I have had this experience with the support portal (https://support.dynatrace.com/supportportal) as well as on this site (https://community.dynatrace.com), and I'm experiencing this on Windows, Linux, Firefox, Chrome, Edge ... you name it, it happens there too.
Analyzing it, you seem to send/receive thousands of DOM requests / Events, to be true, you seem to endlessly check getLoginStatus in your SSO for some reason even if there is no user interaction, hitting the cache of the browser engine constantly (getCookie in this case Gecko/Firefox, but it happens in Chrome too):
With a few tabs of your website open (3-4) this adds up to completely freezing the browser and clocking up the CPU to full potential (even on Ryzen 7 3700X at the moment). I don't know why you're checking the login status 3000 times in 5 seconds, but it sounds quite odd to me.
Every tab uses pretty much exactly 1 CPU core, with multiple tabs open, the browser starts to strugge answering the requests in the base process (like getCookie, which are not in the WebContent engine threads, neither is the GC):
Hey, I see you actually do use Dynatrace and it didn't catch it? Happy to help out to help your customer experience (not completely selfless though ;-)). Just kidding, but would be nice if somebody could look into that, it's really highly annoying and makes using the site/community less fun.
I'm not a SaaS user (Managed installation), so that may be connected ("SaaS login" always shows for me, even if i'm logged into my account - and can obviouisly post).
Solved! Go to Solution.
First of all, I'm sorry for jumping on this so late. Our "bandwidth" was super-limited recently...
I'm really impressed by the level of details you provided - that's very helpful! We love such feedback! 🙂 It made us immediately reach out to the right team to investigate the issue.
Hopefully, I'll have some updates about the fix soon. Let me know if the issue persists or if, in the meantime, you've encountered any other issues.
I know that some time has passed, but I would just like you to know, that we've rewritten the SSO's iframe implementation and we're currently testing it on our preproduction stages. If all is good it should fix the issues on Community (reducing the CPU load by at least an order of magnitude) at the beginning of August. I would love to hear your feedback, once we've got it running.
I don't know if you put that in place (production) already, but the issue seems fixed for me. At least I do not experience the issue at the moment.
At least, seems fixed for me - thanks for caring <3.