Push API: newCall response not reaching sipgate — "Customer server not reachable" for ~45% of calls since Aug 7

[yarek98]

Since Aug 7 2026 ~10:00 CEST, roughly 45% of incoming-call newCall callbacks fail with Customer server not reachable / callback skipped because no callback URL was configured in the webhook log. Zero such errors in the two weeks before. Our endpoint responds in ~0.2s with valid <Response onAnswer= onHangup= /> and is reachable over HTTPS; the affected newCalls do reach us and are processed — only sipgate's callback delivery fails. Onset matches your status-page incident of Aug 7, but the issue persists on Aug 8.

This is affecting a live account. I've also mailed expert@sipgate.io with the account ID and concrete callIds. Could someone from the Push-API team look into the delivery failures? Happy to share account ID and callIds privately.

[fbett]

We have seen the same error message.
In our case, the problem was the usage of internal dial numbers in the dial response.

Assigning external numbers to these groups and referencing them in the dial responses has solved the issue

Sad that this is not working anymore.
It has worked since at least April 2024.

But the documentation now says, only external numbers are allowed.