Cannot build mesos image with docker
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Magnum |
Fix Released
|
Undecided
|
Egor Guz |
Bug Description
Here is my environment:
OS: Ubuntu14.04
RAM: 2G
Disk: 20G
I follow the instruction (https:/
+ /usr/sbin/
Installing for i386-pc platform.
/usr/sbin/
/usr/sbin/
/usr/sbin/
/dev/loop0: [0033]:2440 (/tmp/image.
Here is the full logs: http://
Changed in magnum: | |
assignee: | nobody → Egor Guz (eghobo) |
status: | New → Confirmed |
Changed in magnum: | |
milestone: | none → mitaka-1 |
status: | Fix Committed → Fix Released |
The full logs appear to be incomplete and missing the error and the logs I would like to see where it's trying to partition the loop0 device.
I was receiving this error which I was able to resolve by installing `parted` into the Image ( https:/ /github. com/openstack/ magnum/ blob/master/ magnum/ templates/ heat-mesos/ Dockerfile# L5 ). Since adding that to the Dockerfile and using vivid I have been successfully building the mesos image.
As far as I could tell from the logs at the time it wasn't rewriting the partitions to remove ext2 from where it was installing the bootloader because it was trying to use parted to do that rewrite.
I wonder if there's something slightly different how trusty and vivid manage the loop devices and the way its mapped into the container via --privileged. I run a Vivid desktop and I had to use a Vivid container ... Could you try either setting the FROM to trusty ( assuming that's what you're using ) and/or running this in a Vivid VM ?