I am currently looking into deploying Traefik/Træfik on our service fabric cluster.
Basically I have a setup where I have any number of Applications (services), defined with a tenant name and each of these services is in fact a separate Web UI.
I am trying to figure out if I can configure a single frontend to target a backend so I don't have to define a new frontend each time I deploy a new UI app. Something like
[frontend.tenantui]
rule = "HostRegexp:localhost,{tenantName:[a-z]+}.example.com"
backend = "fabric:/WebApp/{tenantName}"
The idea is to have it such that I can just deploy new UI services without updating the frontend configuration.
I am currently using the Service Fabric provider for my backend services, but I am open to using the file provider or something else if that is required.
Update:
The servicemanifset contains labels, so as to let traefik create backends and frontends.
The labels are defined for one service, lets call it WebUI as an example. Now when I deploy an instance of WebUI it gets a label and traefik understands it.
Then I deploy ANOTHER instance with a DIFFERENT set of parameters, its still the WebUI service and it uses the same manifest, so it gets the same labels, and the same routing. But what I would really want was to let it have a label containing some sort of rule so I could route to the name of the service instance (determine at runtime not design time). Specifically I would like for the runtime part to be part of the domainname (thus the suggestion of a HostRegexp style rule)
I don't think it is possible to use the matched group from the HostRegexp to determine the backend.
A possibility would be to use the Property Manager API to dynamically set the frontend rule for the service instance after creating it. Also, see this for a complete example on using the API.
Related
I try to adjust the template for generating routes within a single namespace.
So basically what openshift does, when I enter a route without settng the host via yaml is generate a route in the following way:
${name}-${namespace}.myapps.mycompany.com
I would prefer to have a base domain for many routes which differs in the path, e.g.:
${namespace}.myapps.mycompany.com/${name}
Is this possible? Especially If I am not an admin of openshift at my company but a dev whose team is responsible just for a few namespaces?
For context: We want to use ArgoCD + Git to use gitops, but do not want to hardcode any infrastructure knowledge like the host or domain in our git repo. We came from using ingresses, but if we omit the host there no routes are generated at all...
Thanks in advance for any help!
You can have path-based routes, e.g., [host]/[path]. If you don't provide your own value for [host], it will use the same OpenShift ${name}-${namespace}.myapps.mycompany.com based values.
I'm not sure that you can change OpenShift's default route template, but you can definitely provide your own path values.
Is there a way to specify a postgresql database name to connect to in the cloud foundry manifest.yml file? I've been raking through the documentation and haven't yet found this specific information.
I'm imagining something like this:
applications:
- name: my-app
routes:
- route: my-app.mybluemix.net
services:
- postgres
dbname: database2
With that approach, a postgresql connection can be made by just the connection string provided by VCAP_SERVICES parsing modules (cfenv in the case of node).
If this is not possible, I will just set a dbname environment variable and build my own connection string.
There is nothing like that in a Cloud Foundry application manifest.yml.
The manifest.yml only takes a list of service instance names and the services with those names will be bound to your app. It does not allow you set other metadata.
https://docs.cloudfoundry.org/devguide/deploy-apps/manifest-attributes.html#services-block
I don't know if these will help, but when you cf bind-service directly there are two additional provisions you can make use of (these are not supported by manifest.yml as of me writing this):
Arbitrary bind parameters. These probably won't help unless your service broker supports them, but it's a way to pass additional info to the service broker. If your broker supported it, you could in theory say give me a database named XYZ by passing it some config this way.
Named service bindings. This provides what amounts to a second name. The intent is that you can create the service with a name of X, but your application can look for a service binding with name Y. You can use this to swap in differently named services, but still expose the same binding name to the application so it will always find the service.
If you are trying to pass in some other service instance related metadata to your application, you'd need to do it some other way. Like if you want to tell it the database name or the connection pool size, etc.. Using environment variables like you mentioned is one option. You could use a config file or cli arguments passed to your application. What you pick is probably a matter of preference/support in the library/framework you're using.
For what it's worth, most service brokers I've seen pass in and tell you a specific database name to use. If the broker said connect to db XYZ and you made your connection to myCoolDb, the connection would fail. Just wanted to mention this. Your mileage may vary.
I have a cluster in Azure AKS with 1 node.
On that cluster I have two back-end services.
Each back-end service is a web app.
I have a domain mydomain.com.
Each app will need to be configured with its own path rule in the ingress object.
Web app 1s (let's call this one the homepage app) target URL needs to be either of the following:
US version of the site: mydomain.com
Swedish version of the site: mydomain.com/se/sv-sv/hem
Any other location/language version of the site: mydomain.com/xx/yy-xx/abcdefgh
Web app 2s (let's call this one the whitepony app) target URL needs to be either of the following:
US version of the site: mydomain.com/us/en-us/whitepony
Swedish version of the site: mydomain.com/se/sv-sv/whitepony
Any other location/language version of the site: mydomain.com/xx/yy-xx/whitepony
(The whitepony apps target path segment is called whitepony regardless of location/language)
Now to my question.
How can I configure these rules in an ingress API object?
Can I use prefixes in the path rules?
Or do I need to use regular expressions?
And what about the special case of the US version of the homepage app, where I'm not using any prefixes/extra URL segments?
Can I use conditions in the ingress object?
Or how would you configure the ingress resource object to meet all the above requirements?
Note that I know and have successfully configured multiple back-end services using path rules in an ingress object.
But without prefixes or extra URL segments.
I won't give you fully working example on how to specify rules in ingress resource to meet your requirements, I would rather like to share with you some hints:
Yes, you will need regular expressions to achieve it, and here is the example of doing it directly with NGINX directives based on example of wordpress multi-language site.
You don't need to define these re-write rules with annotations, you can use for that pure NGINX config style, by supplying appropriate inline NGINX config file inside ConfigMap, here is the example on how to achieve this.
I hope this will help you
Using 5.1.163 version of service fabric run time.
Created a service fabric application with one stateless web api (i.e. using owin communication listener).
Modified the generated code so that listening endpoint to contain partition id/instance id/new_guid (just as is the case for stateful services). This should allow me to create another app instance so that I can have multi-tenancy at application level.
By default, Local.xml file is set to 1 instance for this service.
Deployed it to local machine by F5. Verified that it is deployed to only one instance.
Verified that service is working fine.
Navigated to local service fabric explorer and clicked on the Cluster/Application/AppType node. Clicked on 'Create app instance'.
It successfully created 2nd app instance.
However in this new instance, the service is deployed to all 5 nodes.
I was expecting it deploy the service instance only one node. Is this a bug? But only in this version of service fabric?
When you deploy a Service Fabric application using Visual Studio (or from PowerShell) you use the Deploy-FabricApplication.ps1 that is generated for your application and found in /scripts under your SF project. This script does two things (mainly):
Create/update the application type
Create a new/upgrade existing instance of the application type
The second part there is similar to what you do in the SF Explorer, except this one also considers the publisher profile file you supply. The PS-script actually reads your publisher profile xml files and extracts any parameters in there to a hashset (a dictionary) and passes that as an argument in step 2.
You can create an instance of an SF application type using the PS cmdlets (alternatively you can use FabricClient). The following command does this: New-ServiceFabricApplication. Here you have the chance to supply your own application parameters, including instance count for services in your new application instance (if you have a dynamic parameter for that in your application manifest).
So, when you use the SF explorer to create a new application instance you cannot control how that instance is created, it is always using the default parameter values as specified directly in ApplicationManifest.xml, not values you have specified in your publisher profiles (local1, local5, cloud, etc.).
To controll the creation, run New-ServiceFabricApplication with yor parameters as a hashset.
In my current code base, when i create a VM, DNS name is being dynamically set as same as the instance name. For example, consider if my VM name is "anandInstance", DNS name of the name is being generated as "anandInstance.cloudapp.net". Is there a way to change the DNS name like "dns1.cloudapp.net" during the creation thru REST API??
"Connect to existing VM" , is it possible to achieve this option through REST call? In case "connect to existing.." option , we are getting a list of vms/services to choose and VM is getting created successfully. How to achieve the same using API.
Thanks
In my current code base, when i create a VM, DNS name is being
dynamically set as same as the instance name. For example, consider if
my VM name is "anandInstance", DNS name of the name is being generated
as "anandInstance.cloudapp.net". Is there a way to change the DNS name
like "dns1.cloudapp.net" during the creation thru REST API??
I don't think it is possible. Imagine what a nightmare in the portal would become if you were able to do so? How would you link a Cloud Service (whatever.cloudapp.net) to an actual deployment (MyDemoVm123). However you can use your own domain and have CNAME records pointing to your "want-to-change-for-some-reason.cloudapp.net" (frankly I surely think that soon we will use even longer names)
"Connect to existing VM" , is it possible to achieve this option
through REST call?
Connection to a VM is essentially opening a RDP session. If it a windows VM, you can try using the Download RDP file API call. Once you get the file, just start it with "process.start". If it is linux VM, just start SSH client on port 22 (or one you have defined) from the Cloud Service DNS name you have.
UPDATE
From the azure portal,for stand alone machineoption, we are able to give the dns name with deafult cloudoneapp.net. How to do the same
through the rest api call.any specfic paramter is there to specify the
same?
When you are using the REST API, you first create a Cloud Service (still named hosted service in the REST API) where your machine will be hosted. Here you give the name for that hosted service (the dns name with deafult cloudoneapp.net). Then you call the Create Virtual Machine Deployment API action.
In case "connect to existing.." option , we are getting a list of vms/services to choose and VM is getting created successfully. How to
achieve the same using API.
When you want to get list of all VMs, just get a list of all Hosted Services, then get properties of each and make a guess whether it is a VM or a Cloud Service (maybe by querying for Properties of each service). I don't see a direct access to the list of Virtual Machines. But as this feature being PREVIEW, things might change in the future.
Hope my answer is clear?