ConversationRelay Degradation

What to do

• Check whether your product depends on Twilio. Check your stack

  Paste a package.json, lockfile, Terraform or a list of domains at down.guide/stack to see every provider you depend on.

• Confirm your retries, queues and scheduled jobs caught up after the 0m it lasted.

  Anything that failed between 2026-03-26 22:33 UTC and 2026-03-25 01:30 UTC may need a replay.

• Subscribe to Twilio's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=twilio works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from Twilio 1

1. resolvedMar 25, 2026, 1:30 AM UTC

   Twilio Voice ConversationRelay was degraded for 1 day 18 hours and 54 minutes between 18:29 Pacific Time on 03/24/2026 and 13:23 Pacific Time on 03/26/2026. During this period, ConversationRelay sessions could intermittently fail to terminate when the WebSocket Application sent an end message while still streaming text-to-speech tokens. If all tokens completed before the end message, the session terminated normally. However, if the end message arrived mid-stream, the remaining tokens were discarded and the session entered a state where it could not complete. Affected calls remained silent until the caller hung up or a 60-second safety timeout fired. In both cases, the callback did not receive the end message's handoff data, preventing call transfers from completing. The issue has now been resolved.

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.