Sorry in advance if this is the wrong place for this kind of question.
I'm porting some of my EC2 application's start scripts to the 11.10
ami-21f53948 AMI, from a similar 9.10 (
One step of the process is to raid together the 4 ephemeral drives that come with a
c1.xlarge. In the the previous AMI these could be found at
/dev/sd[b-e]. These devices appear to be absent in the new AMI.
It seems an instance using the new AMI believes they should be there:
$ curl 'http://169.254.169.254/latest/meta-data/block-device-mapping/' && echo ami ephemeral0 ephemeral1 ephemeral2 ephemeral3 root $ curl 'http://169.254.169.254/latest/meta-data/block-device-mapping/ephemeral0' && echo sdb $ curl 'http://169.254.169.254/latest/meta-data/block-device-mapping/ephemeral1' && echo sdc $ curl 'http://169.254.169.254/latest/meta-data/block-device-mapping/ephemeral2' && echo sdd $ curl 'http://169.254.169.254/latest/meta-data/block-device-mapping/ephemeral3' && echo sde
Does anyone have any light to shed on what I can do here? Perhaps they are available under different names (despite the block device mapping) or maybe I can create the needed device files? Any help is appreciated.
While it's not clear exactly how, this issue report may be related. Much like the goggles, passing the block device mapping arguments to
ec2-start-instances (as one post suggested) did nothing.