From what I can see, the downtimes are partly because of the pinging services getting pinged too much.
A simple fix for this is pinging every 4 minutes. Some projects have it pinging every 15 seconds, which is not needed because Glitch projects sleep every 5 minutes.
Imagine 20 projects getting pinged every 15 seconds.
I 100% agree with this. Projects that ping any more than 5 minutes should not be in use.
Yeah, even Glitch said the same thing about this. What’s the difference in pinging every 15 seconds and 5 minutes? Both of them will keep your project online 24/7 the same way and it eases the strain on the servers.
Exactly.
20chars29chars
I support this! I’m moving my Glitch projects that have UptimeRobot to 30 min intervals!
I don’t really know if that would be of much help because your projects would not work for most of the time.
You can keep them at 5, since that keeps them online. The problem is having projects pinging at 15 seconds, not needed.
They are ones I don’t use anymore anyway - like C Web Build is now dead and ScriptingToSaveLives isn’t a thing anymore etc.
I agree with you. You have my vote!
Yes. Totally correct, performance would be amazing if that was the case. I hope that glitch devs see this!!!
it should be that if a project is getting pinged faster than every 4-5mins by one ip that ip will be blocked on that domain (429 per ip?)
We have to fix this! This is a major issue in the community.
Especially with a lot of new monitoring projects offering to ping projects every 15 seconds, this is gonna be an ideal solution.
The reason for the downtimes aren’t usually because of pings, it is the amount of projects using resources on a machine. More machines, more projects, more problems. Reason why Glitch also goes down is that they have to rely on AWS and make sure that they are getting payed. I made a post about this:
But, I would like to know what you would’ve done in this situation (not sarcasticly.)
Maybe it’ll restart someday