Someone in another post showed a decent site that shows cloudflare dependency, so I threw the top 35 actives from observer into it. It's honestly not as bad as I expected it would be, but it is concerning.
It looks like that source also didn't list hexbear.net, which seems to be around the 5th or 6th largest Lemmy instance, at 1.57k monthly active users. Results from the site you linked (thanks, by the way) say that they use Cloudflare for DNS but not for CDN/Proxy/SSL.
Hexbear and lemmynsfw aren't listed on fedidb either, which leads me to believe that they must block whatever discovery mechanism is being used by these fedi stats sites.
You can browse anonymously on any instance that permits doing so, so if you just want to browse during an outage, you can do that anywhere.
IMHO, having an account on a second Threadiverse instance isn't necessarily a terrible idea, not just because of CloudFlare outages, but because instances do have outages for various reasons. I have an account on olio.cafe (PieFed, not on CloudFlare) and on lemmy.today (Lemmy, not on CloudFlare) because I wanted to try out PieFed, and I have fallen back to that to post before if lemmy.today has issues.
That being said, I didn't intentionally try to avoid CloudFlare. I mean, they're used by a lot of major sites, and I don't expect them to have a lot of downtime. I mean, every Threadiverse instance has had downtime for some reason or another. I've had Internet outages, as well as electricity outages. Not all that common or usually an extended thing, but they happen.
While the downtime was most active, most of the top instances were down. Lemmy.ml, feddit.org, discuss.tchncs.de and behaw.org were all up. You can use https://lemmyverse.net/ to browse things and the ones offline all show a “content error” in lemmyverse.
Here is a screenshot that @straycatstrut@discuss.tchncs.de took during the outage
I just came home and learned about the outage. Still can't get the smirk of superiority out of my face from when I noticed my people were still able to shitpost in my absence.
sopuli.xyz didn't have any issues. At least I could access it at any time, I don't know if there is any lag in federation due to unavailability of other instances.
Another important consideration is that Cloudflare outage might have been regional. Same content could be available in one region but not in another. On the Cloudflare status page they mention issues with London specifically.
So, there are a couple of reasons to use CloudFlare, but I suspect that the reason that a lot of people are doing so is to deal with DDoSes, which are hard to deal with otherwise.
Like, my home instance, lemmy.today, doesn't use CloudFlare, so it isn't affected by a CloudFlare outage. But...it was also knocked offline for a few days about a month back by a DDoS.
A lot of major sites do depend on CloudFlare, so they probably aren't going to have a horrendous amount of downtime --- like, any issue that comes up is probably gonna have a lot of engineers banging on it pretty quickly.
I just think clients should auto fail over to another instance. You should be able to pull posts from somewhere else but just be unable to post. No reason we can't pull posts though.
Mine was. And ran better than ever because it didn't have to handle so many updates anymore. I missed the memes though, because the images weren't loading.
56 Comments
reallyzen@lemmy.ml · 39 pts · 258d
Uh? What outage? When?
WalrusDragonOnABike@reddthat.com · 23 pts · 258d
Cloudflare outage. A few hours ago.
Pika@sh.itjust.works · 28 pts · 258d
Someone in another post showed a decent site that shows cloudflare dependency, so I threw the top 35 actives from observer into it. It's honestly not as bad as I expected it would be, but it is concerning.
::: spoiler Lemmy lemmy.world(Active Users: 14512): Cloudflare? Yes(dns,proxy,cname)
sh.itjust.works(Active Users: 2509): Cloudflare? Yes(dns,proxy,cname)
lemmy.ml(Active Users: 2087): Cloudflare? No
lemmy.zip(Active Users: 1704): Cloudflare? Yes(dns,proxy,cname)
lemmy.dbzer0.com(Active Users: 1444): Cloudflare? No
lemmy.ca(Active Users: 1381): Cloudflare? Yes(dns,proxy,cname)
lemmygrad.ml(Active Users: 972): Cloudflare? No
lemmy.blahaj.zone(Active Users: 956): Cloudflare? No
programming.dev(Active Users: 929): Cloudflare? Yes(dns,proxy,cname)
discuss.tchncs.de(Active Users: 778): Cloudflare? No
sopuli.xyz(Active Users: 596): Cloudflare? No
slrpnk.net(Active Users: 371): Cloudflare? No
infosec.pub(Active Users: 331): Cloudflare? No
lemmy.today(Active Users: 314): Cloudflare? No
midwest.social(Active Users: 307): Cloudflare? No
reddthat.com(Active Users: 292): Cloudflare? Yes(dns,proxy,cname)
feddit.nl(Active Users: 290): Cloudflare? No
pawb.social(Active Users: 243): Cloudflare? Yes(dns,proxy,cname)
forum.guncadindex.com(Active Users: 234): Cloudflare? No
mander.xyz(Active Users: 194): Cloudflare? No
lemmings.world(Active Users: 177): Cloudflare? No
ani.social(Active Users: 173): Cloudflare? Yes(dns,proxy,cname)
feddit.it(Active Users: 158): Cloudflare? No
startrek.website(Active Users: 156): Cloudflare? No
feddit.dk(Active Users: 151): Cloudflare? Yes(dns,cname)
leminal.space(Active Users: 126): Cloudflare? Yes(dns,proxy,cname)
ttrpg.network(Active Users: 125): Cloudflare? No
szmer.info(Active Users: 116): Cloudflare? No
lemmy.eco.br(Active Users: 99): Cloudflare? Yes(dns,proxy,cname)
lemy.lol(Active Users: 97): Cloudflare? Yes(dns,cname)
awful.systems(Active Users: 90): Cloudflare? No :::
davidgro@lemmy.world · 15 pts · 258d
I think you missed the second biggest instance (at least it was a while back.)
lemmynsfw.com is using Cloudflare DNS, CDN/Proxy
Edit:
Also noticed:
checkforcloudflare.selesti.com is using Cloudflare DNS, CDN/Proxy.
Pika@sh.itjust.works · 5 pts · 258d
I used observer and sorted by active users, so maybe fediverse observer doesn't host the instance due to the NSFW? idk
Tangent5280@lemmy.world · 2 pts · 258d
Not the goonercave too, Say it ain't so.
Cricket@lemmy.zip · 6 pts · 258d
It looks like that source also didn't list hexbear.net, which seems to be around the 5th or 6th largest Lemmy instance, at 1.57k monthly active users. Results from the site you linked (thanks, by the way) say that they use Cloudflare for DNS but not for CDN/Proxy/SSL.
Hexbear and lemmynsfw aren't listed on fedidb either, which leads me to believe that they must block whatever discovery mechanism is being used by these fedi stats sites.
Pika@sh.itjust.works · 5 pts · 258d
Yea that would make sense, if that endpoint is blocked then it wouldn't be added.
iii@mander.xyz · 17 pts · 258d
Everything looked fine to me. I'm on mander.xyz
cheeseburger@piefed.ca · 17 pts · 258d
dbzer0 was not affected and even had a post about it.
ladel@feddit.uk · 15 pts · 258d
I didn't notice anything on feddit.uk
Akasazh@feddit.nl · 3 pts · 258d
Nor feddit.nl
Skavau@piefed.social · 15 pts · 258d
Piefed was down. Or at least piefed.social.
mesamunefire@piefed.social · 7 pts · 258d
Yeah it uses cloudflare for a number of services if I recall.
MalReynolds@piefed.social · 3 pts · 258d
FWIW I switched VPN egress to Singapore (I have two egress points up permanently and switch with foxyproxy) and was up again immediately.
tal@lemmy.today · 13 pts · 258d
I mean, it's easy to check whether a given instance is using CloudFlare.
You can browse anonymously on any instance that permits doing so, so if you just want to browse during an outage, you can do that anywhere.
IMHO, having an account on a second Threadiverse instance isn't necessarily a terrible idea, not just because of CloudFlare outages, but because instances do have outages for various reasons. I have an account on olio.cafe (PieFed, not on CloudFlare) and on lemmy.today (Lemmy, not on CloudFlare) because I wanted to try out PieFed, and I have fallen back to that to post before if lemmy.today has issues.
That being said, I didn't intentionally try to avoid CloudFlare. I mean, they're used by a lot of major sites, and I don't expect them to have a lot of downtime. I mean, every Threadiverse instance has had downtime for some reason or another. I've had Internet outages, as well as electricity outages. Not all that common or usually an extended thing, but they happen.
yessikg@fedia.io · 13 pts · 258d
I'm pretty sure fedia.io was up the whole time
MyTurtleSwimsUpsideDown@fedia.io · 3 pts · 258d
When was the cloudflare outage? I saw the memes about it on Fedia, but I have no clue if it was ongoing at the time.
yessikg@fedia.io · 3 pts · 258d
Early this morning, from 6 am EST, I think
MyTurtleSwimsUpsideDown@fedia.io · 3 pts · 258d
Thank you! I didn’t get to the web till sometime after 7 EST but it fedia was fine then.
otter@lemmy.ca · 11 pts · 258d
This other post has some discussion: https://lemmy.ca/post/55386786
Here is a screenshot that @straycatstrut@discuss.tchncs.de took during the outage
passepartout@feddit.org · 11 pts · 258d
I just came home and learned about the outage. Still can't get the smirk of superiority out of my face from when I noticed my people were still able to shitpost in my absence.
otter@lemmy.ca · 5 pts · 258d
I appreciate that different teams are doing different things, I'll have to remember to drop by during the next cloudflare outage 😄
guynamedzero@piefed.zeromedia.vip · 10 pts · 258d
I just moved my instance off of cloudflare, you’re all welcome to make an account here!
tobogganablaze@lemmus.org · 10 pts · 258d
lemmus.org was online the entire time.
MadMadBunny@lemmy.ca · 9 pts · 258d
Everything was down for me. Lemmy.ca using Voyager.
otter@lemmy.ca · 9 pts · 258d
We rely on Cloudflare for our instances, so unfortunately lemmy.ca, piefed.ca, and pixelfed.ca were all down during the outage
If anyone is curious, we've discussed why we use Cloudflare here: https://lemmy.ca/post/40252238/15009722
karashta@piefed.social · 7 pts · 258d
sopuli.xyz and fedia.io both stayed up for me
Retro_unlimited@lemmy.world · 6 pts · 258d
Slrpnk worked, I had to use that during the outage.
markz@suppo.fi · 6 pts · 258d
While you were away I was browsing cloudflare memes
IcedRaktajino@startrek.website · 5 pts · 258d
Startrek.website :)
flamingo_pinyata@sopuli.xyz · 5 pts · 258d
sopuli.xyz didn't have any issues. At least I could access it at any time, I don't know if there is any lag in federation due to unavailability of other instances.
Another important consideration is that Cloudflare outage might have been regional. Same content could be available in one region but not in another. On the Cloudflare status page they mention issues with London specifically.
asudox@lemmy.asudox.dev · 5 pts · 258d
mine was
tux0r@feddit.org · 5 pts · 258d
Most.
ekZepp@lemmy.world · 6 pts · 258d
Then tell me which one. Because all of mine was offline.
tux0r@feddit.org · 10 pts · 258d
Mine (feddit.org) was not.
MCTamTam@feddit.org · 5 pts · 258d
Yep❤️
TheMinions@lemmy.dbzer0.com · 5 pts · 258d
Lemmy.world uses Cloudflare.
ekZepp@lemmy.world · 2 pts · 258d
Yep. I kinda realize that.
chunes@lemmy.world · 5 pts · 258d
Say it with me again: your server does not need middlemen that can break your stuff whenever they feel like it.
tal@lemmy.today · 15 pts · 258d
So, there are a couple of reasons to use CloudFlare, but I suspect that the reason that a lot of people are doing so is to deal with DDoSes, which are hard to deal with otherwise.
Like, my home instance, lemmy.today, doesn't use CloudFlare, so it isn't affected by a CloudFlare outage. But...it was also knocked offline for a few days about a month back by a DDoS.
A lot of major sites do depend on CloudFlare, so they probably aren't going to have a horrendous amount of downtime --- like, any issue that comes up is probably gonna have a lot of engineers banging on it pretty quickly.
early_riser@lemmy.world · 1 pts · 256d
I self host a wiki and use Cloudflare as a reverse proxy to protect my IP. If you know of a similarly easy to use solution I'd love to hear it.
_cryptagion@anarchist.nexus · 0 pts · 258d
Sergio@lemmy.world · 4 pts · 258d
baduhai@sopuli.xyz · 4 pts · 258d
sopuli.xyz was fine.
Pika@sh.itjust.works · 4 pts · 258d
sh.itjust.works was down during it as well.
DrDystopia@lemy.lol · 3 pts · 258d
My private archiving instance. But assuming there's still no users and federtion was broken, onlineity doesn't really matter.
TheImpressiveX@lemmy.today · 3 pts · 258d
Not Lemmy, but piefed.social was offline due to Cloudflare.
Limfjorden@feddit.dk · 2 pts · 258d
feddit.dk was unaffected.
ThePantser@sh.itjust.works · 2 pts · 258d
I just think clients should auto fail over to another instance. You should be able to pull posts from somewhere else but just be unable to post. No reason we can't pull posts though.
tal@lemmy.today · 2 pts · 258d
You can monitor instance downtime at:
https://lestat.org/
AlexisFR@jlai.lu · 1 pts · 258d
Ours was ;)
Pazintach@discuss.tchncs.de · 1 pts · 257d
discuss.tchncs.de was fine.
bjoern_tantau@swg-empire.de · 1 pts · 258d
Mine was. And ran better than ever because it didn't have to handle so many updates anymore. I missed the memes though, because the images weren't loading.
hexagonwin@lemmy.sdf.org · 1 pts · 258d
mine was online, but it's pretty buggy recently.
Blaze@piefed.zip · 1 pts · 257d
https://beehaw.org/post/23261600
scroll_responsibly@lemmy.sdf.org · 1 pts · 258d
Surprisingly, lemmy.sdf.org. was up… not that SDF would use cloud flare.