VHFFS Database crash

Posté le lundi 18 décembre 2006, à 15:57 UTC

Oops!, our server hosting the vhffs database has just crashed, we hope a power cycle will be enough to restart it.

Impacted services : Primary mail (smtp,pop,imap,webmail,mailing lists…), Primary DNS, Probably CVS and Subversion, and Jabber

Edit: Well.... the server is dead, two disks has been kicked out of the raid5 storage array, we moved the services on another server and recovered the vhffs database with a snapshot of the datas at Mon Dec 18 16:00:00 CET 2006, if you have made any changes in the panel between Mon Dec 18 16:00:00 CET 2006 and Mon Dec 18 17:40:00 CET 2006 you are invited to make those changes again, everything is back online.

Public mailing lists archives are now generated by MHonArc

Posté le dimanche 17 décembre 2006, à 21:17 UTC

Hi, some of you asked for a good public mailing list archives system, so mailing lists archives are now generated by MHonArc and are still available at the same address.

We are still looking for a way to generate global statistics for months and years like gnome as others does, if you know how to do this, contact us !

Files systems optimizations

Posté le lundi 11 décembre 2006, à 14:11 UTC

Hi, we are going to improve the performance on several files systems in the next few minutes, all services will be impacted...

Edit: Everything is back online, we will see the performance gain in the next hours, tell us if anything is wrong, strange, etc. for you :)

Download service outage

Posté le mardi 5 décembre 2006, à 13:34 UTC

Hi, it seems there was a network problem on the ISP hosting the downloads service for 2 hours.

Everything is back on line ;)

Server in inconsistent state

Posté le lundi 4 décembre 2006, à 19:50 UTC

Hi, we just had kernel issues with one of our servers, some services like primary mail exchanger, pop, imap, primary dns server, panel, jabber, webmail, ... were down.

Everything is back online !

Sorry for the inconvenience ;)

Mail service DDoSed...

Posté le samedi 2 décembre 2006, à 12:20 UTC

The mail service is currently suffering under a DDoS, it caused the whole plateform to crash because we had reached the maximum connections on a database server, this problem is now solved.

Now everything works correctly except the mail, the primary MX has been shut down so we just have to wait now...

Edit: Well, in fact, a file system issue was responsible of this problem, wrong analyse, we will do better later ;-)