Noticed this when I couldn't pay my internet bill, and also could not browse a tea site I normally do. Amusingly, this means https://downdetector.com/ is down Even more amusingly, the detector of...
Noticed this when I couldn't pay my internet bill, and also could not browse a tea site I normally do.
and while centralized, all-eggs-in-one-basket thinking are one reason why i hate the current internet in some ways, these amusing sites are why i still love it, in some ways. got a good hell of a laugh out of all this :)
much love, and coffee, and tough friday empathy for any of you that have to deal with this now, or soon, at your given place of work
Looks like Google DNS and AWS CDN - which makes total sense for a site that's got to do that job, but still makes me smile at the idea of some grumpy Cloudflare engineer having to use their...
Looks like Google DNS and AWS CDN - which makes total sense for a site that's got to do that job, but still makes me smile at the idea of some grumpy Cloudflare engineer having to use their competitors' services while complaining about it the whole time!
Is it just me or is this happening far more frequently this year than previously? Maybe I'm just noticing it more because it's happening at times I actually use the Internet
Is it just me or is this happening far more frequently this year than previously? Maybe I'm just noticing it more because it's happening at times I actually use the Internet
From the last outage I think the problem is Cloudflare refused to be a "done" product. Instead, they needed to upsell new products and added bot management, WAF, access, among others. Some of...
From the last outage I think the problem is Cloudflare refused to be a "done" product. Instead, they needed to upsell new products and added bot management, WAF, access, among others. Some of those products run in the critical request path so even if you haven't purchased them or opt into them you're still getting their outages on the shared infrastructure.
The last outage is a bad bot management configuration, a product that costs a lot extra on top of the enterprise plan taking down all plans.
To be fair, WAF and bot protection do seem fairly core to what they do as a company - they originally formed as a spin out from a spam bot honeypot project, and as I understand it their initial...
To be fair, WAF and bot protection do seem fairly core to what they do as a company - they originally formed as a spin out from a spam bot honeypot project, and as I understand it their initial success was from their DDoS protection product.
But yeah, the idea that they might be losing focus as the infinite growth treadmill grinds on is one that definitely concerns me too, because it happens again and again. I desperately wish companies could just be good at something, make enough money for their employees to be comfortable and their investors to make a steady dividend, and continue to do so forever.
It happened less than three weeks ago (mid-november), so it probably feels perceptibly more frequent even just due to it happening so soon after the last one
It happened less than three weeks ago (mid-november), so it probably feels perceptibly more frequent even just due to it happening so soon after the last one
There was also an AWS outage in us-east-1 earlier this year, which has pretty much the same impact on end users, so that contributes to a recency bias as well even though it's not Cloudflare.
There was also an AWS outage in us-east-1 earlier this year, which has pretty much the same impact on end users, so that contributes to a recency bias as well even though it's not Cloudflare.
Looks like they're all back up for me, in North America, including DownDetector though it was down for a few minutes... (spoke too soon, WIRED is still down for me.)
Looks like they're all back up for me, in North America, including DownDetector though it was down for a few minutes... (spoke too soon, WIRED is still down for me.)
Noticed this when I couldn't pay my internet bill, and also could not browse a tea site I normally do.
Amusingly, this means https://downdetector.com/ is down
Even more amusingly, the detector of the detector is not detecting, and thinks it is up https://downdetectorsdowndetector.com/
And EVEN MORE amusingly, https://downdetectorsdowndetectorsdowndetectorsdowndetector.com/ is detecting that the detector-detector-detector is down, which is also true
and while centralized, all-eggs-in-one-basket thinking are one reason why i hate the current internet in some ways, these amusing sites are why i still love it, in some ways. got a good hell of a laugh out of all this :)
much love, and coffee, and tough friday empathy for any of you that have to deal with this now, or soon, at your given place of work
Clouflare has it's own site Cloudflarestatus.com that they update as it goes.
I am aware, I was not using downdetector to actually check the status of Cloudflare. It was just a source of amusement that it was down
I saw that downdetector uses Cloudflare in the previous outage. It is kinda hilarious :-)
How do they host this site?
Looks like Google DNS and AWS CDN - which makes total sense for a site that's got to do that job, but still makes me smile at the idea of some grumpy Cloudflare engineer having to use their competitors' services while complaining about it the whole time!
That is beautifully Python-esque, love it!
There are times I hate the Internet and times I love it.
This definitely falls in the latter camp.
Is it just me or is this happening far more frequently this year than previously? Maybe I'm just noticing it more because it's happening at times I actually use the Internet
From the last outage I think the problem is Cloudflare refused to be a "done" product. Instead, they needed to upsell new products and added bot management, WAF, access, among others. Some of those products run in the critical request path so even if you haven't purchased them or opt into them you're still getting their outages on the shared infrastructure.
The last outage is a bad bot management configuration, a product that costs a lot extra on top of the enterprise plan taking down all plans.
To be fair, WAF and bot protection do seem fairly core to what they do as a company - they originally formed as a spin out from a spam bot honeypot project, and as I understand it their initial success was from their DDoS protection product.
But yeah, the idea that they might be losing focus as the infinite growth treadmill grinds on is one that definitely concerns me too, because it happens again and again. I desperately wish companies could just be good at something, make enough money for their employees to be comfortable and their investors to make a steady dividend, and continue to do so forever.
It happened less than three weeks ago (mid-november), so it probably feels perceptibly more frequent even just due to it happening so soon after the last one
My fingers are very firmly crossed that this is just random clustering and not a sign of Cloudflare getting worse.
There was also an AWS outage in us-east-1 earlier this year, which has pretty much the same impact on end users, so that contributes to a recency bias as well even though it's not Cloudflare.
Dont't forget the Azure outages.
To an average person 'computer not working' is the beginning and ending of their troubleshooting capability.
Looks like they're all back up for me, in North America, including DownDetector though it was down for a few minutes... (spoke too soon, WIRED is still down for me.)