I have noticed that Sitepoint is getting slower and slower. Several times I get this 503 error:
The Lambda function associated with the CloudFront distribution is invalid or doesn’t have the required permissions. We can’t connect to the server for this app or website at this time. There might be too much traffic or a configuration error. Try again later, or contact the app or website owner.
If you provide content to customers through CloudFront, you can find steps to troubleshoot and help prevent this error by reviewing the CloudFront documentation.
A few staff members have mentioned this as well, but it seems to be a regional thing as others have had no issues at all. But I’ll report it up the chain.
When a CloudFront error like the one described is received, there should be an error code at the bottom of the message. It would look something like this:
Generated by cloudfront (CloudFront)
Request ID: nubEILvjW7_z3USPlIxtyZ0PgJSG7kMqimKaJgRjmgppNjVENMVqhA==
It would be helpful to include that information (or better yet, the entire message contents), along with:
The specific URL that you received the message on.
Your IP, or the country you are in when you received that error (or the country you are going through if connecting via a VPN).
If you can reliably reproduce the error (eg. by refreshing the page).
If you clear your cookies for the page, does it affect performance?
Thanks, and sorry for the negative experience.
So far, I have not seen this error personally. I’m in Australia and the servers are in the US, so I would expect that it would probably be slower for me than anyone else. I’m hoping that this means that it’s a very rare occurrence.
I also hope to be looking at options to improve performance soon.
Maybe related or not, but there are a ton of console errors on the site too. Apparently inline scripts that are being blocked due to Content Security Policies (CSP). Maybe something to also look at. Just a snippet below from the latest Chrome…
Thanks Martyr2, but that wouldn’t be related to performance. It’s failures in the network inspector (those CloudFront or Lambda@Edge 5XX error pages basically) that I’m more interested in.
Having said that, I’ve since looked into this some more and can see that some pages are indeed taking a long time. It doesn’t seem to be specific to any particular region, and the problem seems intermittent for page requests (but not all page asset requests). I’ve raised it with Discourse (although things might be a bit slow due to timezone differences that may need to be worked around).
Thanks @Gandalf. I think that’s more inline with what I was seeing yesterday, where the vast majority of errors were for /community/message-bus/<something> which often had a referer of a page of /community/review/<some id>. I’m guessing that that’s what you’ll see in the Network tab of your browser Developer Tools/Web Developer Tools menu with the 503.
I was able to reproduce the issue for these kinds of URLs when querying Discourse directly (bypassing SitePoint’s infrastructure), but I’m still waiting on Discourse to get back to me with an update.
The Lambda function associated with the CloudFront distribution is invalid or doesn’t have the required permissions. We can’t connect to the server for this app or website at this time. There might be too much traffic or a configuration error. Try again later, or contact the app or website owner.
If you provide content to customers through CloudFront, you can find steps to troubleshoot and help prevent this error by reviewing the CloudFront documentation.
Generated by cloudfront (CloudFront) Request ID: oLlCOtqYoISktwTOS15ZTpF64buotJRbSMWyYkv2liSX3HC_Ju8pYQ==
Thanks. I have observed that there are currently issues mainly in the eu-central-1 region, and I suspect that’s where your request is coming from. Traffic seems to be much slower in that area specifically as well for reasons that are not yet clear, but this is under investigation.
I am aware that the UK region is also running slower than normal, although not to the same degree. Hopefully the root cause is the same and fixing one will fix both.
Sorry for the recent issues. I have rolled out an update that I’m hoping has addressed the majority of the problems reported.
Thanks to everyone for your patience and reports throughout this investigation and development.
If you continue to experience site slowdowns or are still regularly experiencing CloudFront errors (or there is a particular error that is reproducible), please let me know. In particular:
The URL (or section of the site if the problem applies broadly)
Your rough location (eg. country)
The Request ID code from the error message, if applicable.
There was one problem that we identified where clicking the sign-out button on the Community forums previously could take longer than normal to load. If you did not wait for confirmation, your browser may have ended up in such a state whereby you would be able to sign back in. If this has happened to you, deleting your site cookies will resolve the issue (eg. the padlock icon → Clear cookies and site data… in Firefox). I apologise for the inconvenience.
Testing and early reports of the update that went out today have so far been positive, so I am hopeful that site performance has improved for all.
it seems better today.
