If you are experiencing timeout or connection failure errors, regardless of the error code, then
it means that your request to our Ably network has timed out. This could happen for a variety of reasons, such as poor network conditions, improper usage of the Ably SDKs, or Ably server issues.
In order to diagnose a timeout issue, we need sufficient information to understand whether the issue is simply a connectivity issue from the client, a problem with the request itself, or a problem with the Ably network or service.
As such, please please raise a support ticket with as much detail about the failing request, including as much of the following as possible so that we can compare performance and request logs at the time:
- When exactly did the timeout / connection errors occur. We prefer timestamps in UTC format.
- Can you provide the complete logs from the SDK at the time of the failed request? Our SDKs retry when a request fails by default, so we'd like to see if the retries succeeded or failed, and as such, need to see the logs from before and after the actual timed out request you are reporting.
- Which Ably endpoints where you hitting? Specifically, do you have a custom client option environment setting, and what operation from the SDK was failing when the error occurred?
- Which SDK(s) are you using, on what platform, and which versions?
- Do you have stack traces?
- Is this problem continuing to happen or has it happened just once?
- How do you know that the issue was not an Internet connectivity issue at the time versus an Ably availability issue? It's common that when a client has intermittent connectivity issues that requests to Ably will fail / timeout for obvious reasons. It's important we know that the issues you are experiencing are not just connectivity issues on your end, so any insights into what other Internet operations were being run at the same time that were succeeding would be useful in proving or disproving this.
- What is the app ID for the app you are connecting with?
- In addition, please see our standard article on the information we typically require to debug issues and steps needed to reproduce the error if possible.