Unable to map more then 9 mount points.

[SpaceCowBoy]

Hello. I am unable to map more then 9 mp. I get the error
vm 101 - unable to parse value of 'mp10' - unknown setting 'mp10'

do I need to do something different then this?
mp10: /somepath,mp=/somepath

 

[dietmar]

We only allow 10 mount points.

 

[morph027]

Ran into the same for a special container and solved it with multiple

lxc.mount.entry = /data/nas data none bind,create=dir 0 0

 

[LnxBil]

You can just patch PVE to get this done once (need to have an automatic patch mechanism for that):

--- /tmp/Config.pm    2016-07-17 18:59:28.884082554 +0200
+++ /usr/share/perl5/PVE/LXC/Config.pm    2016-07-17 18:59:36.811969832 +0200
@@ -16,7 +16,7 @@
my $lockdir = "/run/lock/lxc";
mkdir $lockdir;
mkdir "/etc/pve/nodes/$nodename/lxc";
-my $MAX_MOUNT_POINTS = 10;
+my $MAX_MOUNT_POINTS = 32;
my $MAX_UNUSED_DISKS = $MAX_MOUNT_POINTS;

# BEGIN implemented abstract methods from PVE::AbstractConfig

 

[fabian]

out of interest - what are your use cases for such setups? are all of those media servers in home labs (which seems to be the usual use case that comes up in the forum )?

 

[morph027]

You're right, this setup is very uncommon. I'd rather think about the setups*.

In my special case, there is a small business client i'm taking care of. There are lots of network shares to provide (each user, roaming profiles, scanner and so on) which i divided into atomic ZFS datasets (you know, better snapshotting possibilities, backup, ZFS rocks, ...). These are being mountpointed into a LXC container running the samba fileserver to share the stuff. I'm not quite happy with this**, but for now it works. In all other setups, i'm just having 1, max. 2 MP's.

* Lately, i came up with the idea to spread the datasets over multiple servers, setup multiple samba fileservers and then use DFS to "proxy" these for the clients. Did not start testing so far.

** Performance impacts? Should not occur, but you never know, anything could happen. Also single point of failure.

 

[LnxBil]

I'm running a similar as @morph027 with atomic ZFS shares (e.g. one zfs dataset for each year of images), so you "outgrow" the limit of 10 really fast.