Skip to main content

Microsoft Teams Downtime Alerts After the Connector Shutdown: Workflows, and Who Owns Them

Set up Microsoft Teams downtime alerts with a Workflows webhook now that Office 365 connectors are retired, and keep the workflow alive when its owner leaves.

FLAREWARDEN
FlareWarden Team
7 min read

Microsoft switched off Office 365 connectors between May 18 and May 22, 2026. Every Teams channel that got downtime alerts through an old incoming webhook went quiet that week, and plenty of them haven’t noticed yet. The replacement is a Workflows webhook. It takes about three minutes to set up, and it carries one new risk the old connectors never had: the workflow belongs to a person.

Set Up Microsoft Teams Downtime Alerts in Seven Steps

You need a Microsoft 365 work or school account that can use Workflows. A personal Teams account won’t do. Pick a channel just for alerts, so a 3 a.m. outage doesn’t scroll away under lunch plans.

  1. Open Workflows on the channel. In Teams, click More options (⋯) next to the channel name, then Workflows.
  2. Choose the template “Send webhook alerts to a channel”. Skip the variants for “specific people” or “people in an org”: those expect a signed-in sender, and a monitoring service can’t sign in as one of your users.
  3. Name it and pick the destination. Call it FlareWarden, choose the team and channel, and click Save.
  4. Copy the URL Teams shows. All of it, including the sig= parameter at the end. That signature is what lets an outside service post without signing in.
  5. In FlareWarden, open Integrations, click Add Webhook, and choose Microsoft Teams as the destination type.
  6. Paste the URL, name the destination, pick which alerts and monitors it covers, and click Connect.
  7. Click Send a test message. A card should appear in the channel a few seconds later.

Microsoft’s support article on Workflows webhooks says the webhook templates and trigger don’t need a premium Power Automate license, so nobody has to buy anything for this. Our Microsoft Teams alerts guide has the exact URL formats and every error message.

Treat the workflow URL like a password. Anyone holding it can post into your channel.

Why Your Old Teams Alerts Stopped

If you set up Teams alerts before mid-2024, you probably used an incoming webhook connector, with an address ending in webhook.office.com. Microsoft blocked new connectors in 2024 and then, per its developer blog’s April 2026 update, disabled the rest in a rollout from May 18 to May 22, 2026. “After these dates, Office 365 Connectors will no longer function.”

Two other traps from the same migration:

  • Old Power Automate URLs died too. Workflow trigger URLs on logic.azure.com moved to environment.api.powerplatform.com, and the old ones stopped working on November 30, 2025. A workflow you created early in the migration may need its URL copied again.
  • Some tools quietly kept posting to the dead address. The sender gets an error, but if nobody watches the sender’s delivery log, the channel just stays silent. Silence looks exactly like “everything is up.”

FlareWarden refuses a webhook.office.com address outright and says why, which beats accepting it and failing later.

The Workflow Belongs to a Person

This is the part the setup guides skip. A Teams workflow is a Power Automate flow, and it has an owner: whoever clicked through those seven steps. When that person leaves the company and their account is disabled, their flows can be turned off with it. Your organization’s admin can also block Workflows, or the connectors a flow uses, with a data policy.

Either way, alerts stop and the channel says nothing. Here’s what we’d do on day one:

  • Add a co-owner. Open the workflow in the Workflows app and add a second owner, ideally a shared admin account rather than another individual.
  • Write down who owns it. Put the owner’s name in the destination name in FlareWarden (“Teams #alerts, owned by Priya”). Future you will thank present you.
  • Send a test after any staff change. One click, and you find out today instead of during the next outage.
  • Keep email on for critical sites. A chat channel is a good second signal and a poor only signal.

Why a Test Can Pass and Still Post Nothing

Teams answers a webhook with 202 Accepted as soon as the workflow starts, before the card is actually posted. If a later step fails (the channel was deleted, the owner lost access, the card was refused), the sender never hears about it. Any monitoring tool, ours included, can only report what Teams told it, and Teams said “accepted.”

So when a test shows as sent but no card arrives, the answer lives in Teams, not in your monitoring tool. Open the Workflows app, select the workflow, and read its run history. The failed step is usually obvious.

The opposite failures are easier, because Teams does answer them:

Teams answersWhat it usually meansFix
401 or 403The trigger only allows people in your organizationEdit the workflow, set who can trigger it to Anyone, save, paste the new URL
404The workflow was deleted, or it’s an old logic.azure.com URLCopy the current URL from the workflow’s trigger
202, but no cardA step after the trigger failedCheck the workflow’s run history in Teams

FlareWarden keeps each destination’s delivery history, including Teams’ answer, and retries a failed delivery at 1, 5 and 15 minutes before giving up.

What Arrives in the Channel

FlareWarden posts Adaptive Cards rather than raw JSON. Each card has a coloured headline with a status emoji, the URL, the cause, where the failure was verified from, how long it was down, and a button to the monitor. The same destination can carry SSL expiry and certificate changes, DNS failures and record changes, and response-time anomalies, and you choose which.

Two limits come from Microsoft, not from us. Workflows posts as the Workflows bot, and Microsoft’s blog states that custom names and icons aren’t supported for Adaptive Card payloads, so the footer on each card is what tells your team where it came from. And Microsoft’s incoming webhook documentation caps a message at 28 KB, so we shorten long values (a long error, a big list of DNS records) to keep every card under it.

The bigger question is what deserves a card at all. A channel that pings on every single-region blip trains people to mute it within a week. We only alert once other regions have re-checked a failure and agreed, which is the reasoning behind our article on false positive downtime alerts. One verified card beats five maybes.

Teams, Slack or Google Chat

Pick the tool your team already reads. If you’re on Microsoft 365, Teams is the obvious home and nobody needs a new app. If the team lives in Slack, see our guide to Slack downtime alerts instead; Google Workspace teams can use Google Chat downtime alerts. Each destination counts toward your plan’s webhook allowance, and the free plan includes two.


Key Takeaways

  • Old connector URLs are dead. Office 365 connectors stopped working in May 2026; anything still pointed at webhook.office.com posts nothing.
  • Use the “Send webhook alerts to a channel” template. It needs no premium license, and the trigger must let Anyone run it.
  • Give the workflow a second owner. A flow owned by one person can switch off when that person leaves, and the channel goes silent.
  • A 202 isn’t a delivered card. When a test passes but nothing shows, read the workflow’s run history in Teams.
  • Send verified alerts only. A channel that cries wolf gets muted, and then Teams alerts are worse than none.

Want outages to land in Teams only once they’re confirmed from more than one region? Start monitoring free with FlareWarden: 15 monitors, two alert destinations, no credit card.