You’re offline. This is a read only version of the page.
Dynamics 365 Business Central SaaS is proudly built-in on top of one of the highest Service Level Agreement (SLA) but it also requires its own service maintenance strategy. This is done in order to have the less downtime and its perception by the user and relative business processes.
This is generically described in the official documentation
FAQ About Using Business Central - Business Central | Microsoft Docs
...
Both business functionality and service components are monitored continuously and updated as appropriate.
...
And also referenced in
Service Overview for Business Central Online - Business Central | Microsoft Docs
Business Central runs globally in many Microsoft Azure datacenters. All parts of the infrastructure and services are continually monitored and optimized to deliver the best possible experience. Most service operations and optimizations happen without users being aware of them. In some situations, user interaction is needed, like reconnecting to Business Central or refreshing the browser. Users will be notified directly in the browser if any action is required on their part. Users are asked to take action before the optimizations are applied on the service. This way, they can continue to work without further interruptions. Service operations happen all day, every day, to always provide best experience.
...
These aforementioned service maintenances do have very low downtime and have its User Experience (UX) side expressed with the typical “sorry to bother you” message as shown below:
There is a second, probably a bit more hidden, service maintenance side effects that could impact integrations through REST and OData APIs, mainly.
When there is a whole cluster service maintenance that involves the whole infrastructure that wraps around the service tiers, these per definition cannot redirect traffic to another infrastructure to guarantee zero downtime nor emit signals around what is happening available to partner telemetries.
This is typically happening sporadically, during non-working business hours and for a very limited time - as much as possible -.
When this happens, if you have REST or OData API calls you might have them fail with an error message (HTTP 503) stating:
Tenant is under maintenance: Upgrade
Just as pure single example, performing a sampling of one AAD tenant operating in CET business hours (UTC+1) and per 30 days this happened.
|
Date |
Time (UTC) |
Duration (approx.) |
|
17th January |
7 PM CET |
4 min |
|
28th January |
2 PM CET |
6 min |
|
29th January |
7 PM CET |
12 min |
|
31st January |
2 PM CET |
10 min |
|
1st February |
7 PM CET |
2 min |
|
3rd February |
7 PM CET |
11 min |
With a total of 116 requests reporting HTTP 503 in 30 days.
The solution to cope with service maintenance is to catch HTTP 503 error message and implement a retry logic, similar to what is described for HTTP 429 in the documentation
Working with API Limits in Dynamics 365 Business Central - Business Central | Microsoft Docs
If you would like this behavior to be changed and enhanced in future versions, you might think of vote for the following IDEAS
https://experience.dynamics.com/ideas/idea/?ideaid=b6444927-96c9-eb11-ba5e-0003ff45d51f
https://experience.dynamics.com/ideas/idea/?ideaid=d87720c9-6cfb-eb11-ba5e-0003ff45da2e
Thank you for sharing Duilio!
Share
Report
All responses (
Answers (
3,666
