This is perplexing. I have 2 sites pointing to the same branch on the same repo of a Quasar SSR/PWA app. Both sites are identical. The only difference is that one has a temporary domain and the other a custom domain.
Everything is fine on the site with the temporary domain. But on the custom domain, not so much.
• The app on the custom domain does not boot as performant as the temp domain site. The below pic is a side-by-side comparison of each site loading after clearing each site cache and opening in a private browser. The site on the left is the temporary domain, and the data that is fetched on the boot-up of the app is painted almost immediately (as expected). However, the same code on the site with a custom domain hangs. As a result, our Lighthouse is way down compared to the other site even though it’s the same code.
• I see a hydration error that shows up only on the custom-domain site, and only on the first page loaded. It goes away on refresh or subsequent visits. Again, strangely, this only happens on the site with a custom domain. The temporary domain doesn’t have any hydration errors. And because the error on the custom domain goes away on refresh, I can’t do any process to isolate the error on a breakpoint/re-run method.
This feels very weird to ask but are there any ideas out there as to why I’m getting an error and slower performance from a site with a custom domain compared to that with a temp domain that shares identical code repo and branch?
I wasn’t able to see the same hydration console error for the custom domain site… though, it does seem that the custom domain takes longer to load. I restarted your app to see if that might clear up a bottleneck somewhere and it seems like it might have. Does the custom domain appear to load faster on your end?
@amiedema, as always your time and feedback is much appreciated! Thanks for taking a look at restarting the app. The discrepancy between both sites still persists though. I’ve been looking at other SSR framework forums to see if there are any similar issues and am now thinking that this must be a slow DNS host. I’ve changed the nameservers from that of the host company to the NS addresses of the Digital Ocean drop I’m using for the app. Hopefully, that update resolves the issue once it propagates. Or perhaps I need to transfer the domain entirely. In a wait and see at the moment. Thanks again.
Yes, that helps! I’ve been researching the steps I need to do to avoid downtime with the transfer. Looking at this link, am I correct to set up SSL to Full (Strict), seeing how Cleavr provides SSL on the server already? Step 2: Setting up SSL with Cloudflare - Tutorial - Cloudflare Community
I think Full Strict can only be used if using a CF SSL cert - which you can do. If using Let’sEncrypt, should select Flexible and then make sure the CF proxy for the domain is turned off, otherwise LetsEncrypt won’t be able to add cert nor renew it.
Right on. The switch is (partially) made. Apparently, I turned the proxies on too soon and had the site down for a moment. I’ll try again after waiting a day to see how much the needle moves. Thanks again @amiedema. This Cleavr board is a gem. Thank you for your time and effort!
