C <igor.duarte.cardoso@intel.com>
Change-Id: I7a00edb5c986b26ec4be739bf8b01ac362abec90
|
|
|
|
Visual Studio Code automatic code formatting was
performed on the 4 modified DCM files.
Issue-ID: MULTICLOUD-1143
Signed-off-by: Igor D.C <igor.duarte.cardoso@intel.com>
Change-Id: I24d2b44181b5c0737df7cb4541f9eede0950211a
|
|
Add functionality to support the PUT api for v2/projects/{project-name}
Also add unit tests for the PUT api
Issue-ID: MULTICLOUD-1130
Change-Id: Ia271569c5f0dec3152952e64171fd5a182aaa333
Signed-off-by: Larry Sachs <larry.j.sachs@intel.com>
|
|
Fixes the build warning 'E004 File did not end with a newline'
Issue-ID: MULTICLOUD-1131
Signed-off-by: Larry Sachs <larry.j.sachs@intel.com>
Change-Id: I0a69fe25f93feb6a700c140c9dae9c214d5d4dd0
|
|
|
|
In this patch, implemented two separate routes for getting all
genericPlacemnetIntents and getting all deployment intents.
Issue-ID: MULTICLOUD-1128
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I81fadb126752788da7c459aa43896fafa583310e
|
|
In this patch implemented a new route to
query all compApps under a project.
Issue-ID: MULTICLOUD-1127
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I5d5d4861539edd2ca0eccff16d94b75439e14db7
|
|
In this patch implement the route: /v2/projects.
Issue-ID: MULTICLOUD-1126
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: Ia38aa560a74c26b8528d6bd579038c1b80b4d3c9
|
|
Handle creation and deletion of the ResourceBundleState
Custom Resource per app/cluster in the rsync microservice
instead of adding it to the resources of the composite
application in the orchestrator service.
Issue-ID: MULTICLOUD-1125
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: Id0c0639ff62e788628e78f972107a7603c27abfb
|
|
Issue-ID: MULTICLOUD-1005
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: Ifca8f939b0d2e8baccb7c97b3a1c857e16b59fa3
|
|
|
|
Changed Rsync to use ordered install.
Changed to use cli-runtime instead of
go-client. Based on code from repo
https://github.com/johandry/klient
Issue-ID: MULTICLOUD-1005
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: I4c2537cb74bd4d24a409cc1f0b7f9ee0875a4e39
|
|
In this patch, identified the root cause
of failure due to triple dash and fixed
it. Also reverted back to original helm charts for
M3db operator.
Issue-ID: MULTICLOUD-1114
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I175cd134cc751bf4b59e72c9c79e0e658a62f348
|
|
Complete the basic terminate support for the
orchestrator and ncm services.
1. When terminate REST API is invoked on a deployment
intent group, call the uninstall grpc api to rsync
and then remove the app context.
2. When terminate REST API is invoked on a cluster,
add the uninstall grpc api call to rsync to remove
the network resources from the clusters.
Issue-ID: MULTICLOUD-1040
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I181e891a8c7c973970af061f9ff07d80c3bb64f9
|
|
This patch provides the basic framework for supporting
monitoring of composite application resources in clusters.
1. Updates to the monitor files for use with v2.
2. Invokes the Watcher process per cluster/app when the
app is instantiated.
3. Adds a ResourceBundleState CR resource to the cluster/app
so that monitor will be able to update status to it.
4. Watcher updates appropriate appcontext status object
when updates are made in clusters by monitor
5. Update appcontext library to define a status handle
and object at the app/cluster level
6. Labels resources with an appropriate tracking label
to coordinate with the ResourceBundleState CR
Issue-ID: MULTICLOUD-1042
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: If007c1fd86ca7a65bb941d1776cfd2d3afed766b
|
|
Status CR gets updated when there is
status update for the resources scheduled
by Rsync. On change to status CR the watch
in the patch will be triggered and the
update can be handled.
Issue-ID: MULTICLOUD-1042
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: I2a46654dca29c05586dbf1e009bec89b0fa75b67
|
|
Removed replace directives that were causing the local build to fail.
Issue-ID: MULTICLOUD-1101
Signed-off-by: Larry Sachs <larry.j.sachs@intel.com>
Change-Id: Ie6604f98a342356af0e477613d5b3840352b6b38
|
|
Signed-off-by: Manjunath Ranganathaiah <manjunath.ranganathaiah@intel.com>
Issue-ID: MULTICLOUD-1005
Change-Id: Iaa8b70f38cf1fc1f89cf2d95fbe37c242fc44f65
|
|
|
|
This patch adds app and resource 'order' and 'dependency'
instructions to the appcontext, as these are currently
expected by rsync.
Adds the rsync client and call to rsync to instantiate the
appcontext.
Issue-ID: MULTICLOUD-1064
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: Iae0da9de4a0ae82bd3ab7ccc72da4abf8b7f2295
|
|
|
|
In this patch, changed the error message for orchestrator
startup without the TLS.
Earlier the message was bit ambiguous.
Issue-ID: MULTICLOUD-1099
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: Idbeefdf5b725f46a52b13ff76981e4dadbe5c86f
|
|
Issue-ID: MULTICLOUD-1005
Signed-off-by: Manjunath Ranganathaiah <manjunath.ranganathaiah@intel.com>
Change-Id: I60e11aaad97b60efc24a02866dc0e580507e5296
|
|
Fix the annotation key value for cnitype - should be 'type'
instead of 'cniType'.
Issue-ID: MULTICLOUD-1094
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I5b12049e9ed114dbc245c5fb219b5a735d8b7e2c
|
|
This operator monitors the status of
resources like pods, services, deployments,
daemonsets, configmaps etc. and updates the
status in the CR accordingly.
Issue-ID: MULTICLOUD-1047
Signed-off-by: Srivahni Chivukula <srivahni.chivukula@intel.com>
Change-Id: I7d92584a44c8add2df69f2985140a55b460ac037
|
|
|
|
listen function doesn't work with ip address.
Binding on all interfaces
Issue-ID: MULTICLOUD-1019
Signed-off-by: Ritu Sood <ritu.sood@intel.com>
Change-Id: I57e1082d8823b349660e0fb207e29ccc9c91a762
|
|
The patch makes grpc calls for context updation
for a given list of controllers and deletes the
extra set of cluster handles for each anyOf
cluster after context updation
Issue-ID: MULTICLOUD-1064
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I4b946f5f130300628ef4f655213639a2444be2cc
|
|
As part of this patch, have implemented
a priorityQueue based contoller list which shall
contain placement contollers and action contollers
listed based on their priority.
Issue-ID: MULTICLOUD-1064
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: Id1beac6e3b6ec5156f64a169042d171ff28e38d6
|
|
Adds a contextupdate client function that can
be invoked for any action or placement controller
to process intents for that controller.
Adds the corresponding server side handling for
the ovnaction controller.
Issue-ID: MULTICLOUD-1019
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: Ifdfc99e522288a530735bc4c1402d45449f6d057
|
|
Split out the cluster provider and cluster APIs
and packages into a separate microservice to align
with the architecture.
Issue-ID: MULTICLOUD-1029
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I08f357b5a488004a2389b72a178ae33e101d1540
|
|
|
|
Issue-ID: MULTICLOUD-1005
Signed-off-by: Manjunath Ranganathaiah <manjunath.ranganathaiah@intel.com>
Change-Id: I90c8ff0d548690ea19d325612d1bcd3e5049dc36
|
|
Reorginize the ncm packges to delineate
- cluster provider and cluster management
- virtual and provider netowrk intent management
- intent scheduler
- internal ovn4k8s network controller
Overall, no code changes, just moving things around.
Issue-ID: MULTICLOUD-1029
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I3d43c7e4eb6f285b51c0385ba18626d3511a14f5
|
|
Split out part of ncm microservice to act as the
Onv4k8s network action controller for the orchestrator.
No code changes really - just moving around to fit the
architectural plan.
Issue-ID: MULTICLOUD-1029
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I17292ac72d041050269f05fc4a0c2a6ca741aeb5
|
|
As part of this patch, we assign groupNumbers
for the set of clusters which are under anyOf, or
in other words are optional for deployement of
app.
Also refactored the instantiation flow by separating
out the etcd interactions
Issue-ID: MULTICLOUD-1064
Signed-off-by: Rajamohan Raj <rajamohan.raj@intel.com>
Change-Id: I21ece189daf6e6b3a7cfdba5df22d57b3d33ca78
|
|
Correct yaml format of Network and ProviderNetwork
CRs produced by ncm so that they can be deployed in
a cluster.
Issue-ID: MULTICLOUD-1072
Signed-off-by: Eric Multanen <eric.w.multanen@intel.com>
Change-Id: I27e55c0c7428ccee6ddb720445e104054eefcee3
|