Service Fabric

Do you have an idea, suggestion, or feedback based on your experience with Azure Service Fabric? We would love to hear from you.

Please take a few minutes to submit your feedback in one of the forums available on the right or vote up an idea submitted by another Azure Service Fabric customer. All the feedback shared in these forums are being monitored and reviewed by the Microsoft Azure Service Fabric team.

Please remember this forum is for Azure Service Fabric, and if you have feedback on other Microsoft products, services or platforms, please submit your feedback in the appropriate forum (found on the right hand side).

Thanks for your feedback – we look forward to improving your satisfaction with Azure Service Fabric

  1. Custom Actor's Queue Extesibility

    Custom Actor's Queue Extensibility

    the Actor's message queuing is one of the model's fundamentals

    Exposing extensibility point for Actor's queue replacement
    will allow to developers to address scenario that may not supported on
    the current Actor queuing model.

    developers can use it to alter the queue strategy using
    Azure Service Bus or Azure Queue.

    by doing so the queue will become reliable for cluster crash
    or mistakable deployment (without upgrade)

    the current implementation rely on WCF which may
    make it relative easy by replacing the WCF transport channel
    (similar to the Net MSMQ binding)

    the Actor class can use virtual…

    20 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  2. Ability to customize rollout of new versions.

    I want the ability to control how quickly a new versions of services are rolled out and what percent of load gets sent to them.

    For example, I want 10% of the load to hit the new version for 1 hour before it is rolled out to any more nodes.

    93 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    under review  ·  2 comments  ·  Flag idea as inappropriate…  ·  Admin →
  3. Add ability to turn off entire cluster.

    In testing scenarios i would like to be able to turn off and delete the entire cluster from Azure portal.

    8 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    4 comments  ·  Flag idea as inappropriate…  ·  Admin →
  4. The cluster is stuck in "deploying" state

    Cluster is stuck in "deploying" state. There is no way to stop it or delete it. Explorer shows nothing.

    5 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →

    We will need more information inculding traces to diagnose any possible issue here. Is it at the time of creating the cluster, or deploying an application? Can you try resetting the cluster by right-clicking the service fabric explorer taskbar icon?

    Mani Ramaswamy
    Microsoft

  5. Reliable priority queue

    More reliable collections would be helpful. Especially useful would be a reliable priority queue. While this has multiple applications, the most obvious is to prioritize background tasks that need to be performed. Ideally, it should be possible to update priority queue entries, for example because their priority has increased.

    39 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    under review  ·  1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  6. Orchestration tooling for local development and test environments

    One thing I feel lacking from what has been displayed from the current tooling around Azure Service Fabric is a way to orchestrate your services and their dependencies during local development or automated testing.

    The two primary situations (for my own use-case at least) would be:


    • I'm a developer that needs to do some changes in Service Fabric application "Application1". A service within this application calls upon an actor or a service "fabric:/Application2/MyServiceOrActor". If I check out "Application1", perform the changes and deploy it locally to test them, it will fail unless I also know before hand about the dependency…
    30 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    under review  ·  1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  7. Service Fabric Cluster Option for Quick Deploy

    Add an option for Clusters not in production environment or live to have the ability or switch option for rapid deployment. Using rapid deployment would take the cluster down, but updates would happen much faster. The lag and time to deploy slowing is a killer in lower environments, where a goof up can cost you hours for the cluster to update.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  8. UWP client applications need a high-level library for using Service Fabric Reliable Actors (ActorProxy support for UWP clients)

    Azure Service Fabric Reliable Actors seem like a great server-side technology for multi-user applications. Unfortunately, UWP client applications cannot use the use the ActorProxy to call methods in remote actor objects (implementing IActor interfaces) because the Microsoft.ServiceFabric.Actors NuGet package only supports .NET 4.5 -- not the UWP platform.

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  9. Be able to check if an Actor exists in the Service Fabric

    When an Actor ID comes from an API endpoint, calling the Actor might result in creating a new Actor (when it doesn't exist) while this was not needed/wanted. It would be nice to be able to check if an Actor exists.

    I'm not sure how the Actor framework works internally, but I think internal it has an IReliableDictionary so a Key lookup should be easy?

    8 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  10. More visibility into cost when creating/stopping various services

    Provide an alert/notification/warn system that will notify users when cost might still be involved despite stopping or deleting services.

    Example: Service Fabric Cluster creates virtual machines scales sets but when the services is stopped/deleted the actual instances are still running therefore involving costs.
    There is no warning or notification whatsoever right now that can alert user.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  11. All services should be able to be put into a disabled no-cost mode

    There should be an option to disable any service to prevent it incurring a charge without being required to delete that service and all dependant services.

    Example of a problem workflow:
    Set up a test Service Fabric Cluster - 5 minutes
    Wait for it to provision - 15+ minutes
    Upload a service - 5 minutes
    (Time to actually use it)
    Delete all attached VMs, Virtual Networks, Storage Locations, Static IPs, and the cluster itself - 15 minutes of active work because there's no way to group delete things.

    Correct workflow:
    Enable the SFC - 1-5 minutes
    (Time for testing)
    Disable…

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  12. Change node property values at run time

    It would be great if we can change a node property value at runtime so that we can control the placement of a stateful service. We have a unique requirement (may not be unique to us) where when you add a new node to the cluster our stateful services can't function until certain tasks are done which takes about 5-15 mins. Meantime, we would like to avoid stateful services in the new nodes (service movement or new service creation).

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  13. Unable to use the backup and restore service

    I am trying to use the backup and restore service with a windows authenticated cluster. I get an error when I do a post to specify a backup policy. Here is my post request

    $ScheduleInfo = @{

    Interval = 'PT15M'
    
    ScheduleKind = 'FrequencyBased'

    }

    $StorageInfo = @{

    Path = '\\devsvcfabric01\SFBackups'
    
    StorageKind = 'FileShare'

    }

    $BackupPolicy = @{

    Name = 'BackupPolicy1'
    
    MaxIncrementalBackups = 20
    Schedule = $ScheduleInfo
    Storage = $StorageInfo

    }

    $body = (ConvertTo-Json $BackupPolicy)
    $url = "http://localhost:19080/BackupRestore/BackupPolicies/$/Create?api-version=6.2-preview"

    Invoke-WebRequest -Uri $url -Method Post -Body $body -ContentType 'application/json' -UseDefaultCredentials

    Here is the error

    Invoke-WebRequest : {"Error":{"Code":"E_UNEXPECTED","Message":"Object reference not set to…

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  14. Add tooltip showing local time zone

    Show local timezone when hovering over GMT times (e.g. Last Activation Time and Last Successful Activation Time). Right now, it just repeats the same time in the tooltop.

    2 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  15. A F HR HARNRU.FIENDIRN S

    DJICNEHDIT SIL NH EC N DIA BDR NDIHACKERF NDIBERNDHUSKIENCHUFTGMA DHY TSJIW NDIO HR EJ HDRN SIR DENHCII ASEBFL EM DUIBSIBW DRJNDIKNE VI DEN AIRN DEU CHACKERB DR WILLIAM EI B SIRNAHIEM H SINE EE ND H DIC DEK LUDIEH ES NDI HADIENF H SEN CUK RN CUR DEN CUH CIC NDRC BDFIN HUC NEFRU CNH SER BSR.VMJ SE BR EM SE

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    6 comments  ·  Flag idea as inappropriate…  ·  Admin →
  16. Improve cluster configuration repair and recovery

    We've run into an issue with cluster configuration where the security settings made the cluster inoperable. We could not even connect to it. The only solution that worked was to use cleanfabric.ps1
    We propose a 2-fold solution.
    1) Make the test-servicefabricconfiguration and test-servicefabricclustermanifest also ensure the security settings are valid
    2) Allow manual repair of the cluster through RDP, provide script that runs locally not through the rest-based API

    7 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    1 comment  ·  Flag idea as inappropriate…  ·  Admin →
  17. 3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  18. 1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    2 comments  ·  Flag idea as inappropriate…  ·  Admin →
  19. Modify version sorting in Service Fabric Explorer

    The sorting of the version string is currently alphabetical. That means that 1.0.10 is sorted before 1.0.2. Seeing as the versioning system is segmented, not decimal, 1.0.10 should be sorted after 1.0.2

    1 vote
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
  20. Modify Visual Studio 2017 Actor template to generate .Net Standard 2.0 projects for Actor Interfaces

    Currently, if you create an actor via the wizards in Visual Studio 2017, it will create a interfaces library (used for client communication), but it targets the same runtime as the actor. However, one can have a client in a completely different runtime (.Net Framework/.Net Core) and communication should still work. But, you can't reference the library directly in such projects. Targeting .Net Standard 2.0 would fix this issue and encourage better portability practices for those projects.

    3 votes
    Vote
    Sign in
    (thinking…)
    Sign in with: Microsoft
    Signed in as (Sign out)
    You have left! (?) (thinking…)
    0 comments  ·  Flag idea as inappropriate…  ·  Admin →
1 2 5 7 9 14 15
  • Don't see your idea?

Service Fabric

Categories

Feedback and Knowledge Base