Forums

This is after the Nov5th update. The web app is used in production. Please explain why is there so much unreliability and what is expected in the future

Hi @FinishLineGelatoAcc, it looks like you were caught up in a outtage on a particular server which hosts your files. Unfortunately it seems like there's a gap in our alerting and nobody was woken up to deal with the issue so it carried on for some time before fixing itself. It looks like it's functioning correctly now but as I've just been notified , we don't currently have any more information about what went wrong. I'll let you know when I know more

Still not working, can you check, should i restart web app or always on tasks?

I’m experiencing issues with asynchronous tasks (always on tasks), and it appears that both of my PythonAnywhere applications are not functioning as expected. According to client feedback, these issues began earlier this week and have been escalating each day. Today, the applications are completely down, impacting all operations.

@FinishLineGelatoAcc -- is the problem fixed now? We're not aware of any issues after the outage that @sboyd posted about earlier, though.

@hyllervianna -- I see that you emailed us about this. Your issue appears to be something different (having started earlier this week), and we'll help you work it out over email.

The problem got fixed later in the day.

Can you give me more details on what exactly happened? What are the steps that you are taking to ensure that problems like these don't happen again? How much more time will this migration take? What is the frequency of scheduled downtimes? We have a paid account and in the last 2 months we have faced 3 long downtimes. Our customers are suffering business losses because of the same.

Hi there -- the migration is complete. We had storage server issues for a number of months -- more information here -- and the fix for those issues went in on Tuesday.

Unfortunately it appears that one of the existing storage servers had a number of problems handling SQLite locks immediately after that upgrade (and as a result of it), but we believe they are fixed now.

We have put an alerting system in place so that if the SQLite issues return, the on-call engineer is paged immediately -- the workaround from our side is something that can be run in seconds, so if the issue does come back, you should not expect any significant downtime. And, of course, if it does come back, we'll immediately start investigating further to see why the fix that we currently have in place didn't prevent the recurrence.

mine the text fields especially passwords stopped working like its disabled or something but i did not touch anything it just stopped working

What do you mean by "stopped working"?