Why just not lower the price for a paid subscription?
Before all of this happened and projects “upgrades” was announced i wanted to support glitch, but when i saw the price i couldn’t understand, why? Ten dollars for what other services offer for half or one third of the price? Yes, glitch has some advantages, but do they really worth this much?
Maybe it’s just me, because i’m fine with limits, besides apps falling asleep. Just thinking that many users would paid for the upgrade if the cost was $2-3.
So you suggesting just go for AWS, because there are really no point in staying with glitch?
And people will just go away, because why would they pay excessive money.
Glitch wasn’t about making money, so what really changed for the past couple month?
Why it’s impossible to offer cheap paid plan without increased disk and memory limits?
If you don’t want to use the glitch editor and its ease of use functionality and prefer working locally and then zipping, FTP’ing and then extracting your files on a server, go right ahead.
Downtime is good sometimes, I have a discord bot with a game, that people type a command to accumulate points since the last time someone typed the command. Even if the bot goes down points are accumulated so when the bot is up again and no one notices I get lots of points .
Thanks for the notice tasha! I was wondering why I kept getting errors from UTR
My only complaint about this is that it would have been nice to receive this notice when it happened, or at least within hours, instead of the next day.
Thanks for the work you guys are doing, and good luck with your reliability-making (I think that’s proper grammar? Welp it is now)
According to a recent post by the Glitch Staff, they have not banned UptimeRobot but they have said in their latest blog post about pausing pinging service to Glitch projects.
For incidents going forward, we may temporarily block pinging services as an alternative to blocking traffic and shutting down projects, as we have done in the past. This will lower the impact on the community as a whole, but anyone still using pinging services will likely receive more notifications that their apps are down.
The error that @RyanLandDev might be because of the above reason, because a 403 error means a forbidden request (where server understood the request but refuses to authorize it).
UptimeRobot is having issues with it’s pinging service (Or they disabled the pinging stuff, but that is unlikely). Use Awake! or other pinging services that pinging your project in 5/10 seconds. Awake pings your project at every 60 seconds.
This is really bad I have glitch pro or whatever, but so many people rely on glitch for discord bots to stay up, but they will no longer stay up, this is just really bad
Hi glitch, did the servers get a refresh? My discord bot is now reinstalling itself which it rarely does
Also why the ping block if there’s no incident according to status.glitch.com, is this a test to see how well things can be after a ping service block
Here’s what I think may have happended. The node modules for all glitch servers were deleted and ping services were disabled to minimize the amount of projects up when this was happening. Most users might be seeing their apps take a long time to start because everything has to be reinstalled.
well i am…
but since those pingos are now blocked, i think i might create a puppeteer / deno webview project that monitors/pings my projects that use less resources and have lower priority than those i boosted.
Update:
We will be blocking 100% of ping traffic today to prevent our API response time from slowing down. This block will most likely last the rest of this week. If this changes, I will post another update here.
Many professional bot developers use glitch for their projects, but now that it’s premium only to stay online, there’s absolutely no reason to use it. Glitch is a step up from Heroku, but now is on the same level, because even Heroku couldn’t keep projects online. Many people will probably move away from glitch, as always online for free was what people had used it for from the start. Though this is only a temp block, I expect that many people will move away from glitch, instead of paying. People chose glitch because it was free, and that there was a way to keep it always online, and that it was the only option. Some people are students and kids getting introductions to coding, people who don’t have money to pay for hosting, and half of my IRL data display projects (built in real life) use glitch projects as an API & data storage. I can’t afford to spend $9/month on glitch, as well as a payed host for my main project. I know it’s hard to keep projects online, and manage all of these servers, but it would be awesome if you found a way to make it possible. I am supportive of the decision, just hopeful that this will change soon. Thanks for your hard work. You’ve helped people around the world who want to get into coding, get better jobs, start a fun project, and more be able to start on an easy and amazing platform without having to pay.
Looks like I’m gonna have to jump ship as well. I have projects that are contacted programmatically via my website and applications for things like statistics/data, but Glitch is blocking it all since it’s classified as “ping services” when it isn’t. Sad. Glitch used to be good but now it isn’t.
Hmmm, idea! A discord bot that pings other bots. Discord bots do spend a lot of time idling so they will use up server resources. Maybe there could be one that can check and activate other bots by watching for you prefix and checking if they are in the server
Though this is a good idea, discord bot’s don’t idle and use too much resources. Every time a message is sent, or an action is sent, Discord sends a request to the server containing information. Based on the way that your Discord API wrapper is set up, it will cache the users & data involved in the request. If glitch had a way of forcing the cache clearing, or limiting the ram, forcing users to use the cache clearer, that would be a more realistic solution. Knowing the Discord API and slightly how glitch works, this wouldn’t be possible, as the bot has to “Log In” to receive requests, and, owning a bot in 5.6k servers (on glitch until about 1k, though I still have other projects here), there would be no reason to have every single server use another bot just to keep it online.
There are many other reasons why this isn’t a realistic solution, based on the Discord API’s current limitations, the way Glitch works, and the way the API wrappers work, though there could be many other solutions.
Sorry, but that’s exactly what I’m doing with glitch too. Because the glitch editor is a horrible mess, and most of the glitch website tries to be all funky with non-intuitive buttons and icons anyway. I much prefer working and running my stuff locally and then pushing it whenever ready, unless it’s like a quick one-line change. For any actual work, the editor is unusable.
For autopingers I suggest they send head requests to minimize bandwith usage. I believe this can result in a 2x or more bandwith saving but i’m not sure if this will still wake the projects
So there is a solution for you. Only works on firefox.
Download the addon here and just click on the website’s title and click with the right button and should this popup so select 1/5 minute(s) to ping your project. And please note that the glitch’s block is temporary. If you want 24/7 just use a laptop with full charged battery in it.
Glitch has temporarily blocked pinging services and as a result of that, projects return a 403 Forbidden error. To know more about this temporary blockage of pinging service, see this blog post by Glitch:
Ping services are mainly monitoring services that are used to keep projects online 24/7 (like UptimeRobot and Awake!). I do not think that is the reason because 503 means “service unavailable” although I expected Glitch projects (and experienced it) to return a 403 Forbidden status.
OK, so Glitch has turned off the ping service for now, I get that.
As we were getting intermittent timeout errors on the 11th and 12th and are now getting consistent 403 errors on every integration since the 13th when they turned off pings, I assume that interaction between [Google Assistant + IFTTT + Glitch] falls under the ping protocol.
I also see where we could do a paid subscription to keep 5 projects awake, but “awake” doesn’t seem to be the issue, ping-ability seems to be the issue and even at our low usage we exceed 5 projects so that would still be useless. Plus awake or not, they were working before with no issues, so I don’t believe sleeping projects are the problem.
So, basically, whenever Glitch experiences an issue and decides to turn off pings, our home automation will break, would that be correct?
Our verification service has been receiving usually high activity lately which is what leads me to believe that Glitch has blocked https://evida.elliebot.xyz.
Hi, so this is my first complete Code and Running Project in Glitch Platform,
and I saw a lot of youtube videos and read some topics in glitch about How to Make bot running 24/7 in Glitch
a lot of people recommended UptimeRobot and Node-keepalive,
I’m using both of them in my project, but the project won’t stay awake and it will turn off about 5 hours later or something
so here I am begging for help!
My server.js Code:
const { Client, Collection } = require(“discord.js”);
const { readdirSync } = require(“fs”);
var TOKEN = process.env.TOKEN;
const client = new Client({ disableMentions: “everyone” });
also when my project is online the UptimeRobot will say it’s still offline, so I have no idea what to do!
and every time I’m starting the project the app will use npm i in console, so is there any way to get rid of it too?
my project link: https://glitch.com/~asqar
but it is installed before, so whats the point of installing again?
i really don’t want this to happen because it takes a lot of times to start the project!
watch.json auto-restart is only for the time you edit something in your project(i guess), and i already disabled that, if the thing you are saying is not the same thing im saying, please give me example of watch.json !
ok, here is the thing,
i already have this file
and this restart type that glitch will do when we are editing is like killing the active process and use node index.js again,
but the restart i’m talking about is not like this, because as you said it will install all packages AGAIN
its something like using npm i or something as i said!
i know, but thats not my problem, my problem is like
im going to run the project after 12 hours and it will install all the packages that were installed before
and i dont want this!
Hello Glitch.
Over the past two days, I’ve had to reinstall express 50 times.
It just randomly decides to go away killing the server.
Please take a look into that.
Thanks.
Yes, you can use the watch.json to specify which files trigger a container restart or install when edited as well as change the delay time between making edits and those changes being applied.
We are receiving reports of some webhooks that are unrelated to the use of ping services not working right now. Our team is looking into this and as soon as I learn more, I will give you an update! Thanks for flagging this!
Glitch doesn’t know themselves how long its gonna take to fix the issues their facing, therefore there is no ETA on how long it will take to get the ping services block lifted.
Okay Thanks For Responding I Hope They Get In Up And Running Soon
But I’m Still Wondering Is The Block Temporary Or The Blocking Pinging For Premium Only Or Is It Just Overall Blocked
For anyone having the problem that the console errors that a module cannot be found, glitch resetted the node_modules to prevent further issues to their and your systems. And how do you help that out?
• The simplest way is going to your package.json,
• Look at dependencies
• REMOVE them all, make sure the bot sees this by having it refresh/restart if you have a watch.json file.
• Once this finishes, paste back all the information you just removed from the dependencies list.
• Restart/refresh once again and that should help. *
If this doesn’t, try searching for solutions like enable-pnpm or by doing npm i <module>.
Hi there @RonnieBoy82!
As you can see from the recent update on status.glitch.com, the team are working on fixing the current outage. I imagine the block will be removed when the problem is identified and a fix is implemented. It would be very hard for the Glitch Support Team to know when this would happen as it depends on the severity of the issue
Hope this helps clarify some more,
Eddie
It’s like a river of discord bots and glitch is a dam and the repl.it is a dam after it. With glitch blocking pinging services I’m pretty sure most discord bot coders will flow down to services like repl