> > "Bloody cushty mate" ;) > Nice :) Glad to hear you got it worked out... RHEL kernels do have ext4 enabled by default but in 5.4 (I think....) it was labelled ext4dev to indicate that there was the *possibility* of changes to it as a development system that could potentially result in manual steps required due to low level changes - for example see how the btrfs stuff changed at one point in the main kernel development line.... In 5.5 (I think... would need to double check the relevant release notes to be sure... I may be one point release in front) the label was changed from ext4dev to ext4 to indicate that no low level changes would occur any more (ie the structure of the filesystem was finalized) however it was still considered a technology preview.... There was something (a tunefs change needed if I recall correctly off the top of my head) to allow old ext4dev systems to be relabelled as ext4 systems to be recognised and mountable.... That's the blah of it ;) Of course RHEL6 will have ext4 by default and not just supported (IIRC) and I imagine that the next point release in the 5 series will take it from technology preview to a formally supported filesystem... Is disk I/O an issue for you? I'll be testing RHEL6 beta 2 with our applications in the next few weeks and I've been wondering whether it would be wise to tune the mount options to improve disk I/O as it is pretty intensive for us at times (lots of read and writing in functional testing)... I recall when ext4 first came with Ubuntu Karmic I had to tune the mount options to have our stuff perform sensibly on the development desktops here... James