Discussion:
Setting up containers with multiple logical volumes
(too old to reply)
Joshua Schaeffer
2014-02-11 15:11:09 UTC
Permalink
I'm trying to figure out how to create an LXC using multiple LVM logical
volumes. I'd like to dohave the root filesystem of the LXC on its own
LV, the /var directory of the LXC on its own LV, and the /tmp directory
of the LXC on its own LV. Here is what I've done sofaron the host:

=====================================
root at reaver:~# vgdisplay vg_lxc1
--- Volume group ---
VG Name vg_lxc1
System ID
Format lvm2
Metadata Areas 1
Metadata Sequence No 21
VG Access read/write
VG Status resizable
MAX LV 0
Cur LV 4
Open LV 3
Max PV 0
Cur PV 1
Act PV 1
VG Size 261.70 GiB
PE Size 4.00 MiB
Total PE 66994
Alloc PE / Size 11264 / 44.00 GiB
Free PE / Size 55730 / 217.70 GiB
VG UUID Zb2ume-1Ty9-s8Ze-NEZc-7Vee-A3h7-A2IvGd
=====================================

=====================================
root at reaver:~# lvdisplay vg_lxc1
--- Logical volume ---
LV Path /dev/vg_lxc1/lv_ldap_var1
LV Name lv_ldap_var1
VG Name vg_lxc1
LV UUID VfCZgf-ZOGA-lJgn-GFO8-fOAd-IMzl-QF32aY
LV Write Access read/write
LV Creation host, time reaver, 2013-11-30 13:46:11 -0700
LV Status available
# open 1
LV Size 10.00 GiB
Current LE 2560
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:3

--- Logical volume ---
LV Path /dev/vg_lxc1/lv_ldap_tmp1
LV Name lv_ldap_tmp1
VG Name vg_lxc1
LV UUID GDru3y-oLJB-Iv06-tjv3-wHuq-p8Fi-xBUscf
LV Write Access read/write
LV Creation host, time reaver, 2013-11-30 13:46:16 -0700
LV Status available
# open 1
LV Size 4.00 GiB
Current LE 1024
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:4

--- Logical volume ---
LV Path /dev/vg_lxc1/lv_ldap_root1
LV Name lv_ldap_root1
VG Name vg_lxc1
LV UUID u1tAUH-t7CB-tEdp-M84c-H93p-9CEn-7xXzAM
LV Write Access read/write
LV Creation host, time reaver, 2014-02-10 20:19:30 -0700
LV Status available
# open 1
LV Size 10.00 GiB
Current LE 2560
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:2
=====================================

I manually created the LV's that I would use for /tmp (lv_ldap_tmp1) and
/var (lv_ldap_var1), but the LV for the rootfs (lv_ldap_root1) I created
using the lxc-create command:

=====================================
lxc-create -n ldap_baneling -t debian -B lvm --vgname vg_lxc1 --lvname
lv_ldap_root1 --fstype xfs --fssize 10G
=====================================

This seems to work correctly for the rootfs, but I can't seem to get the
container to use the /var and /tmp LV's. This is what I've changed in
my container's config file:

=====================================
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
=====================================

I've mounted the LV's under their respective directories. However, when
I run df /var or df /tmp under the container it appears that that those
directories are still mounted under the rootfs. Am I just going about
this incorrectly? Any direction would be appreciated.

Thanks,
Josh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140211/547feb13/attachment.html>
Alvaro Miranda Aguilera
2014-02-11 20:13:52 UTC
Permalink
On Wed, Feb 12, 2014 at 4:11 AM, Joshua Schaeffer
Post by Joshua Schaeffer
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
you are mounting the same path in the same path?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140212/1258a5ed/attachment.html>
Joshua Schaeffer
2014-02-11 20:34:52 UTC
Permalink
Based on the documentation I read, this can be the same, however all I
really care about is that the LV gets mounted to that location on the host.


On Tue, Feb 11, 2014 at 1:13 PM, Alvaro Miranda Aguilera
On Wed, Feb 12, 2014 at 4:11 AM, Joshua Schaeffer <
Post by Joshua Schaeffer
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
you are mounting the same path in the same path?
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140211/2fbfcc8e/attachment.html>
Alvaro Miranda Aguilera
2014-02-13 00:02:09 UTC
Permalink
sorry, i should have been more clear.

seems you are trying to mount /a into /a that won't work.

at those level, where are the logical volumes mounted?

Say you want to share

/media/lv1

into container as /media/lv1

then, the line should be:

lxc.mount.entry = /media/lv1
/var/lib/lxc/ldap_baneling/rootfs/media/lv1 none bind 0 0

do you see the difference with your line?

if you have already mounted your lv inside the container, unmount it, mount
it somewhere else, and try as I tell you, for me it works.

if you have time, I wrote this:

http://kikitux.net/lxc/lxc.html





On Wed, Feb 12, 2014 at 9:34 AM, Joshua Schaeffer
Post by Joshua Schaeffer
Based on the documentation I read, this can be the same, however all I
really care about is that the LV gets mounted to that location on the host.
On Tue, Feb 11, 2014 at 1:13 PM, Alvaro Miranda Aguilera <
On Wed, Feb 12, 2014 at 4:11 AM, Joshua Schaeffer <
Post by Joshua Schaeffer
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
you are mounting the same path in the same path?
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140213/9d4b093d/attachment.html>
Joshua Schaeffer
2014-02-13 16:15:16 UTC
Permalink
Thanks, I'll give this a try.


On Wed, Feb 12, 2014 at 5:02 PM, Alvaro Miranda Aguilera
Post by Alvaro Miranda Aguilera
sorry, i should have been more clear.
seems you are trying to mount /a into /a that won't work.
at those level, where are the logical volumes mounted?
Say you want to share
/media/lv1
into container as /media/lv1
lxc.mount.entry = /media/lv1
/var/lib/lxc/ldap_baneling/rootfs/media/lv1 none bind 0 0
do you see the difference with your line?
if you have already mounted your lv inside the container, unmount it,
mount it somewhere else, and try as I tell you, for me it works.
http://kikitux.net/lxc/lxc.html
On Wed, Feb 12, 2014 at 9:34 AM, Joshua Schaeffer <
Post by Joshua Schaeffer
Based on the documentation I read, this can be the same, however all I
really care about is that the LV gets mounted to that location on the host.
On Tue, Feb 11, 2014 at 1:13 PM, Alvaro Miranda Aguilera <
On Wed, Feb 12, 2014 at 4:11 AM, Joshua Schaeffer <
Post by Joshua Schaeffer
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
you are mounting the same path in the same path?
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140213/79f1ea7a/attachment.html>
Joshua Schaeffer
2014-02-17 16:20:27 UTC
Permalink
I still can't get this to work, I'm sure I'm missing something simple or
obvious. To recap, I'm trying to use an LVM logical volume for my
container's /var and /tmp partitions. Since I've been unable to get this
to work, I've just beenfocusing on getting /tmp to work. I changed my
mount points and my container's config:

Here are my logical volume onthe host:

root at reaver:~# lvdisplay /dev/vg_lxc1/lv_ldap_tmp1
--- Logical volume ---
LV Path /dev/vg_lxc1/lv_ldap_tmp1
LV Name lv_ldap_tmp1
VG Name vg_lxc1
LV UUID GDru3y-oLJB-Iv06-tjv3-wHuq-p8Fi-xBUscf
LV Write Access read/write
LV Creation host, time reaver, 2013-11-30 13:46:16 -0700
LV Status available
# open 1
LV Size 4.00 GiB
Current LE 1024
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:4

I've mounted this LV to: /mnt/lxc/ldap/tmp on the host. I then changed
my container's config file:

root at reaver:~# cat /var/lib/lxc/ldap_baneling/config |grep lxc.mount.entry
lxc.mount.entry = proc
/var/lib/lxc/ldap_baneling/rootfs/proc proc nodev,noexec,nosuid
0 0
lxc.mount.entry = devpts
/var/lib/lxc/ldap_baneling/rootfs/dev/pts devpts defaults 0 0
lxc.mount.entry = sysfs
/var/lib/lxc/ldap_baneling/rootfs/sys sysfs defaults 0 0
#lxc.mount.entry = /mnt/lxc/ldap/var
/var/lib/lxc/ldap_baneling/rootfs/var none defaults 0 0
*lxc.mount.entry = /mnt/lxc/ldap/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none defaults 0 0*

When I start my container I don't see the new mount:

root at baneling:~# df -h
Filesystem Size Used Avail Use% Mounted on
tmpfs 3.1G 0 3.1G 0% /run/shm
rootfs 10G 537M 9.5G 6% /
tmpfs 801M 16K 801M 1% /run
tmpfs 5.0M 0 5.0M 0% /run/lock

Do, I need to put something in the container's fstab?

Thanks,
Josh
Post by Alvaro Miranda Aguilera
sorry, i should have been more clear.
seems you are trying to mount /a into /a that won't work.
at those level, where are the logical volumes mounted?
Say you want to share
/media/lv1
into container as /media/lv1
lxc.mount.entry = /media/lv1
/var/lib/lxc/ldap_baneling/rootfs/media/lv1 none bind 0 0
do you see the difference with your line?
if you have already mounted your lv inside the container, unmount it,
mount it somewhere else, and try as I tell you, for me it works.
http://kikitux.net/lxc/lxc.html
On Wed, Feb 12, 2014 at 9:34 AM, Joshua Schaeffer
Based on the documentation I read, this can be the same, however
all I really care about is that the LV gets mounted to that
location on the host.
On Tue, Feb 11, 2014 at 1:13 PM, Alvaro Miranda Aguilera
On Wed, Feb 12, 2014 at 4:11 AM, Joshua Schaeffer
<jschaeffer0922 at gmail.com <mailto:jschaeffer0922 at gmail.com>>
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/var
/var/lib/lxc/ldap_baneling/rootfs/var none bind 0 0
lxc.mount.entry = /var/lib/lxc/ldap_baneling/rootfs/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none bind 0 0
you are mounting the same path in the same path?
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
<mailto:lxc-users at lists.linuxcontainers.org>
http://lists.linuxcontainers.org/listinfo/lxc-users
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
<mailto:lxc-users at lists.linuxcontainers.org>
http://lists.linuxcontainers.org/listinfo/lxc-users
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140217/dcbdf74e/attachment.html>
Serge Hallyn
2014-02-17 16:55:17 UTC
Permalink
Post by Joshua Schaeffer
I still can't get this to work, I'm sure I'm missing something
simple or obvious. To recap, I'm trying to use an LVM logical volume
for my container's /var and /tmp partitions. Since I've been unable
to get this to work, I've just beenfocusing on getting /tmp to work.
root at reaver:~# lvdisplay /dev/vg_lxc1/lv_ldap_tmp1
Could you please create a new container, 'testme1', and just add to its
/var/lib/lxc/testme1/config the entry

lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap defaults,create=dir 0 0

Then do

lxc-start -n testme1 -l trace -o testme1.out

and send testme1.out here (assuming that fails to start)?

thanks,
-serge
Joshua Schaeffer
2014-02-17 17:07:38 UTC
Permalink
Yes it failed to start:

1. lxc-create -n testme1 -t debian
2. root at reaver:~# cat /var/lib/lxc/testme1/config | grep lxc.mount.entry
lxc.mount.entry = proc proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = sysfs sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap
defaults,create=dir 0 0
3. root at reaver:~# lxc-start -n testme1 -l trace -o testme1.out
lxc-start: No such file or directory - failed to mount
'/dev/vg_lxc1/lv_ldap_tmp1' on '/usr/lib/x86_64-linux-gnu/lxc/lv_ldap'
lxc-start: failed to setup the mount entries for 'testme1'
lxc-start: failed to setup the container
lxc-start: invalid sequence number 1. expected 2
lxc-start: failed to spawn 'testme1'

Thanks,
Josh
Post by Serge Hallyn
Post by Joshua Schaeffer
I still can't get this to work, I'm sure I'm missing something
simple or obvious. To recap, I'm trying to use an LVM logical volume
for my container's /var and /tmp partitions. Since I've been unable
to get this to work, I've just beenfocusing on getting /tmp to work.
root at reaver:~# lvdisplay /dev/vg_lxc1/lv_ldap_tmp1
Could you please create a new container, 'testme1', and just add to its
/var/lib/lxc/testme1/config the entry
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap defaults,create=dir 0 0
Then do
lxc-start -n testme1 -l trace -o testme1.out
and send testme1.out here (assuming that fails to start)?
thanks,
-serge
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140217/945f62b9/attachment.html>
-------------- next part --------------
lxc-start 1392656494.694 DEBUG lxc_conf - allocated pty '/dev/pts/12' (4/5)
lxc-start 1392656494.695 DEBUG lxc_conf - allocated pty '/dev/pts/13' (6/7)
lxc-start 1392656494.695 DEBUG lxc_conf - allocated pty '/dev/pts/14' (8/9)
lxc-start 1392656494.695 DEBUG lxc_conf - allocated pty '/dev/pts/15' (10/11)
lxc-start 1392656494.695 INFO lxc_conf - tty's configured
lxc-start 1392656494.695 DEBUG lxc_console - using '/dev/tty' as console
lxc-start 1392656494.695 DEBUG lxc_start - sigchild handler set
lxc-start 1392656494.695 INFO lxc_start - 'testme1' is initialized
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/' (rootfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/sys' (sysfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/proc' (proc)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/dev' (devtmpfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/dev/pts' (devpts)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/run' (tmpfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/' (xfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/sys/fs/selinux' (selinuxfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/run/lock' (tmpfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/boot' (ext4)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/tmp' (xfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/var' (xfs)
lxc-start 1392656494.696 DEBUG lxc_cgroup - checking '/sys/fs/cgroup' (cgroup)
lxc-start 1392656494.696 INFO lxc_cgroup - [1] found cgroup mounted at '/sys/fs/cgroup',opts='rw,relatime,perf_event,blkio,net_cls,freezer,devices,memory,cpuacct,cpu,cpuset,clone_children'
lxc-start 1392656494.696 DEBUG lxc_cgroup - get_init_cgroup: found init cgroup for subsys (null) at /


lxc-start 1392656494.696 DEBUG lxc_cgroup - cgroup /sys/fs/cgroup has flags 0x2
lxc-start 1392656494.725 INFO lxc_cgroup - created cgroup '/sys/fs/cgroup//lxc/testme1'
lxc-start 1392656494.725 DEBUG lxc_cgroup - checking '/var/lib/nfs/rpc_pipefs' (rpc_pipefs)
lxc-start 1392656494.773 DEBUG lxc_start - Dropped cap_sys_boot

lxc-start 1392656494.773 INFO lxc_conf - 'testme1' hostname has been setup
lxc-start 1392656494.773 DEBUG lxc_conf - mounted '/var/lib/lxc/testme1/rootfs' on '/usr/lib/x86_64-linux-gnu/lxc'
lxc-start 1392656494.774 DEBUG lxc_conf - mounted 'proc' on '/usr/lib/x86_64-linux-gnu/lxc/proc', type 'proc'
lxc-start 1392656494.774 DEBUG lxc_conf - mounted 'sysfs' on '/usr/lib/x86_64-linux-gnu/lxc/sys', type 'sysfs'
lxc-start 1392656494.774 ERROR lxc_conf - No such file or directory - failed to mount '/dev/vg_lxc1/lv_ldap_tmp1' on '/usr/lib/x86_64-linux-gnu/lxc/lv_ldap'
lxc-start 1392656494.774 ERROR lxc_conf - failed to setup the mount entries for 'testme1'
lxc-start 1392656494.774 ERROR lxc_start - failed to setup the container
lxc-start 1392656494.774 ERROR lxc_sync - invalid sequence number 1. expected 2
lxc-start 1392656494.774 ERROR lxc_start - failed to spawn 'testme1'
lxc-start 1392656494.774 DEBUG lxc_cgroup - get_init_cgroup: found init cgroup for subsys (null) at /


lxc-start 1392656494.774 DEBUG lxc_cgroup - destroying /sys/fs/cgroup//lxc/testme1

lxc-start 1392656494.861 DEBUG lxc_cgroup - '/sys/fs/cgroup//lxc/testme1' unlinked
Serge Hallyn
2014-02-18 05:24:07 UTC
Permalink
Post by Joshua Schaeffer
1. lxc-create -n testme1 -t debian
2. root at reaver:~# cat /var/lib/lxc/testme1/config | grep lxc.mount.entry
lxc.mount.entry = proc proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = sysfs sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap
defaults,create=dir 0 0
Wait a sec. I mistyped. There's no fstype in there :)

That should be

lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap ext4 defaults,create=dir 0 0

or whatever fstype it is in place of ext4. Sorry.
Post by Joshua Schaeffer
3. root at reaver:~# lxc-start -n testme1 -l trace -o testme1.out
lxc-start: No such file or directory - failed to mount
'/dev/vg_lxc1/lv_ldap_tmp1' on
'/usr/lib/x86_64-linux-gnu/lxc/lv_ldap'
lxc-start: failed to setup the mount entries for 'testme1'
lxc-start: failed to setup the container
lxc-start: invalid sequence number 1. expected 2
lxc-start: failed to spawn 'testme1'
Serge Hallyn
2014-02-18 05:30:57 UTC
Permalink
Post by Serge Hallyn
Post by Joshua Schaeffer
1. lxc-create -n testme1 -t debian
2. root at reaver:~# cat /var/lib/lxc/testme1/config | grep lxc.mount.entry
lxc.mount.entry = proc proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = sysfs sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap
defaults,create=dir 0 0
Wait a sec. I mistyped. There's no fstype in there :)
That should be
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap ext4 defaults,create=dir 0 0
or whatever fstype it is in place of ext4. Sorry.
Actually I see that will fail too, because 'create=dir' has a
slight problem. We need to remove create=dir from the mount
options which we pass along to the mount syscall.

so for the sake of testing just do

lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 mnt ext4 defaults 0 0

-serge
Joshua Schaeffer
2014-02-18 16:16:11 UTC
Permalink
Okay thanks for the pointers. I'll be able to try this out tonight and
report back.
Post by Joshua Schaeffer
Post by Serge Hallyn
Post by Joshua Schaeffer
1. lxc-create -n testme1 -t debian
2. root at reaver:~# cat /var/lib/lxc/testme1/config | grep
lxc.mount.entry
Post by Serge Hallyn
Post by Joshua Schaeffer
lxc.mount.entry = proc proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = sysfs sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap
defaults,create=dir 0 0
Wait a sec. I mistyped. There's no fstype in there :)
That should be
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap ext4
defaults,create=dir 0 0
Post by Serge Hallyn
or whatever fstype it is in place of ext4. Sorry.
Actually I see that will fail too, because 'create=dir' has a
slight problem. We need to remove create=dir from the mount
options which we pass along to the mount syscall.
so for the sake of testing just do
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 mnt ext4 defaults 0 0
-serge
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140218/f16346d1/attachment.html>
Joshua Schaeffer
2014-02-19 02:04:33 UTC
Permalink
Okay, I think it's all figured out. I made the change you suggested to
my test container and it mounted my LV to /mnt successfully in the
container. I then took another look at my other container and it looks
like I had a bad lxc.mount.entry entry (knew it had to be something
simple). I had:

root at reaver:~# cat /var/lib/lxc/ldap_baneling/config | grep lxc.mount.entry
lxc.mount.entry = proc
/var/lib/lxc/dns_defiler/rootfs/proc proc
nodev,noexec,nosuid 0 0
lxc.mount.entry = devpts
/var/lib/lxc/dns_defiler/rootfs/dev/pts devptsdefaults00
lxc.mount.entry = sysfs
/var/lib/lxc/dns_defiler/rootfs/syssysfsdefaults0 0
lxc.mount.entry = /mnt/lxc/ldap/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmpnonedefaults 00

I'm not sure where I pulled "none" from for the fstype, but when you
mentioned that you didn't specify that in your config line either it
made me check mine (mine is xfs not ext4 but the point is the same). I
also changed thefile system line to "/dev/vg_lxc1/lv_ldap_tmp1" and
changed the mount point to just "tmp". Now I have:

lxc.mount.entry = proc
proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = devpts
dev/pts devpts defaults 0 0
lxc.mount.entry = sysfs
sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_var1
var xfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1
tmp xfs defaults 0 0

Everything seems to be working correctly now. Thanks for all your help,
Josh
Post by Serge Hallyn
Post by Serge Hallyn
Post by Joshua Schaeffer
1. lxc-create -n testme1 -t debian
2. root at reaver:~# cat /var/lib/lxc/testme1/config | grep lxc.mount.entry
lxc.mount.entry = proc proc proc nodev,noexec,nosuid 0 0
lxc.mount.entry = sysfs sys sysfs defaults 0 0
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap
defaults,create=dir 0 0
Wait a sec. I mistyped. There's no fstype in there :)
That should be
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 lv_ldap ext4 defaults,create=dir 0 0
or whatever fstype it is in place of ext4. Sorry.
Actually I see that will fail too, because 'create=dir' has a
slight problem. We need to remove create=dir from the mount
options which we pass along to the mount syscall.
so for the sake of testing just do
lxc.mount.entry = /dev/vg_lxc1/lv_ldap_tmp1 mnt ext4 defaults 0 0
-serge
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140218/4e86ab79/attachment.html>
Jon Brinkmann
2014-02-18 07:34:39 UTC
Permalink
I'm trying to build containers in Scientific Linux, a clone of RHEL.
When building from the git source for v1.0.0_rc3, configure has no
problems identifying the distribution:

Environment:
- compiler: gcc
- distribution: redhat
- init script type(s): sysvinit
- rpath: no
- GnuTLS: no
- Bash integration: yes

However, when I attempt to create a container, I get:

# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat

lxc_container: bad template: redhat

lxc_container: Error creating container system1

and looking in /usr/local/share/lxc/templates:

# ls /usr/local/share/lxc/templates/
lxc-alpine lxc-centos lxc-fedora lxc-oracle lxc-ubuntu-cloud
lxc-altlinux lxc-cirros lxc-gentoo lxc-plamo
lxc-archlinux lxc-debian lxc-openmandriva lxc-sshd
lxc-busybox lxc-download lxc-opensuse lxc-ubuntu

There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?

Jon
Tamas Papp
2014-02-18 08:05:45 UTC
Permalink
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
When building from the git source for v1.0.0_rc3, configure has no
- compiler: gcc
- distribution: redhat
- init script type(s): sysvinit
- rpath: no
- GnuTLS: no
- Bash integration: yes
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
lxc_container: bad template: redhat
lxc_container: Error creating container system1
# ls /usr/local/share/lxc/templates/
lxc-alpine lxc-centos lxc-fedora lxc-oracle lxc-ubuntu-cloud
lxc-altlinux lxc-cirros lxc-gentoo lxc-plamo
lxc-archlinux lxc-debian lxc-openmandriva lxc-sshd
lxc-busybox lxc-download lxc-opensuse lxc-ubuntu
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
There is no official template for redhat.... I guess because there is no
public redhat repository available.

tamas
Fajar A. Nugraha
2014-02-18 08:50:33 UTC
Permalink
Post by Tamas Papp
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
My question would be "what would you need redhat template for"?
Post by Tamas Papp
There is no official template for redhat.... I guess because there is no
public redhat repository available.
Correct. lxc uses public repositories and some kind of bootstrapping
(yum, debootstrap, etc) to setup the guest container. Redhat does not
provide access to such container.

Submitting a scientific linux (SL) template would be a good idea, if
you use it often and have something speicifc that only SL can provide.
There's already oracle linux and centos template, If you just need a
working redhat-clone.

An alternative is to clone a working RHEL installation, and customize
it for lxc (using the content of lxc-centos template script as
example)
--
Fajar
Jon Brinkmann
2014-02-19 00:01:24 UTC
Permalink
Thanks for your help.

I'm hesitant to use Centos repos, mainly due to its "acquisition" by
RH as a test bed for RHEL. SL is and will remain more conservative,
cloning RHEL. I'm not familiar with Oracle's policies, but I found
differences between their lxc instructions and Centos'.

I'll work on a SL template, which uses the SL repos, and submit it to
lxc-devel.

Thanks again,

Jon
Post by Fajar A. Nugraha
Post by Tamas Papp
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
My question would be "what would you need redhat template for"?
Post by Tamas Papp
There is no official template for redhat.... I guess because there is no
public redhat repository available.
Correct. lxc uses public repositories and some kind of bootstrapping
(yum, debootstrap, etc) to setup the guest container. Redhat does not
provide access to such container.
Submitting a scientific linux (SL) template would be a good idea, if
you use it often and have something speicifc that only SL can provide.
There's already oracle linux and centos template, If you just need a
working redhat-clone.
An alternative is to clone a working RHEL installation, and customize
it for lxc (using the content of lxc-centos template script as
example)
--
Fajar
Daniel Veillard
2014-02-20 08:25:55 UTC
Permalink
Post by Jon Brinkmann
Thanks for your help.
I'm hesitant to use Centos repos, mainly due to its "acquisition" by
RH as a test bed for RHEL. SL is and will remain more conservative,
cloning RHEL.
I guess I have to react to that, first historically SL did diverge
more than than CentOS (in order to better serve their audience, a
deliberate choice), also it has been made clear that the way packages
were built into CentOS is not changing, just the guys doing it
work for Red Hat. There is work being done to provide new variants of
CentOS serving more specific space and diverging from the base platform,
but if you want to stick to the RHEL clone, just use the base platform.
http://www.centos.org/variants/
http://wiki.centos.org/SpecialInterestGroup

I know this list may not like Red Hat a lot, and not trust me in
particular since I'm the original author of libvirt, but I still think
I need to correct the above misconception.
Post by Jon Brinkmann
I'm not familiar with Oracle's policies, but I found
differences between their lxc instructions and Centos'.
I'll work on a SL template, which uses the SL repos, and submit it to
lxc-devel.
It would actually be interesting to find out what changes need to be
done for the SL template compared to the CentOS one, if any are
required, my guess is that you will end up with the same except for
the resources URLs, release package name, and possibly the path for the
cache and release file.

Daniel
Post by Jon Brinkmann
Thanks again,
Jon
Post by Fajar A. Nugraha
Post by Tamas Papp
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
My question would be "what would you need redhat template for"?
Post by Tamas Papp
There is no official template for redhat.... I guess because there is no
public redhat repository available.
Correct. lxc uses public repositories and some kind of bootstrapping
(yum, debootstrap, etc) to setup the guest container. Redhat does not
provide access to such container.
Submitting a scientific linux (SL) template would be a good idea, if
you use it often and have something speicifc that only SL can provide.
There's already oracle linux and centos template, If you just need a
working redhat-clone.
An alternative is to clone a working RHEL installation, and customize
it for lxc (using the content of lxc-centos template script as
example)
--
Fajar
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
--
Daniel Veillard | Open Source and Standards, Red Hat
veillard at redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/
http://veillard.com/ | virtualization library http://libvirt.org/
Fajar A. Nugraha
2014-02-20 12:04:53 UTC
Permalink
On Wed, Feb 19, 2014 at 12:01:24AM +0000, Jon Brinkmann wrote:> I'm not
familiar with Oracle's policies, but I found
Post by Jon Brinkmann
differences between their lxc instructions and Centos'.
I'll work on a SL template, which uses the SL repos, and submit it to
lxc-devel.
It would actually be interesting to find out what changes need to be
done for the SL template compared to the CentOS one, if any are
required, my guess is that you will end up with the same except for
the resources URLs, release package name, and possibly the path for the
cache and release file.
IIRC Oracle template make it to the official repo befor the official centos
one. Some of the differences came from different approach in error handling
(e.g. difference in RPM library version between host and container).

Personally I suggest using the centos one as base for SL template, since it
came later, and have had some review and suggestions from the community.
--
Fajar
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140220/7a49c8cb/attachment.html>
Michael H. Warfield
2014-03-08 03:32:36 UTC
Permalink
Sorry for joining this discussion late but I've been out of pocket since
2/16 and this is the first I've had a chance to comment...
Post by Jon Brinkmann
Post by Jon Brinkmann
Thanks for your help.
I'm hesitant to use Centos repos, mainly due to its "acquisition" by
RH as a test bed for RHEL.
Time out... I don't recall hearing that it was for a "test bed for
RHEL" at all. That's what they support Fedora for. The statement was
made that they were supporting CentOS as a community supported version
but very much in parallel with RHEL as it has been. RHEL releases first
and CentOS makes the OpenSource pieces available after. Doesn't sound
like a "test bed" to me.

This is almost getting back to their roots of the pre Fedora RHEL split.
I have no heartburn over that and even welcome it. CentOS has suffered
from an on-again, off-again commitment to it's ongoing development and
stability (there was some serious heartburn for several months that
CentOS 6 would EVER come out). Having a corporate backer is a good
thing in that it removes some doubt at to the long term support and
longevity.
Post by Jon Brinkmann
SL is and will remain more conservative,
Post by Jon Brinkmann
cloning RHEL.
Which is cool and which is why and SL template would be a good thing
too.
Post by Jon Brinkmann
I guess I have to react to that, first historically SL did diverge
more than than CentOS (in order to better serve their audience, a
deliberate choice), also it has been made clear that the way packages
were built into CentOS is not changing, just the guys doing it
work for Red Hat. There is work being done to provide new variants of
CentOS serving more specific space and diverging from the base platform,
but if you want to stick to the RHEL clone, just use the base platform.
http://www.centos.org/variants/
http://wiki.centos.org/SpecialInterestGroup
I know this list may not like Red Hat a lot, and not trust me in
particular since I'm the original author of libvirt, but I still think
I need to correct the above misconception.
I've become sort of the defacto maintainer for the Fedora and CentOS
templates. The list is not anti-redhat, quite the contrary. There are
a lot of contributions and direction from the Ubuntu crowd but I've
found that Serge and Stephane are both very happy with having rpm based
distros such as Oracle, CentOS, Suse, and Fedora (as long as it's
someone else doing the supporting). I have not encountered any
antipathy from any of the crowd here against that. I've gotten stuck
with leading the charge into the systemd fire, so I guess it's an
attitude that they would rather have it be me than them in that
flamefest.
Post by Jon Brinkmann
Post by Jon Brinkmann
I'm not familiar with Oracle's policies, but I found
differences between their lxc instructions and Centos'.
I'll work on a SL template, which uses the SL repos, and submit it to
lxc-devel.
It would actually be interesting to find out what changes need to be
done for the SL template compared to the CentOS one, if any are
required, my guess is that you will end up with the same except for
the resources URLs, release package name, and possibly the path for the
cache and release file.
I've looked at it. It's, as you say, largely going to be centered
around the initial bootstrap of the cache and, yeah, it's going to be
mostly the url's for the initial packages and repos. IIRC from my last
look, I had some heartburn over retro releases (i.e. building an SL 5.0
container) because some of the repo directories were restructured
between 5 and 6 that was going to be problematical. If it's just
structured for the latest release, like OpenSuse is, it's not that
difficult.

I think the CentOS template is the best place to start there (no
aspersion cast upon Dwight and the Oracle template - I've been grabbing
great ideas from him and he's done great work with that template). I
mentioned SL some time ago and I may take a look at it myself if someone
else doesn't beat me to it.

Several people have commented that these 4 templates (Oracle, CentOS,
SL, and Fedora) should be so close they could almost be consolidated but
I don't see that happening any time soon and I don't see a support
mechanism (other than maybe symlinking them, which I don't like).
Another fish to fry on another day...
Post by Jon Brinkmann
Daniel
Mike
Post by Jon Brinkmann
Post by Jon Brinkmann
Thanks again,
Jon
Post by Fajar A. Nugraha
Post by Tamas Papp
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
My question would be "what would you need redhat template for"?
Post by Tamas Papp
There is no official template for redhat.... I guess because there is no
public redhat repository available.
Correct. lxc uses public repositories and some kind of bootstrapping
(yum, debootstrap, etc) to setup the guest container. Redhat does not
provide access to such container.
Submitting a scientific linux (SL) template would be a good idea, if
you use it often and have something speicifc that only SL can provide.
There's already oracle linux and centos template, If you just need a
working redhat-clone.
An alternative is to clone a working RHEL installation, and customize
it for lxc (using the content of lxc-centos template script as
example)
--
Fajar
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
--
Michael H. Warfield (AI4NB) | (770) 978-7061 | mhw at WittsEnd.com
/\/\|=mhw=|\/\/ | (678) 463-0932 | http://www.wittsend.com/mhw/
NIC whois: MHW9 | An optimist believes we live in the best of all
PGP Key: 0x674627FF | possible worlds. A pessimist is sure of it!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 482 bytes
Desc: This is a digitally signed message part
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140307/e63d587c/attachment.pgp>
Michael H. Warfield
2014-03-08 03:58:36 UTC
Permalink
Post by Michael H. Warfield
Sorry for joining this discussion late but I've been out of pocket since
2/16 and this is the first I've had a chance to comment...
Post by Jon Brinkmann
Post by Jon Brinkmann
Thanks for your help.
I'm hesitant to use Centos repos, mainly due to its "acquisition" by
RH as a test bed for RHEL.
Time out... I don't recall hearing that it was for a "test bed for
RHEL" at all. That's what they support Fedora for. The statement was
made that they were supporting CentOS as a community supported version
but very much in parallel with RHEL as it has been. RHEL releases first
and CentOS makes the OpenSource pieces available after. Doesn't sound
like a "test bed" to me.
This is almost getting back to their roots of the pre Fedora RHEL split.
I have no heartburn over that and even welcome it. CentOS has suffered
from an on-again, off-again commitment to it's ongoing development and
stability (there was some serious heartburn for several months that
CentOS 6 would EVER come out). Having a corporate backer is a good
thing in that it removes some doubt at to the long term support and
longevity.
Post by Jon Brinkmann
SL is and will remain more conservative,
Post by Jon Brinkmann
cloning RHEL.
Which is cool and which is why and SL template would be a good thing
too.
Post by Jon Brinkmann
I guess I have to react to that, first historically SL did diverge
more than than CentOS (in order to better serve their audience, a
deliberate choice), also it has been made clear that the way packages
were built into CentOS is not changing, just the guys doing it
work for Red Hat. There is work being done to provide new variants of
CentOS serving more specific space and diverging from the base platform,
but if you want to stick to the RHEL clone, just use the base platform.
http://www.centos.org/variants/
http://wiki.centos.org/SpecialInterestGroup
I know this list may not like Red Hat a lot, and not trust me in
particular since I'm the original author of libvirt, but I still think
I need to correct the above misconception.
I've become sort of the defacto maintainer for the Fedora and CentOS
templates. The list is not anti-redhat, quite the contrary. There are
a lot of contributions and direction from the Ubuntu crowd but I've
found that Serge and Stephane are both very happy with having rpm based
distros such as Oracle, CentOS, Suse, and Fedora (as long as it's
someone else doing the supporting). I have not encountered any
antipathy from any of the crowd here against that. I've gotten stuck
with leading the charge into the systemd fire, so I guess it's an
attitude that they would rather have it be me than them in that
flamefest.
Post by Jon Brinkmann
Post by Jon Brinkmann
I'm not familiar with Oracle's policies, but I found
differences between their lxc instructions and Centos'.
I'll work on a SL template, which uses the SL repos, and submit it to
lxc-devel.
It would actually be interesting to find out what changes need to be
done for the SL template compared to the CentOS one, if any are
required, my guess is that you will end up with the same except for
the resources URLs, release package name, and possibly the path for the
cache and release file.
I've looked at it. It's, as you say, largely going to be centered
around the initial bootstrap of the cache and, yeah, it's going to be
mostly the url's for the initial packages and repos. IIRC from my last
look, I had some heartburn over retro releases (i.e. building an SL 5.0
container) because some of the repo directories were restructured
between 5 and 6 that was going to be problematical. If it's just
structured for the latest release, like OpenSuse is, it's not that
difficult.
Oh, yeah... Just peeked at it again. Now I remember where I was
getting that migraine from.

Up there at http://ftp.scientificlinux.org/linux/scientific/

Sooo... For the 6.x release, you've got these directories (at SL):

6, 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, and 6rolling.

What the hell is the difference between 6 and 6rolling? CentOS, 6 is
the rolling release and it's what that template uses for that rev.

Now for 5.x you've got this:

5rolling, 5x, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, and 510.

Sigh... Different directory paradigm. No 5 at all. Presumably 5x.
But no "." in any of them. So the logic to support both the 5.x and 6.x
versions is going to be butt ugly.

Supporting JUST the rolling releases, I could pull 6rolling for 6 and
5rolling for 5. Getting minor releases working will be a pain and
probably not needed anyways.

Except... The directory structure under 6rolling is not the same as
5rolling.

The packages in 5rolling are here:

5rolling/x86_64/SL/

The packages in 6rolling are here:

6rolling/x86_64/os/Packages/

Just one more thing that needs to be handled.
Post by Michael H. Warfield
I think the CentOS template is the best place to start there (no
aspersion cast upon Dwight and the Oracle template - I've been grabbing
great ideas from him and he's done great work with that template). I
mentioned SL some time ago and I may take a look at it myself if someone
else doesn't beat me to it.
Several people have commented that these 4 templates (Oracle, CentOS,
SL, and Fedora) should be so close they could almost be consolidated but
I don't see that happening any time soon and I don't see a support
mechanism (other than maybe symlinking them, which I don't like).
Another fish to fry on another day...
Post by Jon Brinkmann
Daniel
Mike
Post by Michael H. Warfield
Post by Jon Brinkmann
Post by Jon Brinkmann
Thanks again,
Jon
Post by Fajar A. Nugraha
Post by Tamas Papp
Post by Jon Brinkmann
I'm trying to build containers in Scientific Linux, a clone of RHEL.
# lxc-create -t redhat -n system1 -P /systems/
lxc_container: No such file or directory - bad template: redhat
There's no sign of a redhat template. How do I proceed? Should I create
a template for it from lxc-centos and submit it to lxc-devel?
My question would be "what would you need redhat template for"?
Post by Tamas Papp
There is no official template for redhat.... I guess because there is no
public redhat repository available.
Correct. lxc uses public repositories and some kind of bootstrapping
(yum, debootstrap, etc) to setup the guest container. Redhat does not
provide access to such container.
Submitting a scientific linux (SL) template would be a good idea, if
you use it often and have something speicifc that only SL can provide.
There's already oracle linux and centos template, If you just need a
working redhat-clone.
An alternative is to clone a working RHEL installation, and customize
it for lxc (using the content of lxc-centos template script as
example)
--
Fajar
_______________________________________________
lxc-users mailing list
lxc-users at lists.linuxcontainers.org
http://lists.linuxcontainers.org/listinfo/lxc-users
--
Michael H. Warfield (AI4NB) | (770) 978-7061 | mhw at WittsEnd.com
/\/\|=mhw=|\/\/ | (678) 463-0932 | http://www.wittsend.com/mhw/
NIC whois: MHW9 | An optimist believes we live in the best of all
PGP Key: 0x674627FF | possible worlds. A pessimist is sure of it!

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 482 bytes
Desc: This is a digitally signed message part
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20140307/f625606c/attachment-0001.pgp>
Fajar A. Nugraha
2014-02-18 05:28:16 UTC
Permalink
On Mon, Feb 17, 2014 at 11:20 PM, Joshua Schaeffer
Post by Joshua Schaeffer
I still can't get this to work, I'm sure I'm missing something simple or
obvious.
lxc.mount.entry = /mnt/lxc/ldap/tmp
/var/lib/lxc/ldap_baneling/rootfs/tmp none defaults 0 0
root at baneling:~# df -h
Filesystem Size Used Avail Use% Mounted on
tmpfs 3.1G 0 3.1G 0% /run/shm
rootfs 10G 537M 9.5G 6% /
tmpfs 801M 16K 801M 1% /run
tmpfs 5.0M 0 5.0M 0% /run/lock
What does "cat /proc/mounts" in the containter show?
--
Fajar
Continue reading on narkive:
Loading...