WTF of the Week · #3
Resolved Feature toggle, then a message loopAn app posted billions of messages, and Slack's free workspaces went quiet
- Who
- Slack, free-plan workspaces
- When
- 1 October 2026
- How long
- About 6 h 20 min
- Read
- 3 min
What happened
At 05:39 Pacific time on 1 October, Slack said some free-plan workspaces were blocked from sending messages. The block came from the daily message limit on the free plan, and it was firing in what Slack called an unusual spike pattern.
Seventeen minutes later, Slack found the source: a third-party app stuck in a loop that had sent billions of messages. At 06:58 they added more detail. A feature toggle ramp-up had made an app integration post messages again and again in a processing loop. That pushed free-plan workspaces over the message volume threshold, and the sending block switched on.
Slack had a likely fix at 07:31. The rollout started at 11:52, and at 11:59 they said the block was lifted for every affected free-plan workspace. Slack has said it will do a full investigation into the trigger, the cause and how to prevent it.
The WTF
The message limit is there to protect the free plan. On this day it worked exactly as built, and it stopped the teams themselves, not the app. For about six hours they couldn't talk in their own workspace.
The lesson
Slack has not published its full report yet, so we only know what the status page says. Still, two things are worth checking in our own systems.
First, what happens when one integration sends far more than normal? A loop in a webhook or bot can look like real use. Second, who does your limit block? A limit per workspace stops everyone in that workspace, even the people who did nothing unusual.
If you have a limit, check who it stops when the extra traffic comes from someone else.
Would you have noticed?
From the outside, Slack itself was up. A check that only loads the app would have stayed green. A check that sends a test message from a free workspace would have failed from the start.
Outage bingo
4 of 9. How many has your team hit?
- Config change, stamped
- Bad deploy
- It's always DNS
- Database
- FREE: WTF?!, stamped
- Runaway retries, stamped
- Limit hit, stamped
- Upstream down
- Network
Hug of the Week
To the Slack team: thank you for posting the cause within 20 minutes, in plain words, while the incident was still going on.
Source: Slack status page
Slack has promised a full investigation. We'll update this story when their report is out.