Hi All, Last Wednesday there was an airconditioning failure in our machine room. This crashed among other things the dirac database server. Unfortunately the scale of the database corruption only became apparent days later. Our last good backup is from Wednesday 12:00 midday. We have tried various approaches to recover the lost database entries, however there are still large inconsistencies. We now intend to roll back the dirac file catalogue back to Wednesday 1/2/17 12:00. This means that any files you have deleted again will start reappearing in the files catalogue. This is something we can eventually fix by comparing with what is on disk, but it will take a long time to as we'd have to compare each file in the catalogue with the files present on the storage elements. It also means anything you have written since 1/2/17 12:00 will have to be re-registered. (The actual files on the storage elements are unaffected). In some cases (e.g. where the data has been written to Imperial) we will be able to assist this process in a fairly straightforward manner. Other cases are probably less straight forward and we'd ask you to please contact us directly so we can try and fix it. This is clearly not good enough for a production level service, but this came as a surprise to us as well. We have since bought and installed a UPS which should hopefully cover any more power cuts and we have a more robust procedure in place should this happen again. Apologies. Daniela -- Sent from the pit of despair ----------------------------------------------------------- daniela.bauer@imperial.ac.uk HEP Group/Physics Dep Imperial College London, SW7 2BW Tel: +44-(0)20-75947810 http://www.hep.ph.ic.ac.uk/~dbauer/