Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    estibent10
    @estibent10
    Hi! right now I can use Dapr for production environment? Dapr is ready for Apps on production? Thanks!
    2 replies
    aktxyz
    @aktxyz
    looking thru the actor api specs ... is there a way to delete an actor?
    https://github.com/dapr/docs/blob/master/reference/api/actors_api.md#deactivate-actor
    answering my own question from above ... reminders are definitely per actor instance (actorid)
    Artur Souza
    @artursouza
    @aktxyz The actor instance is deactivated automatically after some time (idle time). No need to actively invoke deactivate.
    aktxyz
    @aktxyz
    @artursouza thanks ... more asking about truly deleting ... ie removing from the state store and removing any saved timers/reminders
    meant to say removing any saved reminders (timers don't get persisted !)
    3 replies
    clintsinger
    @clintsinger
    I have a service that starts trying to send messages to actors before the Dapr runtime is ready so starts throwing a bunch of exceptions. Is there a mechanism to tell if dapr is available and ready before I send anything to it?
    2 replies
    Artur Souza
    @artursouza
    @aktxyz This behavior can be implemented as a method in your actor class.
    Artur Souza
    @artursouza
    @clintsinger You can create a method to ping the health endpoint for the Dapr's sidecar. There are two issues related to this: dapr/dapr#1933 and dapr/dapr#1493
    Amit Hansda
    @amitHansda
    Hi, Anyone using DaprClient dotnet sdk?
    I am facing issues while trying to Publisher messages since this morning? Have not faced same previous week.
    Can anyone confirm on the same?
    3 replies
    aktxyz
    @aktxyz
    is there anything to help with deactivating actors when a memory threshold is reached? or is this left up to the process hosting the actors?
    another interesting tidbit, may be working as intended ... but when I remove all state for an actor, it's entry in the state store (I am using postgres) does get removed ... but the reminder entries in the key="actors||<actor-name>" state entry are still there for the now cleared out actors ... this is probably as intended since a reminder should fire for any actor, not just an actor that happens to persist some state ... but was neat to figure out nonetheless :)
    Gokhan Altinoren
    @altinoren
    I'm trying to understand if Dapr to Dapr calling across namespaces is still an open issue or not: dapr/dapr#1275
    Is this implemented, or is there an issue that I can follow this? Ingress to Dapr requires a Dapr sidecar on Ingress gateway, but that also means they need to be in the same namespace, which is not possible based on our design.
    1 reply
    Yaron Schneider
    @yaron2
    Hey everyone, community call will be starting in a few minutes: https://us02web.zoom.us/j/85305980190. Password: eWRhSklVTjJjSnhTaURDcFZaU2ZzQT09
    aktxyz
    @aktxyz
    +1 on the community calls ... learned a lot ... will have to go back and watch the earlier ones !
    Eduardo Monteiro de Barros
    @eduardomb08

    Hey everyone, community call will be starting in a few minutes: https://us02web.zoom.us/j/85305980190. Password: eWRhSklVTjJjSnhTaURDcFZaU2ZzQT09

    Is there a recurrent calendar event for it?

    Mark Chmarny
    @mchmarny
    If you are using our go SDK we would like to hear from you. We've updated the client and serving (callback) portions recently and would like to hear about your experience. How are our docs? What's missing feature-wise? How can we make things easier? We can chat here, do hangouts/teams/zoom session, or just DM me (@mchmarny) on Twitter.
    aktxyz
    @aktxyz
    are there any rules/limits for dapr actor keys ... I tried to use a key with a slash in it "123/ABC" ... I got the vague exception "ServerErrorNoMeaningFulResponse" ...
    with a slightly less vague inner exception "'N' is an invalid start of a value. Path: $ | LineNumber: 0 | BytePositionInLine: 0."
    this is using the dapr dotnet-sdk
    aktxyz
    @aktxyz
    seems to just like not like /'s in the actor id string, maybe related to an internal format for the id that uses slashes ?
    https://github.com/dapr/dotnet-sdk/blob/900727f35187bb887de9a79b5b9df18f51ec9f67/src/Dapr.Actors/Constants.cs#L32
    Nkosinathi Sangweni
    @malotho
    level=fatal msg="failed to start internal gRPC server: error from authenticator CreateSignedWorkloadCert: error from sentry SignCertificate: rpc error: code = Unavailable desc = connection error: desc = \"transport: authentication handshake failed: x509: certificate signed by unknown authority (possibly because of \\"x509: ECDSA verification failure\\" while trying to verify candidate authority certificate \\"cluster.local\\")\"" app_id=leadinboundsvc instance=leadinboundsvcapp-6f6697f466-6dlbr scope=dapr.runtime type=log ver=0.10.0
    We are getting the error since upgrading to v0.10.0
    We are running on Dapr on AKS
    Nkosinathi Sangweni
    @malotho
    I have verified that we get this error only if we upgrade Dapr, but if we unisntall and reinstall the deployment works.
    2 replies
    Santiago Molina
    @smolina74
    hi, I'm trying to use uppercase middleware (middleware.http.uppercase) in an echo service example (from Learning Dapr book). I'm adding a pipeline.yaml file in the root of the project like this:
    `apiVersion: dapr.io/v1alpha1
    kind: Configuration
    metadata:
    name: pipeline
    spec:
    httpPipeline:
    handlers:
        - type: middleware.http.uppercase
          name: uppercase`
    and running the example like this: dapr run --app-id rust-web --app-port 8088 --port 8080 --config ./pipeline.yaml cargo run
    the echo service works perfect except for the uppercase middleware not doing anything. The echo response is not converted to uppercase. Am I missing something in the configuration?
    (I also have a components folder in the project where I configured another yaml called uppercase.yaml with the following info: apiVersion: dapr.iop/v1alpha1 kind: Component metadata: name: uppercase spec: type: middleware.http.uppercase
    2 replies
    Any idea?
    Edward.Chan
    @EdwardChange4
    Hi , guys . I wonder if the actor can subscribe topic ?
    aktxyz
    @aktxyz
    @NullEntity has not been very confusing to me ... and am quite familiar with both
    aktxyz
    @aktxyz
    if I have multiple containers running a set of actors ... and then I want to spin one container down ... is there a way to mark a container/node/process so that the dapr runtime will stop sending actor requests to it?
    3 replies
    Davide Bedin
    @dabedin

    I am attempting to save state to a Cosmos DB with strong consistency and first-write-wins concurrency, which should be supported for this store.
    The code:

    Console.WriteLine($"ETag {state.ETag}");
    var options = new StateOptions() {Concurrency = ConcurrencyMode.FirstWrite, Consistency = ConsistencyMode.Strong};
    await state.SaveAsync(options);

    I receive the ERR_STATE_SAVE error from Dapr

    Grpc.Core.RpcException: Status(StatusCode=Unknown, Detail="ERR_STATE_SAVE: BadRequest, Value 'strong' specified for the header 'x-ms-consistency-level' is invalid.

    I am missing any configuration?

    4 replies
    aktxyz
    @aktxyz
    is there an easy way to flip between running against the latest tag/release (0.10.0) and master ? on windows ... wanting to try things out on master before opening any issues/questions
    4 replies
    aktxyz
    @aktxyz

    I am doing to local mode testing with 4 processes (3 apps each with 1 actor ... and 1 client)

    start 3 dapr apps, each hosts a single actor type
    dapr dashboard (shows all 3 apps propertly)

    dapr run --dapr-http-port 3500 --app-id Xa0 --app-port 5000 ./bin/Debug/netcoreapp3.1/Test.Dapr.Xa0.exe
    dapr run --dapr-http-port 3501 --app-id Xa1 --app-port 5001 ./bin/Debug/netcoreapp3.1/Test.Dapr.Xa1.exe
    dapr run --dapr-http-port 3502 --app-id Xa2 --app-port 5002 ./bin/Debug/netcoreapp3.1/Test.Dapr.Xa2.exe

    run a dapr client, this connects to the default port (3500) and attempts to create an actor in each app and call a method

    dotnet run --no-build

    about 50% of the time, at least one of the actor calls will result in "context deadline exceeded"

    any thoughts on how to get more info to figure out what is going on?

    aktxyz
    @aktxyz
    if I don't start one of the apps, I get this error "error finding address for actor type XaTest1 with id 0", which makes total sense because the app that hosts the actor is not running ... and also makes me think the "context deadline exceeded" message is a comms issue of some sort
    aktxyz
    @aktxyz
    I have a feeling there is some sort of daprd.exe intercommunication via grpc that is failing or perhaps just timing out
    Artur Souza
    @artursouza
    @aktxyz Dapr does not use GRPC for Actor communication (yet).
    aktxyz
    @aktxyz
    @artursouza thanks! is grpc used in daprd <---> daprd communication?
    Yaron Schneider
    @yaron2
    @aktxyz yes, and thr context deadline exceeded occurs because there was a connection timeout (probably) so indeed looks like a comms issue.
    aktxyz
    @aktxyz
    @yaron2 thanks, going to try in wsl2/ubunto ... see if things behave any better there :)
    aktxyz
    @aktxyz
    I am still getting the "context deadline exceeded" error in my multi-app setup on wsl/ubuntu ... then I tried restarting the containers (placement/redit/zipkin) ... turns out that restarting the placement container before starting everything up ... very reliably solves the issue
    aktxyz
    @aktxyz

    running the current (and master) placement service ... I am seeing these errors in the placement output when the "context deadline exceeded" issue occurs

    ERRO[0049] error updating host on lock operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge
    ERRO[0049] error updating host on lock operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge
    ERRO[0049] error updating host on update operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge
    ERRO[0049] error updating host on update operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge
    ERRO[0049] error updating host on unlock operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge
    ERRO[0049] error updating host on unlock operation: rpc error: code = Unavailable desc = transport is closing  instance=DESKTOP-S1BHL3F scope=dapr.placement type=log ver=edge

    will just add a restart of the placement container to my run script for now

    clintsinger
    @clintsinger
    I'm trying to follow the instructions to set up on a kubernetes cluster and I'm running into an issue following the helm instructions. helm repo add dapr https://daprio.azurecr.io/helm/v1/repo errors with Error: looks like "https://daprio.azurecr.io/helm/v1/repo" is not a valid chart repository or cannot be reached: error unmarshaling JSON: while decoding JSON: json: unknown field "acrMetadata". Anyone else getting that issue too?
    Gerwim
    @gerwim

    Hey guys! I've started with Dapr yesterday and got it working with two apps on my local development machine. However, I can not get it working on Kubernetes. The sidecar starts and restarts and the logs don't tell me anything meaningful. Anyone has a clue on why it's in this restart loop?

    Logs:

    time="2020-09-22T18:41:30.206767322Z" level=info msg="starting Dapr Runtime -- version 0.10.0 -- commit 6032dc2" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.206823944Z" level=info msg="log level set to: info" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.207100519Z" level=info msg="metrics server started on :9090/" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.metrics type=log ver=0.10.0 
    time="2020-09-22T18:41:30.207775089Z" level=info msg="loading default configuration" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.207828164Z" level=info msg="kubernetes mode configured" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.207837849Z" level=info msg="app id: web" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.20785422Z" level=info msg="mTLS enabled. creating sidecar authenticator" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0 
    time="2020-09-22T18:41:30.208101852Z" level=info msg="trust anchors and cert chain extracted successfully" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime.security type=log ver=0.10.0 
    time="2020-09-22T18:41:30.208114517Z" level=info msg="authenticator created" app_id=web instance=admintst-78f6f8cc45-xvmvd scope=dapr.runtime type=log ver=0.10.0
    @clintsinger I have not, but since I have other issues with my deployment, I will try a helm deployment and see if that changes anything and will let you know ;-)
    Gerwim
    @gerwim
    @clintsinger I have succesfully deployed the Helm chart. Are you sure you are using v3?
    But I still have aforementioned issue :-/
    clintsinger
    @clintsinger
    @gerwim I'm using v3 version.BuildInfo{Version:"v3.3.3", GitCommit:"55e3ca022e40fe200fbc855938995f40b2a68ce0", GitTreeState:"clean", GoVersion:"go1.14.9"}