rolling update

Author: thomas-tacquet

faq/serverless-containers.mdx

@@ -26,6 +26,21 @@ Yes. Because Serverless Containers supports any containerized application, you c
 
 With serverless, you only pay for the compute resources you actually use. There are no upfront provisioning costs or paying for idle capacity. When your application traffic is low, the cost scales down, and when traffic spikes, the platform automatically scales up, ensuring you never overpay for unused resources.
 
+## Does updates of Serverless Containers can cause downtime?
+
+No, when deploying a new version of your Serverless Container generates a **rolling update**. This means that a new version of the service is gradually
+rolled out to your users without downtime. Here is how it works:
+
+* When a new version of your container is deployed, the platform automatically starts routing traffic to the new version incrementally, while still serving requests from the old version until the new one is fully deployed.
+* Once the new version is successfully running, we gradually shifts all traffic to it, ensuring zero downtime.
+* The old version is decommissioned once the new version is fully serving traffic.
+
+This process ensures a seamless update experience, minimizing disruption to users during deployments. If needed, you can also manage traffic splitting between versions during the update process, allowing you to test new versions with a subset of traffic before fully migrating to it.
+
+## Can I update Serverless Container ressources (vCPU and RAM) at any time?
+
+Yes, Serverless Containers ressources can be changed at any time without causing downtime, see previous question.
+
 ## How does scaling work in these serverless services?
 
 Scaling in Serverless Containers and Serverless Functions is handled automatically by the platform. When demand increases—more requests or events—the platform spins up additional instances to handle the load. When demand decreases, instances spin down. This ensures optimal performance without manual intervention.

faq/serverless-functions.mdx

@@ -26,6 +26,21 @@ With serverless, you only pay for the compute resources you actually use. There
 
 Scaling in Serverless Containers and Serverless Functions is handled automatically by the platform. When demand increases—more requests or events—the platform spins up additional instances to handle the load. When demand decreases, instances spin down. This ensures optimal performance without manual intervention.
 
+## Does updates of Serverless Function can cause downtime?
+
+No, when deploying a new version of your Serverless Function generates a **rolling update**. This means that a new version of the service is gradually
+rolled out to your users without downtime. Here is how it works:
+
+* When a new version of your function is deployed, the platform automatically starts routing traffic to the new version incrementally, while still serving requests from the old version until the new one is fully deployed.
+* Once the new version is successfully running, we gradually shifts all traffic to it, ensuring zero downtime.
+* The old version is decommissioned once the new version is fully serving traffic.
+
+This process ensures a seamless update experience, minimizing disruption to users during deployments. If needed, you can also manage traffic splitting between versions during the update process, allowing you to test new versions with a subset of traffic before fully migrating to it.
+
+## Can I update Serverless Function ressources (vCPU and RAM) at any time?
+
+Yes, Serverless Functions ressources can be changed at any time without causing downtime, see previous question.
+
 ## How do I integrate my serverless solutions with other Scaleway services?
 
 Integration is straightforward. Serverless Functions and Containers can be triggered by events from Queues, SQS and SNS and can easily communicate with services like Managed databases or Serverless databases. Serverless Jobs can pull data from Object Storage or output processed results into a database. With managed connectors, APIs, and built-in integrations, linking to the broader Scaleway ecosystem is seamless.

faq/serverless-jobs.mdx

@@ -26,6 +26,12 @@ Serverless Jobs allows you to run large-scale batch processing and computational
 
 Integration is straightforward. Serverless Functions and Containers can be triggered by events from Queues, SQS and SNS and can easily communicate with services like Managed databases or Serverless databases. Serverless Jobs can pull data from Object Storage or output processed results into a database. With managed connectors, APIs, and built-in integrations, linking to the broader Scaleway ecosystem is seamless.
 
+## Can I update Serverless Jobs ressources (vCPU and RAM) at any time?
+
+Yes, ressources of your Job Definition can be updated at any time.
+
+Ongoing Job Runs will remain using the ressources definied at the time it started.
+
 ## How am I billed for Serverless Jobs?
 
 ### Principles

serverless/containers/concepts.mdx

@@ -185,6 +185,17 @@ A queue trigger is a mechanism that connects a container to a queue created with
 For each message that is sent to a queue, the trigger reads the message and invokes the associated container with the message as the input parameter.
 The container can then process the message and perform any required actions, such as updating a database or sending a notification.
 
+## Rolling update
+
+When deploying a new version of a Serverless Container, a rolling update happens by default. This means that the new version of the service is gradually rolled out to your users without downtime.
+Here is how it works:
+
+* When a new version of your container is deployed, the platform automatically starts routing traffic to the new version incrementally, while still serving requests from the old version until the new one is fully deployed.
+* Once the new version is successfully running, we gradually shifts all traffic to it, ensuring zero downtime.
+* The old version is decommissioned once the new version is fully serving traffic.
+
+This process ensures a seamless update experience, minimizing disruption to users during deployments. If needed, you can also manage traffic splitting between versions during the update process, allowing you to test new versions with a subset of traffic before fully migrating to it
+
 ## Stateless application
 
 A stateless application is a computer program that does not save client data between sessions. Data generated in one session is not saved for use in the next session with that client.

serverless/functions/concepts.mdx

@@ -109,6 +109,17 @@ A queue trigger is a mechanism that connects a function to a queue created with
 For each message that is sent to a queue, the trigger reads the message and invokes the associated function with the message as the input parameter.
 The function can then process the message and perform any required actions, such as updating a database or sending a notification.
 
+## Rolling update
+
+hen deploying a new version of a Serverless Function, a rolling update happens by default. This means that the new version of the service is gradually rolled out to your users without downtime.
+Here is how it works:
+
+* When a new version of your function is deployed, the platform automatically starts routing traffic to the new version incrementally, while still serving requests from the old version until the new one is fully deployed.
+* Once the new version is successfully running, we gradually shifts all traffic to it, ensuring zero downtime.
+* The old version is decommissioned once the new version is fully serving traffic.
+
+This process ensures a seamless update experience, minimizing disruption to users during deployments. If needed, you can also manage traffic splitting between versions during the update process, allowing you to test new versions with a subset of traffic before fully migrating to it.
+
 ## Runtime
 
 The runtime is the execution environment of your function. Regarding Serverless Function, it consists of the languages in which your code is written.