Closed Hermanverschooten closed 8 years ago
@Hermanverschooten (long story short) go along something like ...
dd if=/dev/zero of=extra.img bs=1G count=16
coreos-xhyve-run
after the LPC_DEV
line ...
IMG_HDD="-s 4,virtio-blk,extra.img"
should work...
@AntonioMeireles I already tried that way, coreos just crashes on the boot
@rimusz i didn't have too much time to look (busy atm) but, at sight, i suspect that masking the lvm services may do the trick ... (lvm2-{lvmetad,monitor,pvscan@}) as are them that seem to be messing with this.
@AntonioMeireles no worries, as I do not have much time to play with it too. I just gave a shot
Just tried this as well. Stops while trying to start udev on boot.
@jbergstroem it works (tested with tip (built from source) xhyve). trick is to just format the volume in advance ...
Ah, obviously. Now to conveniently doing that on a mac :)
brew install e2fsprogs
:smiley:
Thanks man, now I feel slightly RTFM'ed!
..actually, it still freezes:
$ /usr/local/opt/e2fsprogs/sbin/fsck.ext4 ~/vm/coreos/extra.img
e2fsck 1.42.12 (29-Aug-2014)
coreos-extra: clean, 11/1048576 files, 109841/4194304 blocks
$ sudo MEMORY=2048 IMG_HDD="-s 4,virtio-blk,/Users/jbergstroem/vm/coreos/extra.img" \
SSHKEY="$(cat ~/.ssh/foo.pub)" ./coreos-xhyve-run
<snip>
[ OK ] Started Rebuild Dynamic Linker Cache.
[ OK ] Started Rebuild Hardware Database.
Starting udev Coldplug all Devices...
[ OK ] Started udev Coldplug all Devices.
Starting udev Wait for Complete Device Initialization...
[ OK ] Found device /dev/ttyS0.
[ *** ] A start job is running for udev Wai...vice Initialisation (59s / 3min)./coreos-xhyve-run: line 46: 92205 Killed: 9
I've tried the beta channel to no avail.
@jbergstroem humm, are you using a xhyve binary built from a xhyve git tip checkout (and not the one, right now, in homebrew) ?
brew install --head xhyve
edit coreos-xhyve-run and in CMDLINE cut acpi=off
and try again plz. (that one is not yet merged - it's part of #34)
..and I'm in. Thanks, again @AntonioMeireles.
great to hear! thanks to you for your patience!
@AntonioMeireles @jbergstroem no needs to pre-format the extra.img with e2fsprogs
.
I use format-ephemeral.service
to check if the /dev/vda
is formatted, if not it gets formatted.
This is the snippet from my user-data:
...
- name: fleet.service
command: start
- name: format-ephemeral.service
command: start
content: |
[Unit]
Description=Formats the ephemeral drive
[Service]
Type=oneshot
RemainAfterExit=yes
ExecStart=/bin/sh -c '/usr/sbin/blkid -pi /dev/vda | grep TYPE= || /usr/sbin/mkfs.ext4 /dev/vda'
- name: var-lib-docker.mount
command: start
content: |
[Unit]
Description=Mount ephemeral to /var/lib/docker
Requires=format-ephemeral.service
After=format-ephemeral.service
Before=docker.service
[Mount]
What=/dev/vda
Where=/var/lib/docker
Type=ext4
- name: docker-tcp.socket
...
When I try to run one of my docker-composed containers I quickly run out of space. I see the disk is currently only 512Mb, is there a way to increase this?