mirror of https://github.com/dapr/docs.git
Merge branch 'v1.9' into fix-2964
This commit is contained in:
greenie-msft
GitHub
commit
ab01a3587f
|
|
@ -34,7 +34,8 @@ The table below shows the versions of Dapr releases that have been tested togeth
|
|||
|
||||
| Release date | Runtime | CLI | SDKs | Dashboard | Status |
|
||||
|--------------------|:--------:|:--------|---------|---------|---------|
|
||||
| December 2nd 2022 | 1.9.5</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.3 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported (current) |
|
||||
| February 1st 2023 | 1.9.6</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.3 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported (current) |
|
||||
| December 2nd 2022 | 1.9.5</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.3 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported |
|
||||
| November 17th 2022 | 1.9.4</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.3 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported |
|
||||
| November 4th 2022 | 1.9.3</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.3 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported |
|
||||
| November 1st 2022 | 1.9.2</br> | 1.9.1 | Java 1.7.0 </br>Go 1.6.0 </br>PHP 1.1.0 </br>Python 1.8.1 </br>.NET 1.9.0 </br>JS 2.4.2 | 0.11.0 | Supported |
|
||||
|
|
|
|||
|
|
@ -28,8 +28,6 @@ spec:
|
|||
value: "false"
|
||||
- name: cleanSession
|
||||
value: "false"
|
||||
- name: backOffMaxRetries
|
||||
value: "0"
|
||||
```
|
||||
|
||||
{{% alert title="Warning" color="warning" %}}
|
||||
|
|
@ -49,6 +47,19 @@ The above example uses secrets as plain strings. It is recommended to use a secr
|
|||
| caCert | Required for using TLS | Certificate Authority (CA) certificate in PEM format for verifying server TLS certificates. | `"-----BEGIN CERTIFICATE-----\n<base64-encoded DER>\n-----END CERTIFICATE-----"`
|
||||
| clientCert | Required for using TLS | TLS client certificate in PEM format. Must be used with `clientKey`. | `"-----BEGIN CERTIFICATE-----\n<base64-encoded DER>\n-----END CERTIFICATE-----"`
|
||||
| clientKey | Required for using TLS | TLS client key in PEM format. Must be used with `clientCert`. Can be `secretKeyRef` to use a secret reference. | `"-----BEGIN RSA PRIVATE KEY-----\n<base64-encoded PKCS8>\n-----END RSA PRIVATE KEY-----"`
|
||||
|
||||
### Enabling message delivery retries
|
||||
|
||||
The MQTT pub/sub component has no built-in support for retry strategies. This means that the sidecar sends a message to the service only once. If the service marks the message as not processed, the message won't be acknowledged back to the broker. Only if broker resends the message, would it would be retried.
|
||||
|
||||
To make Dapr use more spohisticated retry policies, you can apply a [retry resiliency policy]({{< ref "policies.md#retries" >}}) to the MQTT pub/sub component.
|
||||
|
||||
There is a crucial difference between the two ways of retries:
|
||||
|
||||
1. Re-delivery of unacknowledged messages is completely dependent on the broker. Dapr does not guarantee it. Some brokers like [emqx](https://www.emqx.io/), [vernemq](https://vernemq.com/) etc. support it but it not a part of [MQTT3 spec](http://docs.oasis-open.org/mqtt/mqtt/v3.1.1/os/mqtt-v3.1.1-os.html#_Toc398718103).
|
||||
|
||||
2. Using a [retry resiliency policy]({{< ref "policies.md#retries" >}}) makes the same Dapr sidecar retry redelivering the messages. So it is the same Dapr sidecar and the same app receiving the same message.
|
||||
|
||||
### Communication using TLS
|
||||
|
||||
To configure communication using TLS, ensure that the MQTT broker (e.g. mosquitto) is configured to support certificates and provide the `caCert`, `clientCert`, `clientKey` metadata in the component configuration. For example:
|
||||
|
|
@ -70,8 +81,6 @@ spec:
|
|||
value: "false"
|
||||
- name: cleanSession
|
||||
value: "false"
|
||||
- name: backoffMaxRetries
|
||||
value: "0"
|
||||
- name: caCert
|
||||
value: ${{ myLoadedCACert }}
|
||||
- name: clientCert
|
||||
|
|
@ -109,8 +118,6 @@ spec:
|
|||
value: "false"
|
||||
- name: cleanSession
|
||||
value: "true"
|
||||
- name: backoffMaxRetries
|
||||
value: "0"
|
||||
```
|
||||
|
||||
{{% alert title="Warning" color="warning" %}}
|
||||
|
|
|
|||
|
|
@ -32,10 +32,6 @@ spec:
|
|||
value: "default"
|
||||
- name: persistent
|
||||
value: "true"
|
||||
- name: backOffPolicy
|
||||
value: "constant"
|
||||
- name: backOffMaxRetries
|
||||
value: "-1"
|
||||
- name: disableBatching
|
||||
value: "false"
|
||||
```
|
||||
|
|
@ -49,20 +45,16 @@ spec:
|
|||
| token | N | Enable Authentication. | [How to create pulsar token](https://pulsar.apache.org/docs/en/security-jwt/#generate-tokens)|
|
||||
| tenant | N | The topic tenant within the instance. Tenants are essential to multi-tenancy in Pulsar, and spread across clusters. Default: `"public"` | `"public"` |
|
||||
| namespace | N | The administrative unit of the topic, which acts as a grouping mechanism for related topics. Default: `"default"` | `"default"`
|
||||
| persistent | N | Pulsar supports two kind of topics: [persistent](https://pulsar.apache.org/docs/en/concepts-architecture-overview#persistent-storage) and [non-persistent](https://pulsar.apache.org/docs/en/concepts-messaging/#non-persistent-topics). With persistent topics, all messages are durably persisted on disks (if the broker is not standalone, messages are durably persisted on multiple disks), whereas data for non-persistent topics is not persisted to storage disks. Note: the default retry behavior is to retry until it succeeds, so when you use a non-persistent theme, you can reduce or prohibit retries by defining `backOffMaxRetries` to `0`. Default: `"true"` | `"true"`, `"false"`
|
||||
| backOffPolicy | N | Retry policy, `"constant"` is a backoff policy that always returns the same backoff delay. `"exponential"` is a backoff policy that increases the backoff period for each retry attempt using a randomization function that grows exponentially. Defaults to `"constant"`. | `constant`、`exponential` |
|
||||
| backOffDuration | N | The fixed interval only takes effect when the `backOffPolicy` is `"constant"`. There are two valid formats, one is the fraction with a unit suffix format, and the other is the pure digital format that is processed as milliseconds. Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h". Defaults to `"5s"`. | `"5s"`、`"5000"` |
|
||||
| backOffInitialInterval | N | The backoff initial interval on retry. Only takes effect when the `backOffPolicy` is `"exponential"`. There are two valid formats, one is the fraction with a unit suffix format, and the other is the pure digital format that is processed as milliseconds. Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h". Defaults to `"500"` | `"50"` |
|
||||
| backOffMaxInterval | N | The backoff initial interval on retry. Only takes effect when the `backOffPolicy` is `"exponential"`. There are two valid formats, one is the fraction with a unit suffix format, and the other is the pure digital format that is processed as milliseconds. Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h". Defaults to `"60s"` | `"60000"` |
|
||||
| backOffMaxRetries | N | The maximum number of retries to process the message before returning an error. Defaults to `"0"` which means the component will not retry processing the message. `"-1"` will retry indefinitely until the message is processed or the application is shutdown. Any positive number is treated as the maximum retry count. | `"3"` |
|
||||
| backOffRandomizationFactor | N | Randomization factor, between 1 and 0, including 0 but not 1. Randomized interval = RetryInterval * (1 ± backOffRandomizationFactor). Defaults to `"0.5"`. | `"0.5"` |
|
||||
| backOffMultiplier | N | Backoff multiplier for the policy. Increments the interval by multiplying it with the multiplier. Defaults to `"1.5"` | `"1.5"` |
|
||||
| backOffMaxElapsedTime | N | After MaxElapsedTime the ExponentialBackOff returns Stop. There are two valid formats, one is the fraction with a unit suffix format, and the other is the pure digital format that is processed as milliseconds. Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h". Defaults to `"15m"` | `"15m"` |
|
||||
| persistent | N | Pulsar supports two kinds of topics: [persistent](https://pulsar.apache.org/docs/en/concepts-architecture-overview#persistent-storage) and [non-persistent](https://pulsar.apache.org/docs/en/concepts-messaging/#non-persistent-topics). With persistent topics, all messages are durably persisted on disks (if the broker is not standalone, messages are durably persisted on multiple disks), whereas data for non-persistent topics is not persisted to storage disks.
|
||||
| disableBatching | N | disable batching.When batching enabled default batch delay is set to 10 ms and default batch size is 1000 messages,Setting `disableBatching: true` will make the producer to send messages individually. Default: `"false"` | `"true"`, `"false"`|
|
||||
| batchingMaxPublishDelay | N | batchingMaxPublishDelay set the time period within which the messages sent will be batched,if batch messages are enabled. If set to a non zero value, messages will be queued until this time interval or batchingMaxMessages (see below) or batchingMaxSize (see below). There are two valid formats, one is the fraction with a unit suffix format, and the other is the pure digital format that is processed as milliseconds. Valid time units are "ns", "us" (or "µs"), "ms", "s", "m", "h". Default: `"10ms"` | `"10ms"`, `"10"`|
|
||||
| batchingMaxMessages | N | batchingMaxMessages set the maximum number of messages permitted in a batch.If set to a value greater than 1, messages will be queued until this threshold is reached or batchingMaxSize (see below) has been reached or the batch interval has elapsed. Default: `"1000"` | `"1000"`|
|
||||
| batchingMaxSize | N | batchingMaxSize sets the maximum number of bytes permitted in a batch. If set to a value greater than 1, messages will be queued until this threshold is reached or batchingMaxMessages (see above) has been reached or the batch interval has elapsed. Default: `"128KB"` | `"131072"`|
|
||||
|
||||
### Enabling message delivery retries
|
||||
|
||||
The Pulsar pub/sub component has no built-in support for retry strategies. This means that sidecar sends a message to the service only once and is not retried in case of failures. To make Dapr use more spohisticated retry policies, you can apply a [retry resiliency policy]({{< ref "policies.md#retries" >}}) to the MQTT pub/sub component. Note that it will be the same Dapr sidecar retrying the redelivery the message to the same app instance and not other instances.
|
||||
|
||||
### Delay queue
|
||||
|
||||
When invoking the Pulsar pub/sub, it's possible to provide an optional delay queue by using the `metadata` query parameters in the request url.
|
||||
|
|
|
|||
Loading…
Reference in New Issue