BAD GATEWAY: ERROR 502

The gerbils are probably getting old and tired.
Perhaps the hosting company is doing some new year cleaning the cage and installing new wheels [servers]
 


Perhaps the hosting company is doing some new year cleaning the cage and installing new wheels [servers]

If it was me, I'd be filling the hosting company's inbox with support tickets.

But, in this case, I know nothing about the hosting. It could be a dedicated box with no support. Heck, it could be some server in Rob's closet.

This is my uptime:


That's a daily downtime of 26 seconds - or about 2.5 hours per year.

The Linux.org downtime is 5 minutes 16 seconds per day - or about 1 day 7 hours 47 minutes per year.
 
Screenshot from 2023-01-06 06-53-33.png
 
It hasn't been *that* bad today. So, we've got that going for us!


The outages can indeed be annoying. I suspect we could improve the uptime if folks were kind enough to donate for hosting fees. Then @Rob can buy some more gerbils to keep the server up and running!

(Also, we're a PG13-ish site.)
 
I figured I'd write this to show how hard it can be to keep a site online 24/7. This report is pretty tame - and it's just from a single week at Linux-Tips.us. The numbers are often higher, sometimes by a wide margin:

hUZMdmh.png


Hopefully you can read that. I had to zoom out quite a bit to get it all into one screenshot.

That's just from a single layer of my site's defense. That's the stuff that made it through other layers or stuff made it through this layer.

There are constant attacks. This is a mild weekly report.

I guess what I'm trying to say is that it's difficult to keep a site online all the time. As frustrating as it is, there's a whole lot that's beyond our control.
 
Here's another very annoying problem...that still happens...
2023-01-11-10-32.png


Got it just now.
t1935.gif
 
Got it just now.
t1935.gif

If the tab has been open for a while, refresh it before typing anything and pressing the post reply button.

The problem you posted can be either an outage or you're no longer considered 'secure' (see the Cloudflare loading screen that 'scans you') by CloudFlare.

So, if the tab has been open for a while, hit refresh first.

I'm not sure how long that 'while' is, however. If I had to guess, I'd say it's at least a 30 minute window. It's less if you've gone messing about with your IP address via a VPN! (So I have learned through repeated spankings from Cloudflare.)
 
Everything is exactly the same as it was since I joined the Forum...I'm not the only one this happens to.

It doesn't happen all the time and never know when it's going to...if I hit refresh (F5) it stuffs up my post and can't post...only site this happens.
m1512.gif
 
if I hit refresh (F5) it stuffs up my post and can't post

When I notice that's going to happen, I quickly select all and copy my post. It saves retyping it.

And, you're definitely not alone in this.

Even the uptime monitor can miss some outages, as it only checks every 60 seconds.
 
When I notice that's going to happen, I quickly select all and copy my post. It saves retyping it.

And, you're definitely not alone in this.

I also copy and paste my post when this happens.
m1512.gif
 
I always "Select All", then "Copy" before I click the "Post Reply" button on any forum. If something goes bad, I open a text editor window on my desktop and save the post before dealing with the issue.

After doing my own research, I figured out that forums like Linux.org have installed a special detector in their servers. They installed it just for me. It knows when I forget to "Select All, Copy" ... and that is when it triggers the failure and loses the contents of my reply. :-(
 
Under normal circumstances, the forum will also save a draft for you.

We're behind CF so the boss man could actually enable browser cache - where what you type is stored in your cache until you hit the post button. That might work better than the forum's draft system, considering the regular outages.
 

Staff online


Top