On 2020-04-03 18:16:33, Jonathan Ballet wrote: > Hi, > > On AWS Nitro instances, EBS volumes are exposed as NVMe block devices by > the kernel on the /dev/nvme* paths. > > The AWS documentation "Identifying the EBS Device" says that the Linux > kernel doesn't guarantee the creation order of these devices, as they > are discovered in the order the devices respond. > > This creates a situation where you can start an instance with 2 (or > more) attached EBS and sometimes end up with the "root" EBS being named > /dev/nvme0n1 and the other one /dev/nvme1n1, sometimes the opposite. > > (It's also even more confusing because the device names specified when > attaching the disks when creating the instance is also different...) > > In its own Linux image, AWS apparently ships a set of udev rules + a > script to get these device names out of the information exposed by the > NVMe devices, which create appropriate symlinks in /dev towards each > /dev/nvme* devices. > > This doesn't change the random detection order by the kernel, but this > provides a more stable interface to deal with instances containing > multiple disks. > > I wonder if it would be possible to provide such a mechanism by default > in the official Debian AMI? > Is this type of cloud provider-specific configuration could be accepted? > (I guess that would make sense only for AWS images.) I remember there was similar case with non-deterministic naming convention for ephemeral EBSs long time ago. We took similar approach and devised script doing exactly that thing, but we did it only to our and our customer images. Keeping that in mind I suppose I'm not going to oppose too much if handling of that case would be added to EC2 class. -- |_|0|_| | |_|_|0| "Panta rei" | |0|0|0| -------- kuLa -------- | gpg --keyserver pgp.mit.edu --recv-keys 0x686930DD58C338B3 3DF1 A4DF C732 4688 38BC F121 6869 30DD 58C3 38B3
Attachment:
signature.asc
Description: PGP signature