API Outage | PDFShift

Resolved

The service seems to be now back up and working correctly.
We'll keep a close eye on the service and are already starting to implement a new plan in case something like this happens again.

Our objective now is to setup an architecture that can fail at AWS and still remain operational for all our users.

Posted 28 Feb at 06:42pm CET.

Updated

We are starting to see positives results from the outage, the current deployment seems to handle the issue. We'll keep an eye to ensure it stays on.

Posted 28 Feb at 06:29pm CET.

Updated

A new error is occuring at the SNS endpoint from AWS...

Posted 28 Feb at 05:56pm CET.

Updated

We are currently deploying an update that tries to improve the communication with AWS SQS when their service is failing. The deployment is in progress and the API should be updated in about 20 minutes from now.

Posted 28 Feb at 02:57pm CET.

Updated

There is definitely an issue going on at AWS SQS: Sentry is reporting 12 news issues today, all of them are related to issues connecting to AWS SQS.

We are working on a fix to mitigate this right now.

Posted 28 Feb at 02:36pm CET.

Created

Starting at 10:55 am today, our API started to get slower and slower in providing responses, causing requests to be dropped and long processing time.

We are currently investigating the issue, but the errors we are receiving tends to point toward AWS having some issues with some services we rely on (SQS mostly is causing issues).

We are working on finding a solution to mitigate this and will keep you updated of any news.

Posted 28 Feb at 11:55am CET.