QEMU has native support for Ceph block devices. Starting KVM with an RBD drive looks like this:
file=rbd:rbd/jessie:id=admin:key=xxx==:auth_supported=cephx\;none:mon_host=node1.example.com\:6789\;node2.example.com\:6789\;node3.example.com\:6789,if=none,id=drive-virtio-disk0,format=raw -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1
This could be supported by allowing to pass qemu custom args and making artifact_source optional.
Hey,
There is definitely a need for richer volume support in Nomad. We would like to tackle it in a way that maintains consistency across drivers. Just wanted to comment that this is definitely something we are aware of!
Any updates on this, @dadgar , or is it still just a radar item?
No updates yet! It is a large feature and we want to make sure it is done right!
Unfortunately 'make sure it is done right' often ends in 'is never done at all'. Pity :-(
Most helpful comment
Hey,
There is definitely a need for richer volume support in Nomad. We would like to tackle it in a way that maintains consistency across drivers. Just wanted to comment that this is definitely something we are aware of!