Error: Aborted Happening Very Often, 50% Failure over Past Several Days

Aspose.Slides Cloud Product Family
,
21

Hello @andrey.potapov - do you know when I might expect to hear information from the development team? It has been over 40 days since I opened this ticket. That feels like a long time for a response. The Error: aborted seems like something that they should be able to give me ideas about how to resolve.

22

@beautifulai,
I forwarded your request to our developers and will inform you of any progress.

23

Based on experience, do you think I will get a resolution? How long could it take?

24

@beautifulai,
Unfortunately, I cannot answer your questions. Our developers will do their best to resolve the issue.

25

@beautifulai,
It is hard to determine what is happening. Our developers are able to get an error “500 - Reading the request body timed out” when they perform many (~100) requests to the Cloud at a time via Node.js SDK. They suspect that the requests library is to blame, because when requests are made without the SDK via Axios, such errors do not occur. Our developers will replace the request for Axios in v23.3. They hope it would help, though they cannot guarantee it, because there are a number of things this suggestion cannot explain.

  1. There is not a single trace of the exception thrown by other customers.
  2. The percent of exceptions is much lower than what you reported: 1 or 2, seldom 10+ of 100.
  3. Why did the library suddenly start throwing exceptions? If it’s a bug in the request, then it must be happening all the time, because we’ve been using this library since the beginning.

However, our developers will try to do the update.

26

This does not sound like the situation we’re having. Looking at the attachment, each request is several minutes apart.

image.jpg (64.7 KB)

27

@beautifulai,
I forwarded it to our developers. Thank you for your patience.

28

@beautifulai,
Our developers could not reproduce the issue, but we have migrated our application for AWS to K8S infrastructure, and we hope the bug will disappear because it very much looked like an infrastructure issue. Please inform us if the bug still persists, which we strongly hope not happen.

29

We’ve migrated away to another platform as we could not wait 5 months for a fix.

30

@beautifulai,
We are very sorry that we could not fix this problem for a long time. We apologize for any inconvenience.