Danbooru turns 21 years old today. Happy 21st birthday to Danbooru!
danbooru.donmai.us/posts/1
Danbooru is back!
Politicians, athletes, journalists, NGOs, companies... They all use corporation-owned social networks to communicate with us. Citizens shouldn't have to hand over their data to companies to read content published by public entities. Public Feeds changes that by making posts of popular accounts freely available. Read more.
Danbooru is an English-language imageboard and image hosting website focused primarily on anime-style illustrations. It was launched in 2005 by a programmer known as "Albert" and is frequently described as one of the earliest and most influential "booru" style sites, using collaborative tags to organize images. More on Wikipedia
Danbooru is back!
PS: The tag limit has been raised to 3 if you make an account
We're still not sure when the site will be back up, but expect days at least. The original security issue has been fixed, but currently we're doing a full audit of the site's code to make sure there are no other potential security issues when the site comes back up.
You can see our progress here: github.com/danbooru/danbooru/commits/master/. We're working as fast as we can and we've fixed dozens of issues so far, but there's still many more to fix. If you want to help out, you can help scan for issues yourself and report any you find.
Site down until further notice while we investigate a security incident. Expect a few days at least.
In short: someone uploaded a Javascript worm with an IP grabber that logged the IP address of anyone viewing an infected Flash post.
At this time we believe the only damage done was IP addresses getting leaked for users who viewed Flash posts in a 5 minute window. Still doing a full investigation. Join the Discord for updates: discord.gg/danbooru.
The worm, for the curious: www.decompiler.com/jar/9e7095c4a107358b9d44cc05937f0cc9/b…
Danbooru turns 21 years old today. Happy 21st birthday to Danbooru!
danbooru.donmai.us/posts/1
Site down for planned maintenance. Expected downtime is 2 hours
Site down for planned maintenance. Shouldn't take more than 10 minutes.
Danbooru's 10 millionth post is Mococo! danbooru.donmai.us/posts/10000000
Site is back up. A network switch at the datacenter failed and after that it was discovered one of the PSUs in the database server was going bad
database server is down. no eta.
Danbooru turns 20 years old today. Happy 20th birthday to Danbooru!
danbooru.donmai.us/posts/1
we are so back. the database server had a failed fan and two bad sticks of ram that were replaced
one of the servers had a bad stick of ram. the site will be down while it's being replaced. no idea how long this will take
site down for maintenance
🎉 Happy 19th Birthday to Danbooru! 🎉
danbooru.donmai.us/posts/1
Expect 15 minutes of downtime while the servers are rebooted.
we are so back
Site is down. Possible hardware failure on the database server. Still investigating.
back up
site is down for scheduled maintenance. should be back up in 30 minutes
The site is back up. danbooru.donmai.us/forum_topics/23669
database server was overheating. datacenter is performing maintenance
investigating performance issues
6 million GET
danbooru.donmai.us/posts/6000000
Artists: Email contact@novelai.net to tell NovelAI not to use your works in their AI generator. Danbooru is not affiliated with NovelAI and we can't stop them from using your works in their AI generator.
‼️ RETWEET TO SPREAD AWARENES ‼️
Artists, if your art has been reuploaded to Danbooru, look into getting it taken down ASAP or it will be subjected to being used for AI learning without your consent.
NovelAI is based on Stable Diffusion, which is trained on literal billions of images across the entire internet. Check haveibeentrained.com to see if your works have been included from other sites.
30 minutes of downtime for more database maintenance.
5 minutes of downtime for database maintenance.
Uploads currently down because of a network outage at our hosting provider. status.us.ovhcloud.com/pages/59dd23da8827c804746f1664
Database maintenance. Back in 10 minutes.
Server upgrade complete🎉
Down for a database server upgrade. Back in two hours.
db update. back in 1 min.
deploying update, back in 1-2 minutes
deploying update
deploying update, be back in 2-3 minutes
deploying a quick update
deploying a site update, debugging issues with it
Happy 16th birthday to Danbooru! The first post to Danbooru was uploaded 16 years ago today. danbooru.donmai.us/posts/1
Deploying a quick update
Investigating an outage of the image server. Uploads currently down and some thumbnails failing to load.
Back, it was network maintenance at our hosting provider.
Deploying a quick update to tweak the color scheme danbooru.donmai.us/forum_topics/17913
deploying an update
downtime resolved
currently undergoing an outage, working on fixing it
Deploying a quick update.
Deploying an update, back in a few.
Resolved recent downtime. Site was overloaded, possibly because of a possible DoS attack or badly behaved bot.
site currently under very high load. working on it
deploying a quick update
working on downbooru
issue resolved
Site down while we investigate an outage. Experiencing network connectivity problems with our hosting provider. Join us on Discord while you wait: discord.gg/danbooru
A Cloudflare outage has brought down Danbooru, among other sites. See news.ycombinator.com/item?id=23875692
Down 1-2 minutes for maintenance. Join the Discord while you wait: discord.gg/eSVKkUF
Down for 15-20 minutes while our hosting provider performs a network upgrade. Join us in Discord while you wait: discord.com/invite/8SAF59U.
Danbooru will go down for maintenance for 15-20 minutes two hours from now. Join us in Discord while you wait: discord.gg/8SAF59U
Deploying an update. Expect a few minutes of downtime. Hang with us in Discord while you wait: discord.gg/8SAF59U
Saved searches for gold users now update immediately instead of triggered a delayed update. Depending on performance this may get rolled out to everyone.
migrations are fucked give me an hour
bad postgresql port in the config on reportbooru, which was cascading and causing the popular search report to fail. issue is fixed now.
i think we're good?
I'm adding a dedicated asset server for thumbnails to help with iowait but that code path is uh complicated. Work in progress. I'll also bring another app server online soon.
I know the image sync is broken currently. I've set up a cronjob to run every 10 minutes to sync stuff manually. I'll look into the problem more deeply tomorrow morning.
Site is back up.
Downtime for server migration. Will take 3-4 hours.
Uploads should be fixed
uploads are borked, deploying a fix
site is back
running some quick benchmarks, site will be down for half an hour
fucking around with database replication again, downtime of an hour intermittently
had a real "did i just rm -rf postgres?" moment there
slave database is out of sync so i need to take down the site for an hour or so to recover it
http server hadn't been restarted after ssl certs were renewed. site is back up.
linode restarted all their servers again so IQDB was backlogged. things should be normal again.
experimenting with a new system for uploads. if you use the bookmarklet or drag and drop a file, it should process in the background while you tag. it is probably still a little wonky. please report problems on the forum, or github, or here. thanks!
restoring thumbnails, expect an hour of downtime
looking into the broken thumbnails issue on hijiribe, should be fixed soon
i had two ip ban two bots message me to figure out what went wrong
investigating database slowness
two bugs with tag autocomplete and the forum voting have been fixed
please pardon our dust. some major internal changes were deployed and we're fixing issues as we find them.
uploads are fixed
raid5 partition is scrambled on sonohara. will probably need to rebuild it. will probably take 2-3 days plus a week or so to restore all the back up images
grub on sonohara server is hosed, need to wait until someone at the data center can put in a live cd
one server is down (raid is fucked again), performance will be degraded until it's restored. probably a few days.
IQDB is working again. It may take some time for the worker to process the backlog of images.
another bad hard drive. taking it out of dns for now
i hate mdadm
the sonohara server was down (bad disk probably). i will take it out of dns for now.
letting the background jobs catch up a bit
back. maybe.
some downtime for emergency database maintenance
down for maintenance
some maintenance, should be back in an hour
maintenance is finally over
rsync -a --delete empty/ cache/
this is still taking forever but i've recovered 300GB so far
this is taking forever so i'm going to run at reduce capacity for awhile. if the site is slow you may need to wait for dns to update
deleting some old files, downtime while in io hell
welp this is gonna take longer than expected
doing some database maintenance, will be down for half an hour
reindexing iqdb posts from the past day, let me know on forum or github if any are missing
lots of database timeouts but we just had a major migration. will give it some time to settle before taking action.
admin dashboard is fixed
Deploying meta tags...
first post