This will affect Jabber itself and all transport services, and also this website. And it will take place during the next 24 to 48 hours. Please be patient and expect some service outages. We changed the TTL of our DNS zones to 1 minute but nevertheless it could be that your client still looks for the old server. You might need to relogin to the Internet and/or restart your client.
We will be back here when everything is moved. Until then you also can reach us on Twitter, where a link to this post will also be posted.
The Spectrum transport services are down at the moment. It will take one to three days until these services are up again. Mostly because we are about to move all Jabber related services to a new machine. So please stay tuned.
There were some outages in the past weeks.
A big outage was from yesterday to today, almost one day. Coming back just two hours ago. But everything is fine again. We had some server hardware problems which are solved now. No data loss and no need to restore our backups.
The other outages are much harder to detect. There is a annoying bug in Prosody at the moment which causes database connections to fail at some point. Then a restart is needed. We need the information then (because most of the time it only affects users who want to login, not already logged in users) to restart the server. Another bug which is unknown at the moment is that connections to jabber.ccc.de are failing from time to time. A restart solves it – or causes it. You choose… ;-) But these problems are Prosody related (the first for sure, the second maybe) and we work on it.
So we need your help. If you cannot login, tell us. Here or Twitter, does not matter. If you see your buddies from other servers offline at once, tell us. At the moment we just know about this jabber.ccc.de problem which happened two times.
PS: We are not thinking about cancelling our Jabber services here, do not worry! Some people were afraid.
We upgraded the Jabber server to Prosody 0.9.4 this morning, see the changelog.
We just upgraded the Jabber server to Prosody 0.9.2, see the changelog.
The most important bugfix is the bug related to zlib compression used with encrypted connections. It seems that this is finally fixed! :-)
Due to another two hard disk failures all Jabber services and this website will be down for some time. All Jabber services will go down at about 5:30pm CET, the website will follow at about 8:00pm. We need time to backup all databases and Jabber server data before restoring both hard disks of our RAID1 array… Unbelievable, but that’s how it is…
The services will be back as soon as possible. Everything from one hour to five hours is possible. Please stay tuned…
We upgraded the Jabber server from Prosody 0.8 to 0.9 today! This brings a lot of new features, I just want to name IPv6 support and a Web Admin interface for our Jabber hosting customers. :-)
After migrating from ejabberd to Prosody we had to dump IPv6, which is now coming back. We are very happy about that and cannot wait how statistics will change after we support it again. ;-)
Also we installed several new modules with some new features.
A big thank you to Matt, the chief maintainer of Prosody! :-) The server upgrade only took about 20 minutes until the server was up and running again. Without him this would not have been possible. We run some special configs which had to be taken care of.
After extreme hardware problems on the server machine everything is back up and running again.
Both hard disks crashed so we had to restore the backup from January, 9th 2013, which was about two days old. Sorry about this, but the backup from January, 10th was compomised with most likely corrupt data because of the dying hard disks.
Our server provider Hetzner keeps telling us that hard disks are expendable items and that it is completly normal that there are hard disk crashs. Well, 5 disks in 6 months, we do not believe this. Also because we are running root servers for almost 10 years and had 1 or 2 failing disks until we got this machine here…
Guys, sorry again!