On Fri, Feb 26, 2010 at 01:08:20PM +0100, Marco d'Itri wrote: > On Feb 21, Marco d'Itri <m...@linux.it> wrote: > > > Well, in computer science we tend to believe in causality. > > If it did not work the first time then it will probably not work the > > second time as well. Anyway: > > > > echo add > /sys/block/sdd/uevent > So, are there any news?
This being a production server, the window for debugging was limited. I had to give up and mknod the device manually. Everything worked from there; I was able to partition the drive and add it to the RAID. I find it odd that MAKEDEV refused to make the device, citing udev (so I went looking for how to tell udev the device really was there), but mknod worked anyway. Obviously not a udev bug as such, but I thought it worth mentioning (wishlist bug to add an override flag to MAKEDEV?). We use the same Supermicro motherboard & 1U case w/ SATA hotswap for some of our customer boxes. Next time I have new hardware in or a customer box for maintenance (rahter than someone remotely swapping drives for me on colo'ed production box), I'll attempt to reproduce the problem and provide the desired debugging information. In the mean-time, Supermicro servers aren't exactly rare, so perhaps someone will beat me to it. Suggest tagging moreinfo. Thanks. -- _ivan -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
