Comment 9 for bug 365944

Sorry for not opening a new bug. I will do so next time. It's just too hard to recognize what belongs together (same cause) or just looks the same (same symptoms). Please understand: I'm supposed to search for existing bugs (and add to those) to not produce unnecessary duplicates and on the other hand I'm supposed not to mix things together that don't belong together. That's much easier for you, that you have more insight. Also Alessio reported this half a year ago, so I expected that nothing will happen with this bug anymore anyway if noone adds new information.

Back ontopic: At least for my case i found the root-cause: Some time ago I used "fix partition ordering" of fdisk and there are known incompatibilites between that function and parted. See for example http://www.google.at/search?hl=de&q="Error%3A+Unable+to+satisfy+all+constraints+on+the+partition."+fdisk+fix+order . Of course I didn't know that partman is using parted or libparted until a few hours ago. I don't know if fdisk or parted is to blame now, but it's one of those two. One solution/workaround (the only that I've found) is in the last post here: http://parted.alioth.debian.org/cgi-bin/trac.cgi/ticket/201 . But be very careful with that: In the process of fixing this problem I lost one of my partitions due to overwriting the LUKS-header with a partition table.

I don't know if I was supposed to put that comment already into another, new bug? But I think it's not too bad here, because if I put it into something like "fdisk/parted-incompatibility" noone would find it, if he encounters the same symptoms that i did.