The certificate itself won’t affect that. Running on exceptionally slow hardware can make TLS pages load slower, but that shouldn’t be an issue unless you’re on embedded and/or decades-old hardware. You should look at your browser’s developer tools and see what’s taking so long to load. It’ll break out each and every resource. Maybe you have a resource loading from somewhere else taking forever?
Looks like it’s just not a fast server to me, downloading at 7.5kb/sec.
# curl 'https://bizweb360.com' > out.test
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 91350 0 91350 0 0 7525 0 --:--:-- 0:00:12 --:--:-- 7992
For example, in Chromium or Chrome, if you have the Developer Tools open while loading a page, it will show a “waterfall chart” indicating how long each resource took to load and why.
There does seem to be something very wrong with your site. For example, right now this single static file is taking several minutes to load for me and is only downloading at a few kilobytes per second.
I have a similar problem. I have an instance in AWS LightSail, I set up for the plan with 4GB RAM and 2vCPU. I installed the LetsEncrypt certificate last week and my entire site started to get slow. Including the management part of Wordpress.
Could this have to do with some configuration of my certificate? Does it need to configure something to redirect http to https? Could this be slowing down the site?
Given the way SSL/TLS works, it is close to impossible that a change on the CA side could affect performance this way.
That said, something that could certainly cause this is an issue with the web server configuration. If you used certbot with the apache plugin, which automatically modifies your apache configuration to enable HTTPS, it could even be specific to the HTTPS vhost if the configuration was written in a way that would cause certbot to get confused. If disabling HTTPS reproducibly reduces the loading time, that’s certainly a possibility.
It would be useful if you could share all apache configuration files that were changed from the default configuration (i.e. the configuration after a clean install of apache), plus the files added by certbot (if you used certbot with the apache plugin), plus possibly any relevant logs from apache’s error log.
for example have either of you reached out to amazon to check if there are any datacentre issues or network speed issues?
Start with the observed root cause:
The websites are loading slow
Work on likely possibilities
A) Servers and Networks are slower (have either of you ruled this out)
B) Have you compared HTTP only loads vs HTTPS loads
C) What is the CPU and RAM usage of your servers
D) Do you have any historic records and is the usage higher
E) Have you made any changes (updating OpenSSL libraries, updating Apache/Ngingx)
Generally saying the renewal made things slow down without a shred of evidence or a structured approach is not an approach I would use (and I have been involved in troubleshooting lots of slow applications)
A) Servers and Networks are slower (have either of you ruled this out)
ANS :- I am Useing AWS t2.medium and my http sites working fine. Issue in https only
B) Have you compared HTTP only loads vs HTTPS loads
ANS :- I am useing Instance to Instance LB. For testing i have stopped my loadbalancer.
C) What is the CPU and RAM usage of your servers
This is my server t2.medium vCPU 2 Memory 4MB. And uses is fine.
D) Do you have any historic records and is the usage higher
NO
Have you made any changes (updating OpenSSL libraries, updating Apache/Ngingx)
No only add new domain and run ssl add command. But my ssl renew one day back for all after that sites are slow.
One easy check to see if the speed is an HTTPS issue (either cert or config) is to put up a static file, something somewhat large like a 50 megabyte image and then time downloading it over HTTP and HTTPS using a tool like cURL. Obviously, you can’t have a redirect in place for this test.
Since the server just has to serve a static file, any major speed difference between the two attempts will mostly eliminate the application software as a point of problem.
So for all of the people having slowness problems, have you tried downloading a single large static file over HTTPS to see if that is also slow?
The original user complaining about this, @Manny, has a site where downloading a large JPEG file takes a long time (several minutes!), so there is some underlying problem that makes everything slow. That’s not necessarily the case for every site.
Similarly, have you all tried using something like the Chromium/Chrome developer tools where it displays the times that each resource took to load, and the reasons that each took a particular amount of time?
This site is encrypted through Let’s encrypt also, but doesn’t have the loading time issue .
It is on the same server everything is Same except the site certificate has not been renewed yet after the original setup .
We will removed the ssl and ran the bizweb360 site on a different server it was fine . We are going to do the same on this server and run the site on http and see .
… limits the bandwidth available to each client to 8048 bytes per second. This roughly matches the speed at which I can load the site. I haven’t waited for the full site to load yet, but it’s definitely > 1MB, so a limit of 8 kbyte/s seems rather low here.
(I’m not familiar with mod_cband, but I’d double-check whether any of those directives introduce similar limits.)
One other difference between your wam360.com site and your bizweb360.com is that the wam360 site uses what seems to be a cdn. Most of the content is served up from what looks like a cdn (cdn.wam360.com) which could explain why the wam360 site is not having issues with loading
Thanks for all the help. We were able to fix the issue. The download speed as pointed out was 8 Kb which had been changed . We couldn’t find out the reason , but the site is working fine now .