So I want to mount all shares from each unRAID in every other unRAID. Lets keep this simple and consider only 2 unRAID SERVER1 and SERVER2

 

Mounting specific folders between each server is easy. Let keep it one way this now for ease:

 

root@SERVER1:/# /usr/bin/mkdir "/mnt/server2/tv"

root@SERVER1:/# /usr/bin/chown -R nobody:users /mnt/server2/tv

root@SERVER1:/# mount -t cifs -o username=anonymous,password=anonymous,uid=99,forceuid,gid=100,forcegid,file_mode=0666,dir_mode=0777,rw //server2/TV /mnt/server2/tv

uid=99,forceuid,gid=100,forcegid,file_mode=0666,dir_mode=0777,rw

 

 

This just works.

 

However a similar IPC$ mount this fails miserably:

 

root@SERVER1:/# mount -t cifs -o username=anonymous,password=anonymous,uid=99,gid=100 //server2/IPC$ /mnt/server2

 

We can tell the IPC$ share exists:

 

root@SERVER1:/# smbclient -L //server2/ -U%
WARNING: The "null passwords" option is deprecated
Domain=[iNTRANET] OS=[unix] Server=[samba 3.6.10]

        Sharename       Type      Comment
        ---------       ----      -------
   ...
        IPC$            IPC       IPC Service (Content)
Domain=[iNTRANET] OS=[unix] Server=[samba 3.6.10]
   ...

 

Some permissions is stopping this... does anyone have a step for a clue?

 

Update changed:

 

uid=99,gid=100

 

uid=99,forceuid,gid=100,forcegid,file_mode=0666,dir_mode=0777,rw

 

This more explicitly matches unRAID user and permissions defaults.

 

Still no luck with IPC$. Anyone?

It seems that you simply cannot mount the IPC$ share under Linux.

 

I find this hard to believe since it is trivial in windows. Anyone got something else for me to look into as I am now at a loss.

 

In simple terms I want to mount the whole server under one directory from another server. preferably using cifs but I could live with a hack to make it work at this point.

Gut feeling I have is that LimeTech has made what you want to do very difficult or impossible.  I would suggest you look at the licensing structure and costs to see why.

Gut feeling I have is that LimeTech has made what you want to do very difficult or impossible.  I would suggest you look at the licensing structure and costs to see why.

 

I ponder and ponder and have no clue what this means? What has licensing got to do with mount -t cifs?

Gut feeling I have is that LimeTech has made what you want to do very difficult or impossible.  I would suggest you look at the licensing structure and costs to see why.

 

I ponder and ponder and have no clue what this means? What has licensing got to do with mount -t cifs?

 

Duh!  You could link any number of three-disk servers together  to construct a massive array and thus avoid paying LimeTech for all of the work that had gone and continues to go into maintaining and improving unRAID.

Gut feeling I have is that LimeTech has made what you want to do very difficult or impossible.  I would suggest you look at the licensing structure and costs to see why.

 

I ponder and ponder and have no clue what this means? What has licensing got to do with mount -t cifs?

 

Duh!  You could link any number of three-disk servers together  to construct a massive array and thus avoid paying LimeTech for all of the work that had gone and continues to go into maintaining and improving unRAID.

That would not work.    Any disks that are mounted are effectively outside the array as far as unRAID specific features are concerned.

 

As far as I know the issue with ipc$ is a generic Linux issue - and nothing to specific to unRAID.

Duh!  You could link any number of three-disk servers together  to construct a massive array and thus avoid paying LimeTech

So, your brilliant idea is that you'll save money by making something with 20 data disks and 10 parity disks?  By all means, go for it! 

 

 

OK I now understand what your licensing theory is now. IMO it has nothing to do with this issue so lets get back on topic.

 

Anyone had any success mounting a whole unRAID server with mount -t cifs?

In simple terms I want to mount the whole server under one directory from another server. preferably using cifs but I could live with a hack to make it work at this point.

 

/boot/config/smb-extra.conf:

 

[mnt]
comment = Mount Dir
path = /mnt/
valid users = gfjardim
write list = gfjardim
browseable = No

 

This is what I use to access all files/disks from another computer: just create a new share via smb-extra.conf file.

 

My Ubuntu fstab:

 

//192.168.0.100/mnt /media/unRAID cifs  rw,nounix,iocharset=utf8,_netdev,file_mode=0777,dir_mode=0777,credentials=/home/guilherme/.unraid_credentials 0 0

I use automount with NFS.

 

I can't remember if unRAID has the automount drivers loaded in the kernel however it does make it worthwhile to request it if people find it useful for cross mounting server to server.

 

Anyway using automount and NFS you can set up tables on each client hosts to point to other servers. Simply by cd'ing to the path you can automount and autoumount the remote directories.

 

While it doesn't  answer the original question, I figured I would post some information which may be useful for other ideas.

 

on my slackware development host I have

root@slacky:~# cat /etc/auto.master 
#
# Sample auto.master file
# This is an automounter map and it has the following format
# key [ -mount-options-separated-by-comma ] location
# For details of the format look at autofs(5).
#
/misc   /etc/auto.misc
#
# NOTE: mounts done from a hosts map will be mounted with the
#       "nosuid" and "nodev" options unless the "suid" and "dev"
#       options are explicitly given.
#
/net    -hosts
#
# Include central master map if it can be found using
# nsswitch sources.
#
# Note that if there are entries for /net or /misc (as
# above) in the included master map any keys that are the
# same will not be seen as the first read key seen takes
# precedence.
#
+auto.master
#
/unraid /etc/auto.unraid

 

root@slacky:~# cat /etc/auto.unraid 
*       -rw,rsize=8192,wsize=8192,posix unraid:/mnt/&

 

I believe I had to customize the /etc/exports file for no_root_squash

root@unRAID:~# cat /etc/exports
# See exports(5) for a description.
# This file contains a list of all directories exported to other computers.
# It is used by rpc.nfsd and rpc.mountd.


"/mnt/disk1" -async,no_subtree_check,fsid=11 *(insecure,anongid=100,anonuid=99,rw,no_root_squash)
"/mnt/disk2" -async,no_subtree_check,fsid=12 *(insecure,anongid=100,anonuid=99,rw,no_root_squash)
"/mnt/disk3" -async,no_subtree_check,fsid=13 *(insecure,anongid=100,anonuid=99,rw,no_root_squash)

 

 

root@slacky:~# df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/root             15277288   4183848  11093440  28% /
tmpfs                   512656         0    512656   0% /dev/shm


root@slacky:~# cd /unraid/disk1
root@slacky:/unraid/disk1# ls -l
total 44766
drwxrwx---  11 rcotrone users      704 2013-10-18 21:02 BACKUPS/


root@slacky:/# df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/root             15277288   4183848  11093440  28% /
tmpfs                   512656         0    512656   0% /dev/shm
unraid:/mnt/disk1    2930177104 2415404896 514772208  83% /unraid/disk1

 

In any case, this makes sharing disks via a tree easy.

 

cd /servername/diskname.

 

I've been doing it this way for years in all my networks with great success.

In simple terms I want to mount the whole server under one directory from another server. preferably using cifs but I could live with a hack to make it work at this point.

 

/boot/config/smb-extra.conf:

 

[mnt]
comment = Mount Dir
path = /mnt/
valid users = gfjardim
write list = gfjardim
browseable = No

 

This is what I use to access all files/disks from another computer: just create a new share via smb-extra.conf file.

 

My Ubuntu fstab:

 

//192.168.0.100/mnt /media/unRAID cifs  rw,nounix,iocharset=utf8,_netdev,file_mode=0777,dir_mode=0777,credentials=/home/guilherme/.unraid_credentials 0 0

 

I thought about this but my main worry was unforseen consequences.

 

Can we think of any?

 

I use automount with NFS. ....

 

This is really rather slick.

 

I think combining avahi with cross mounting could be quite a bit of bang for your buck as a future feature enhancement.