There's a per-share config file for setting up security, which is also needed for "write-able user shares".  Both features are being tested now.

Any chance to have NTFS native partition support? I know this is problematic on a number of levels but it would be so much easier in some circumstances to troubleshoot or even update drives locally and then sync.

Concatenated parity drives would be cool, but it seems there is a decent work around.  I'll start my UnRaid with four or five 500GB WD drives and add more as needed.  At some point, I'll start using 750GB drives.  I'll add the first 750, and delete the old parity drive, and make the new 750 the new parity drive, then add the old 500gb parity drive as a new data drive.  Doesn't seem much work compared to the potential concatenated process of adding the 750 as data, and a 250 for additional concatenated parity.

 

That solved, my vote is for S3 suspend support!

My vote goes for NFS support. Why? Because most hardware media player boxes (e.g. Tvix HD-5000,  PixelMagic MB200 etc) struggle to play high bit rate video files over Samba shares, while they're doing better with NFS shares.

Good points about the writable user shares being a "nice have" only.

 

To me, must have would be:

 

1.  Security, for home users would be nice, but this is stopping me from recommending an unRaid server at my work, I am in charge of IT but it would have to be secure, at the user level.

2.  S3 to save power and hence dollars, and the enviro.  Plus the server will last longer.

3.  Shorter drive spin down times, preferably drive specific.

My vote goes for NFS support. Why? Because most hardware media player boxes (e.g. Tvix HD-5000,  PixelMagic MB200 etc) struggle to play high bit rate video files over Samba shares, while they're doing better with NFS shares.

 

I've been eager to see that as well, along with UPnP- both have advantages and disadvantages with the latest crop of players.

Tom has already committed that the next build will have NFS built into it.  However, he stated that Integration with the UI would be in a subsequent release.

 

Another reason for NFS is to provide a second means of sharing the drives incase of Samba issues (like some of us are seeing now with Vista .

Good points about the writable user shares being a "nice have" only.

 

To me, must have would be:

 

1.  Security, for home users would be nice, but this is stopping me from recommending an unRaid server at my work, I am in charge of IT but it would have to be secure, at the user level.

2.  S3 to save power and hence dollars, and the enviro.  Plus the server will last longer.

3.  Shorter drive spin down times, preferably drive specific.

 

Hello, I am a newbie. Liked unRaid very much based on what I have read so far. For me point 2 & 3  and WOL  are the most important factors before I can commit my resources to unRaid. I will try out the free version but will not buy the full blow version as I am a firm believer in energy/environmental conservation. Will watch the development closely.

 

Regards and Good Luck.

Shaurya

Silly request #1

 

An option in "Settings" to view disk temperature in degrees Fahrenheit for those of us who are Celsius challenged...

I really need NFS (UDP & TCP) support.

 

+1

 

I have several TVIX video streaming boxes that are unable to stream HD files with SAMBA but work great with NFS.

 

By the way, what are other people using to stream HD content other then HTPC's?

 

Another Tvix user here who really would like NFS to be at the top of the list

 

Currently trying all manner of Samba/windows/mac solutions to allow my movies to be piped down to the TVIX.

 

I just hate the idea of having to have windows/mac running to run some NFS software to allow the streaming to work properly.. so all in on the Unraid would be superb.

- user defined spin down time in minutes/hours per drive or the option for all drives (not just 1,2,3,4,5 hrs in a dropdown)

- auto refresh of main.htm: would help with parity check or rebuild to get an updated status instead of clicking refresh. this could be a user defined option too.

- ability to modify disk names instead of disk1,disk2,ect.....

- choose which installed network card you would like to use. could be a dropdown and a forced restart after modification of NIC

 

keep up the good work!

Thanks.

my vote is to have user security for each of the automatic network shares/disks that show up in My Network Places like you do for the user created shares (export mode, exceptions, valid users, invalid users, etc).

 

or atleast have an "exceptions" field so we can export Disk shares as "Export read only" w/ the exception for "root"

I vote to bump up the jumbo frame on the list.

I'd say iSCSI over NFS. It's faster and platform independent.

On the topic of security, I made an assumption but never saw it listed. LDAP or AD-integrated user access is the only way it would work for me. No one's going to maintain a separate set of permissions on a medium-large scale.

The List:

- support multi-core processors & memory beyond 1GB

- add cpuspeed support

- support additional file systems such as XFS, NTFS

 

- NFS support

- AFP support

 

- get time from the internet (ntpd)

 

if this is a linux kernel-based solution, isn't this as easy as compiling these in to the kernel (or modules?) - i guess then just adding them into the UI is all that is left...

 

and ntp is real easy i actually don't use ntpd, i use ntpdate, and then i schedule it to run via cron every hour. would be easy to allow the user to define their own servers and how often they want it to update.

 

since this is a storage OS first, all priority should be on storage related things. performance should be key.

 

important:

- increase write performance

- automatic save/restore of Flash 'config' directory to a designated hard disk file

- email alerts

- hot spares

 

less of a concern... to me this is just more bells and whistles that should be very low priority (this is where SOHO units get lost from the primary goal of a reliable and performant storage unit in my opinion...)

- UPnP

- Integrated wizd server

- Slimserver 6.5.0. support

- other misc servers, eg, "Galleon"

- hamachi vpn

 

I would like to see unraid integrated with LinuxMCE. I think Tom could get more "attention" from his product if it worked plug and play with LinuxMCE.

Hi,

 

I would like to vote for (in that order):

 

- memory beyond 1GB (needed for enhancements)

- UPnP (not Slimserver - it's proprietary)

- increase write performance

- email alerts

- automatic save/restore of Flash 'config' directory to a designated hard disk file

- hot spares

- permit some disks to be outside the array. This will let you select devices which will not be included in the parity-protected array.

- get time from the internet (ntpd)

 

In addition: A possible solution if 15TB are not enough ...

 

Thanks

Harald

 

I propose a plugin/drop in architecture for user provided scripts.

 

The purpose is to allow people to provide scripts that are dropped into a startup and/or shutdown directory without needing to edit the go script

 

Here's an example of what I have for my /boot area.

Granted the location of the scripts can change, I think we just need a consistent way of doing it for everyone.

 

In my particular case, I literally modify the /etc/rc.d/rc.shutdown script to call my hook.

I'm sure Tom could add the hook onto the initroot fs to do the calls without my mod.

Still, it should provide ideas of what can be done.

 

 

root@unraid:/boot/config/rc.local# more rc.local_startup 
#!/bin/bash
logger -trc.local_startup -plocal7.info -is "Initiating Local Custom Startup."

for script in /boot/config/rc.local/*
do scriptbase=${script##*/}      # Strip pathname
   if [ $scriptbase = "rc.local_startup"  ] ;then continue; fi 
   if [ $scriptbase = "rc.local_shutdown" ] ;then continue; fi 
   ( echo "Processing $script" 
     fromdos < $script | sh
   ) 2>&1 | logger -t$scriptbase -plocal7.info -is

   # Old way not used.
   # fromdos < $script > /tmp/$scriptbase
   # chmod u+x /tmp/$scriptbase
   # /tmp/$scriptbase
   # rm /tmp/$scriptbase
done


# Setup shutdown script

RCFILE=rc.local_shutdown
if ! grep /boot/config/rc.local/$RCFILE /etc/rc.d/$RCFILE > /dev/null 2>&1
   then echo /boot/config/rc.local/$RCFILE >> /etc/rc.d/$RCFILE
fi

if [ ! -x /etc/rc.d/$RCFILE ]
   then chmod u+x /etc/rc.d/$RCFILE
fi

logger -trc.local_startup -plocal7.info -is "Local Custom Startup Complete."

 

 

Some example scripts I have in my rc.local directory are

 

root@unraid:/boot/config/rc.local# more S01-blockdev 
#!/bin/bash

if [ ${DEBUG:=0} -gt 1 ]
   then set -x -v
fi

LOOP=10
while [[ $LOOP -gt 1 && ! -b /dev/md1 ]]
do
        (( LOOP=LOOP-1 ))
        echo "Waiting for /dev/md1 to come online ($LOOP)"
        sleep 1
done
sleep 1

for disk in /dev/md*
do blockdev --setra 2048 $disk
done



root@unraid:/boot/config/rc.local# more S10-installpkg 
#!/bin/bash
for package in /boot/packages/*.tgz
do echo "Installing $package"
   installpkg $package 2>&1 | logger -tinstallpkg -plocal7.info -i
done 



root@unraid:/boot/config/rc.local# more S20-init.proftpd 
#!/bin/bash

PACKAGE=openssl-solibs-0.9.8g-i486-1
if [ ! -f /var/log/packages/$PACKAGE ]
   then installpkg /boot/packages/$PACKAGE.tgz
fi

PACKAGE=proftpd-1.3.1-i486-1
if [ ! -f /var/log/packages/$PACKAGE ]
   then installpkg /boot/packages/$PACKAGE.tgz
fi

if grep '^ftp' /etc/inetd.conf >/dev/null
   then # echo "ftp already defined in /etc/inetd.conf"
        exit
fi

if ! grep -v '^#' /etc/inetd.conf | grep proftpd >/dev/null; then
cat <<-EOF >> /etc/inetd.conf
ftp     stream  tcp     nowait  root    /usr/sbin/tcpd  proftpd
EOF

fi

read PID < /var/run/inetd.pid
kill -1 ${PID}

# Add default ftp user if desired or does not exist.
if ! grep '^ftp' /etc/passwd
   then groupadd -g ftp 
        useradd -d /home/ftp -s /bin/false -G ftp -m -c "Anonymouse FTP User" ft
p
fi

# My Preference (Anonymous ftp user to user shares)
usermod -d /mnt/user ftp

 

 

Feature request: Overheating harddisks are spun down.

When harddisks reach critical configurable  templevels, spin them down.

 

Failing fans are a fact of life, and the result should be offline data, not fried disks and destroyed data.

/Rene

indeed good idea

 

 

Feature request: Overheating harddisks are spun down.

When harddisks reach critical configurable  templevels, spin them down.

 

Failing fans are a fact of life, and the result should be offline data, not fried disks and destroyed data.

/Rene

 

I like it.  However, the easier and more immediate implementation is to avoid having a single point of failure.  I have two case fans and one fan blowing over each stock of drives (in addition to the CPU and PSU fans).  Even if one (or perhaps two) of those fans dies, there will be enough air circulation to prevent meltdown.  For example, only one stack of drives had a fan originally, the other drives got ~10C higher, but never reached critical.

 

 

Bill

I like the idea - but the overheating drive may be in use!  If so, it will immediately spin up again.  Maybe there could be two temperatures.  If it hits the first it spins down, if it hits the second, it brings down the array and shuts down the box.