Mustangserver 1.3

Mustangserver 1.3

Today, march 24th, 2024, we released a set of changes to both our infrastructure and our Mustangserver server application.

Updates to the infrastructure

  • There are now minimum Service Level Agreements (SLAs) in our contract.
    In the basic configuration we only guarantee 99% availability but we can talk if you need more. We now also have more scalable prices.
    The amount of operations per month when we still consider someone a “standard” customer has been increased five-fold.
  • Authentication using API keys: API keys are easier to use than our standard oauth2 authentication and they can be restricted by time, referrer or IP address. This was possible before but we now resolved the last issues and added it to our documentation.
  • There is now end-to-end-monitoring on both APIs, Mustangserver and Mustangserver-docs. Previously we monitored the OS and the process only, but not it’s responses, which in very rare cases would give us false negatives. Now we periodically perform a full round trip including actual operations on the api manager and the server process.
  • Load balancing is now tested and enabled: We tested the already existing load balancer, now we really “armed” it.

Application updates

Among others,

  • we updated the Phive validator to now also validate e.g. german XRechnung 3.0.1
    (for CII this was previously possible using Mustangservers internal /validate validation),
  • we updated the documentation e.g. on the visualization, i.e. on xmltohtml
  • we made it easier to log messages in a massively parallel scenario and
  • we added a endpoint “combine”, which adds a invoice object as JSON directly to
    a PDF/A file. Previously there was only combineXML, which required the JSON to be converted
    to XML in a previous step.