One of the solutions I am consulting on today is securing a number of APIs with OAuth with client credential flow, using Azure Active Directory as the identity provider. Those APIs are exposed via Azure API Management, which makes the validation of the access tokens provided as simple as injecting a policy at a product, API or operation level.
While this all works nicely, there is one item in this process which is a painpoint from an operations point of view: as they use client certificates to request access tokens, the certificate management is becoming a bit of a chore with the number of APIs and clients that access those APIs increasing.
A lot of those clients are either Azure Functions or Azure Logic Apps, which today provide the ability to using a managed identity to easy the burden of maintaining credentials.
So it was only logical to think about how would we be able to use the managed identities configured in both Azure Functions and Logic Apps to generate a token that can be validated by API Management. Turns out that this is quite possible, but needs a bit of preparation…
Last January, I wrote about this year edition of Global Integration Bootcamp. As you might remember, this year we locked it on 26-27 February (UTC Time) running two different event streams starting at times that can be suitable for the three main time zone areas- APAC, EMEA and Americas.
We’ve received an awesome number of sessions and in the end, the event will count with 21 sessions and 19 speakers from APAC, EMEA and Americas, including Microsoft Cloud Advocates, Azure MVPs, Community Leaders and Industry Experts – covering a long list of topics in the from cloud and data integration to hybrid integration and API design/development. Here is this year’s lineup:
“So to complete our configuration we just need your outbound static IP…” This is something that pops up again and again, specially if you work integrating legacy systems, like banks, government agencies or other older systems that requires a static IP Address to add to firewall inbound rules.
In the past I had to use on of the subscription tiers from Azure API Management or in some cases deploy the code within a self hosted service in a Virtual Machine. And both of them are valid options if you already have one of those components in place. But in a couple of my last projects, Not only I needed to implement a new component (in this case API Management) just to fulfil this requirement. I also had to change my design a bit, because API Management didn’t support one of the streaming requirements I had (but that is a story for another post).
But I wouldn’t be writing a post just to complain about this (not that this has never happened before) – I actually found another solution – one that I’ve tried and discarded earlier this year, but thanks to some good work from the App Services Engineering team, is finally a viable solution: Azure Functions + VNET Integration + NAT Gateways!
Last week I’ve attended the Integrate conference once again. It was my 5th time attending the conference, and the 4th time I’ve participate as a speaker.
But Integrate 2020 was quite a different experience from the last years. Dubbed Integrate 2020 Remote, to reflect the fact that was yet another event that had to adapt to the pandemic reality that assaulted the planet since the end of last year. Kovai, the company formerly know as BizTalk 360, and the event organizer since its inception in 2012, took some time to decide that the event would go ahead, but when decided that it would be in a new format, didn’t pull any stops to get it running. With almost 1000 online attendees – which is impressive for a paid event, Integrate 2020 had 28 speakers, between Microsoft Program Managers and Microsoft MVPs and as many technical sessions distributed across three days.
This year’s MS Build was very different – we all know that. But for most of us, me included, was the first opportunity to join the event officially. And what event it was.
Ran across three time zone, with a mix of live sessions, Teams Live events and smaller Teams events, like focus groups, which allowed the attendees to really interact with the product groups and advocates from Microsoft.
This year there were some interesting announcements around the Azure Integration Services technologies. I’ve recently shared those announcements on a Auckland Azure User Group meetup, and thought that, since I already had everything collated, it would be a good idea to just share this with you on the blog as well. So, let’s talk about what is now available, or just \around the corner for AIS
“It was the week before Christmas, and…” and I was actually super busy! One of my main tasks on that week was to implement notifications when a legacy Dynamics AX, still running on-premises, had orders ready to delivery.
My solution was relatively simple (although needed to be generic enough to include other notifications later):
I had a very simple event data being provided by the notification repository:
I thought that it was quite an easy setup, but I got stuck for a while setting up the Event Grid Logic App trigger. Why? I was expecting that the trigger would support advanced filters out of the box, on the designer experience, but that’s not the case.
During December’s episode of Integration Downunder, Alessandro Moura showed a recap of the main features that announced for Logic Apps throughout the year. If you didn’t watch it on the day, you should take a look at the webcast.
One of the features that caught my attention, which I haven’t seen before, was the trigger condition. The ability to only fire a logic app if the condition is met. This is great for scenarios where you don’t have control over the event which triggers the logic app (like for example Dynamics 365 triggers, which only allow you to execute a logic app when a record for a given entity has been created or updated), but don’t want to implement the checks within the logic apps itself.
This week we’ve been preparing for a Go Live in a project that integrates Dynamics 365 Field Services with an on-premises system . Part of this process was to run a series of migration script to get data from the on-prem system into D365.
Since D365 pushes the information into logic apps, I thought that the safest way to run that migration would be to simply disable the logic apps that would triggered by the process and the event would be “lost”. But to my surprise logic apps triggers are more powerful than that and remember the last event being processed. So when I turned on my logic apps after the migration, I was “rewarded” with thousands of triggers being fired…
To be honest this is quite powerful, because in cases where you had to take the logic app offline because of downstream system issues or updates, upstream systems can continue to work as expected. But how to avoid that in cases like mine – where bulk uploads (initial load or bulk updates) are not expected to flow downstream?
It doesn’t expect the data like showed above (which is fair enough), but also don’t like “AlternateEmail”: null. Instead it expects the AlternateEmail element to be dropped from the payload. Trying to do this with logic apps components would make the workflow really hard to maintain later (and to be honest I don’t even sure if I would be able to pull that off with out of the box components like composite and variables).
Last week I was reviewing a logic app with one of the lead developers at Theta. It was a relatively simple logic app. It needed to call an oData endpoint on a regular basis, and process the value back. The developer original design was to:
Poll the oData endpoint on the agreed interval.
Test if the value array length was bigger then zero
If the array was larger than zero, process each value in the array within a for each.
Else, terminate the instance as cancelled, so he could distinguish between real executions and zero-length polls.
That would work, but if felt wrong for a couple of reason.
The logic app was wasting an action to test if the actual logic would be executed or not.
Another action was being wasted just to tag the logic apps that didn’t actually “fired”.
Then it dawn on me that we should be using splitOn instead. This would avoid the check, Setting up the splitOn is as simple as adding this on your trigger: