Jump to content
Sign in to follow this  
CANARYKING

502 Bad Gateway

Recommended Posts

It’s now such a problem as to render the forum almost unusable (on iPad at least). Here’s hoping that this post actually ‘arrives’...

Share this post


Link to post
Share on other sites
2 hours ago, Norwich R Us said:

Yes it's shocking, probably the busiest time for archant and the service is failing, no good for anyone

It's falling apart precisely because it's busy, need to up the server size, increase the nginx connects or add another server to create a cluster. 

Share this post


Link to post
Share on other sites

The forum is falling apart again....

Seriously though, it's almost unusable at the moment. Really frustrating, especially for us exiles.

Share this post


Link to post
Share on other sites

It would have been nice to celebrate a bit with fellow Pink Un'ers, exiles and assorted loons....and read Ricardo's report.

Share this post


Link to post
Share on other sites

Bit of shame it's not working properly. Continual refreshing seems to get a partial service though.

Sort it out Archant.

Share this post


Link to post
Share on other sites

Such a shame that on a day when we all want to celebrate and bask in the glory of Premier League promotion, the forum is virtually unusable.

Share this post


Link to post
Share on other sites

The bl00dy irony...

 

As someone has put above, I click on a post called 502 Bad Gateway...

And what comes up a poxy white screen with 502 Bad Gateway.

 

Come on Archant sort it out

Share this post


Link to post
Share on other sites

Given the number of ads this site must serve, I would have been sure that this would have been fixed sooner.

Share this post


Link to post
Share on other sites

Sort it out Archant, it only happens on the message board not the rest of the site.

This what it says on the web " The HyperText Transfer Protocol (HTTP) 502 Bad Gateway server error response code indicates that the server, while acting as a gateway or proxy, received an invalid response from the upstream server."

and this is how to fix it. -- https://blog.hubspot.com/marketing/502-bad-gateway

Share this post


Link to post
Share on other sites

🎶 the board,

the board is falling apart again 🎶

  • Haha 1

Share this post


Link to post
Share on other sites

5 minutes ago, CANARYKING said:

ITS STILL BLOODY WELL HERE !!!

yep, its mostly ok, so better than yesterday

Share this post


Link to post
Share on other sites

It depends on whether the site is hosted in house or by a third party. I'm having a bit of a nightmare with a hosting company right now and the technical support has been offshored to some incredibly obtuse and argumentative Eastern Europeans who take 24 hours to respond to each message and insist that it's my fault they have disabled my database.

Yes, they need more servers, or to rewrite the code to try again rather than failing over whenever it fails to make a database connection.

Share this post


Link to post
Share on other sites
9 minutes ago, Petriix said:

It depends on whether the site is hosted in house or by a third party. I'm having a bit of a nightmare with a hosting company right now and the technical support has been offshored to some incredibly obtuse and argumentative Eastern Europeans who take 24 hours to respond to each message and insist that it's my fault they have disabled my database.

Yes, they need more servers, or to rewrite the code to try again rather than failing over whenever it fails to make a database connection.

502 is not a database connection issue or a code issue - it means the upstream server (PHP FPM I expect) is not available. It's also nothing to do with the web browser in usage, or the client type. There's a lot of partial and/or misinformation floating around on this thread.

In simple terms, the web server (nginx in this case) handles the incoming connections from users' web browsers, and then decides what to do with it. For some files, such as images, it may serve them directly to the user, but in the event it determines it should be handled by another server (in this case I expect everything matching a .php file extension), it will try and connect to the other server(s) and return the response they give. In the event it fails, you get a 502 error indicating nginx has tried to contact the upstream server but not been able to.

Whilst the specifics are hugely dependent on configuration, I don't think this forum's problems are resource related; my guess is that the nginx web server and the PHP upstream server run on the same machine. As the nginx server always returns promptly it indicates that RAM and CPU are not particularly stressed. In addition, it usually works 1 in x times if you refresh (although this doesn't help people who have POSTed long posts and lost them), making me think that either the PHP FPM server is not configured to allow enough connections to satisfy demand, and rather than waiting is rejecting connections, or that certain workers processes are dying and not being restarted correctly by the host process.

It is also possible they're running multiple upstream services on different servers, and some are timing out - but usually nginx is intelligent enough to exclude upstreams that fail so I'm not sure it's that.

In summary, I'm pretty sure this is a config issue that should be easy to fix.

  • Like 2

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...