03:00AM A number of servers are offline, due to a problem affecting Hetzner’s Samrand datacenter.
Update: Hetzner informs me that the issue is due to a power outage, affecting “a large number of rows [of servers]” and that their technicians are working on the problem.
05:40AM Hetzner appears to have fixed their problem, and servers are booting up. There will be degraded performance (sluggish websites, timeouts) until disk I/O settles down within 30 minutes or so.
13:20 We’re seeing timeouts and slow responses from websites hosted on Bumblebee, and are investigating the problem.
13:30 The problem has been found and fixed, and everything should be back to normal.
9:50PM We’re working hard to mitigate a DDOS targeting two sites on this server. Sites on this server will be unresponsive until the attack has been stopped.
10:20PM Almost there, we’ve stopped 90% of the attack, now blocking the last sources.
10:33PM Things are almost back to normal. Server iowait is a bit high, so sites will be a bit slow until that is resolved. Thanks for your patience, and apologies for this inconvenience.
21 September 2:50PM
Server Bumblebee 22.214.171.124 is offline. Technicians are investigating.
3:15PM The problem has been found, and technicians are working to fix the hardware fault.
3:26PM The problem has been fixed, and the server will be booted up shortly
3:37PM Still waiting for the server to be booted.
3:45 Server is booting up now. For the next 10-20 minutes or so, websites and emails will be slow while disk IO is high after the reboot.
17 September 2018 8:38PM
We’re working to mitigate a DDOS which is preventing the server from serving sites successfully.
The DDOS has been mitigated. Apologies for the inconvenience.
10:50AM Apache on Lannister is encountering an error, and is not serving any websites on this server. We’re working on the problem.
10:57AM The problem has been fixed, and Apache restarted. All websites should be functioning correctly now.
14:00 Greyworm is currently offline due to a DDOS attack. We’re working on mitigating it.
14:17 We’ve null-routed the target of the attack, and this seems to have stopped the DDOS. Unless another site on this server is targeted, the server should now run normally. Please accept our apologies for this downtime.
17:15 Hetzner is currently experiencing a routing issue in the Samrand data-centre. Connections to some of our servers hosted in that data-centre are affected. Hetzner engineers are working on the problem.
17:28 The problem has been resolved.
06:54 AM The server Littlefinger (126.96.36.199) is currently experiencing high load which is slowing websites to a crawl. We are working on fixing the issue.
07:38 AM Server load has dropped to acceptable levels. We will be monitoring to ensure that the problem doesn’t return.
Baratheon is currently experiencing a problem, with some sites not loading.
We are investigating.
I have found a temporary fix, until cPanel can provide a fix or patch for the problem.
My “temporary” fix appears to be the only fix for this, until Comodo sorts their problem out. Response from cPanel support:
I can see you've already corrected this issue:
# grep -i stapl /usr/local/apache/conf/includes/pre_virtualhost_global.conf
Unfortunately, the 'ocsp.comodoca.com' server is currently facing issues, which would prevent SSL sites from being served. You've disabled SSL Stapling properly. Unfortunately, there isn't much else we can do until Comodo investigates why their OCSP server is not responding.
I certainly apologize for the inconvenience. Please let me know if you have anymore questions or issues.