I'm trying to import an OpenApi3.0 file to Apisix (from Apisix Dashboard), but I can't get it to work as expected (or probably I'm missing something)
I have the following:
An upstream created and set up
A service created and set up using the previous upstream
I'm trying to import a OpenApi3.0 yml file like this (exactly this):
openapi: 3.0.0
info:
title: REST API
description: ""
version: 1.0.0
paths:
/order:
post:
description: Test description.
x-apisix-service_id: 438901282904736682
labels:
environment: test
The service_id is the id of the service previously created
Then, I went to Routes panel, clicked on Advanced->Import, selected "test" as Task Name and uploaded my file.
What I expect is that when I go to the "Configure" page on the newly created route, it shows my service associated, but this didn't happen as shown below. Not only the service is not being associated, but also my labels and description are blank.
The route generated by the import, without service associated, labels and description
I'm using:
Apisix 3.0.0 (I tried with 2.15 but it's the same)
Apisix Dashboard 2.15 (I also tried with 2.14 but it's the same)
What I'm doing wrong? Or maybe it's a bug with Apisix Dashboard?
As far as I know, the version of the OpenAPI definition that still supports this format is 2.13.x.
In the new version 2.14.0 and 2.15.0, the community has introduced a new data loader implementation, which currently aims to prioritize support in the standard OpenAPI 3 specification (but still falls short) and does not add support for the old APISIX custom fields (x-apisix-xxxx) for now.
If you have a strong need for this, you will have to continue using APISIX Dashboard 2.13.x.
Related
I want to use Datadog's Deployment Tracking but to do so I need to dynamically set the version tag. Right now my backend tag is set in the Dockerfile ENV and LABEL and on the frontend it is just in datadogRum.init(). They are currently hardcoded.
I usually do one deployment per day, I would like these version tags to be auto-incremented but haven't found a good way to do it.
I tried using an environment variable from build tools to update these, but could not find a reliable way to do that. I use CodeFresh for backend building and AWS Amplify for frontend.
I am working on a d365 unified interface sandbox environment on a development project.
This environment was setup recently as a clone of the production d365 instance.
Today I have been adding some plugins and finding a strange issue. I can get the plugin code on record create/update firing no problem (I have pre operation create/update and post operation create/update stages defined and the correct code gets hit for each).
But the C# plugin code does not recognise any of the pre or post images that I have added.
In code when we check IPluginExecutionContext.PostEntityImages it does not contain anything.
Any of the pre existing images that were there already when the environment was cloned are firing correctly. We have a process whereby we name all of our pre and post images the exact same for every entity and I know the ones I have created are named exactly as expected.
In this example I have created a Post Operation stage Update plugin on the OOB opportunity entity with a PreImage defined against it but the code just will not recognise it.
Anyone experienced this before?
TIA
Occasionally the sandbox service seems to fail picking up updates on a plugin assembly. In those cases updating the assembly with a different assembly version (build or revision number) can help.
If not, I would advise to simply remove the complete assembly and recreate it again.
If you do not have an automated deployment process in place, follow these steps:
Create a separate solution.
Add the assembly along with its step registrations and images to the solution.
Export the solution.
Remove the assembly using the plugin registration tool.
Import the solution again.
Is there any way to deploy a particular version of Kubernetes to ACS in Azure?
Using Azure resource manager, or az command.
It doesn't seem like template format for container service shows this info.
You can specify the version in ACS in selected regions. See the template example.
https://github.com/weinong/azure-quickstart-templates/tree/master/101-acs-kubernetes-with-version
We will be updating azure cli with this feature soon.
I suggest you to use acs-engine in this case. It is a tool for you to specify some custom definition like "orchestratorversion" for your case, and it can then generate an ARM template for deploying the k8s cluster.
You could download the acs-engine tool here https://github.com/Azure/acs-engine/releases (choose the version based on your need for the k8s version).
To achieve your goal, you have to provide a json file which the template you could find here https://github.com/Azure/acs-engine/blob/master/examples/kubernetes-releases/kubernetes1.7.json. You could alter the attribute "orchestratorversion" to either 1.5, 1.6 or 1.7 to suit your need. (or maybe 1.8 for the latest version)
When the json file is ready, you could turn it into the ARM template files by typing the following command
.\acs-engine.exe generate kubernetes.json
This will create a new directory called "_output" and you could find the azuredeploy.json and azuredeploy.parameters.json files there.
For more information about the attributes in the json file, take a look at https://github.com/Azure/acs-engine/blob/master/docs/clusterdefinition.md.
Another way you could also try is in the Deployment via Azure CLI specified here. https://github.com/Azure/ACS/tree/master/docs
I am migrating an existing app from teiid-8.8.1c1/jboss-as-7.2.0 to WildFly 9.0.2
Teiid 8.13.4. I have been able to update dependencies in all cases except one:
I have one class getting an Admin instance from org.teiid.adminapi.AdminFactory as follows:
AdminFactory.getInstance().createAdmin(...)
So that is can get and check status on a VDB.
But is seems that the AdminFactory is no longer part the teidd adminapi. Is there some new way to get an admin instance?
Per teiid message board, it was moved to teiid-jboss-admin jar:
[TEIID-3503] Development with Embedded teiid should not require dependency on EAP - JBoss Issue Tracker separated out the EAP dependent Admin functionality into the teiid-jboss-admin jar. It doesn't look like that was called out in the release notes or a specific migration doc, so I'll make sure that gets updated.
I am new to Web Services. I was trying to publish a JAX-WS Web Service to apache juddiv3 using eclipse(indigo) Web Service Explorer. But each time I try to publish it following error is produced:
IWAB0135E An unexpected error has occurred.
UDDIException
Message part {urn:uddi-org:api_v2}get_authToken was not recognized. (Does it exist in service WSDL?)
I am using following components:
(i) Eclipse Indigo (EE version).
(ii) juddi-portal-bundle-3.1.2
(iii) Windows 7
While publishing the web service I am using following entries under the 'actions' section of the eclipse 'Web Service Explorer':
Publish=Business
Publication Format=Simple
Publish URL= "http://localhost:8080/juddiv3/services/publish"
UserID = root
Password = Root
Name = Business1
Many times when I press 'Go' button then 'Please wait. The operation is still in progress' message is also displayed even after a long time.
Kindly help me with this issue.
Kindly suggest if there is any other framework other than juddiv3 which could be used to create private UDDIs on local machines.
Thanks
Its a known issue the eclipse plugin. A bug has been filed.
https://bugs.eclipse.org/bugs/show_bug.cgi?id=307202
FYI, the jUDDI team is working on adding v2 endpoints to the v3 server. This use case is one of the drivers. It's almost in a working state now and will be released with the 3.2 version of jUDDI.