Jump to content

Running unRAID with a full Slackware distro


bubbaQ

Recommended Posts

Thanks for your help nick. But moving Makefile and Kconfig into the md directory gives me errors while issuing make oldconfig:

Strange. I just double-checked, and it still works fine for me.  

 

Have you successfully compiled and booted into the kernel (and tested that the unRAID interface works) without this tweak for dm-mod?

 

You're using kernel 2.6.32.9, right?

 

Take a look at the bottom of that Kconfig file.  Make sure it includes the last two lines; it's easy to accidentally miss a line at the bottom of a large copy/paste:

endmenu

endif

 

If that doesn't work, just replace them both with the originals that you copied over from the /unraid/ directory (from the wiki's instructions).  dm-mod might not be absolutely critical...

Link to comment
  • Replies 130
  • Created
  • Last Reply

I have not successfully booted into 2.6.32.9 as my boot disk is on the AOC-SASLP-MV8 controller and I can't for the life of me get the kernel to boot with the scst and mvsas module compiled from http://scst.sourceforge.net/. The only way I can boot into this kernel is if I enable built in support for AOC-SASLP-MV8 but apparently it's not advised to do so with the kernel drivers and instead better to use scst and mvsas.

 

According to another post, ATA support for libsas has been configured in the kernel under

Device Drivers
-> SCSI device support
---->SCSI transports
------>ATA support for libsas (requires libata)

 

I've compiled both mvsas and scst

ls -al  /lib/modules/2.6.32.9-unRAID/extra/
total 2320
drwxr-xr-x 3 root root    4096 Dec 14 16:49 ./
drwxr-xr-x 4 root root    4096 Dec 14 18:31 ../
drwxr-xr-x 2 root root    4096 Dec 14 16:49 dev_handlers/
-rw-r--r-- 1 root root   73889 Dec 14 16:37 mvsas.ko
-rw-r--r-- 1 root root 2276437 Dec 14 16:49 scst.ko

 

On boot up I get a kernel panic something to the effect, "VFS: unable to mount ..."  

Obviously due to the fact that it can't boot the root fs on the device.

 

For those who have installed a slackware system with the AOC-SASLP-MV8, what have you done? Is there a configuration I'm missing? I tried adding modprobe scst and /sbin/modprode mvsas to rc.modules to no avail.

 

Forgot to mention: concerning patches, the following have been used to patch the kernel source:

 

readahead-2.6.32.below11.patch
scst_exec_req_fifo-2.6.32.patch

 

Perhaps I'm missing other patches? Would the fact that dm-mod not loading in some way prevent scst/mvsas from loading as well?

 

Link to comment

I have not successfully booted into 2.6.32.9

I'd strongly recommend getting this to work first, before trying to mess with the dm-mod tweak.

 

 

I think you've answered your own question here :)

On boot up I get a kernel panic something to the effect, "VFS: unable to mount ..."  

Obviously due to the fact that it can't boot the root fs on the device.

 

For those who have installed a slackware system with the AOC-SASLP-MV8, what have you done? Is there a configuration I'm missing? I tried adding modprobe scst and /sbin/modprode mvsas to rc.modules to no avail.

 

The only way I can boot into this kernel is if I enable built in support for AOC-SASLP-MV8

 

If your boot drive is on the raid controller, then you will not be able to boot unless you have the raid controller's drivers built into the kernel.  A module won't work.  Think about it: the module is stored on the hard drive that you're trying to access in order to boot from.  But you can't access that hard drive until the module has been loaded!

 

Another option is to create an initrd (initial ramdisk) which contains the relevant modules. If there's some legitimate reason that the mvsas driver should be built as a module rather than built into the kernel, then this is what you should do.  This isn't something I've ever had great success with (potentially just from lack of really trying) and I've generally avoided the process.  There should be a plethora of guides on how to do it available on Google.  Here's a site that explains what these are and why you might need them.

 

The easiest way to work around all of this, though, (and what I chose to do) is to just put your boot drive somewhere that is easily and directly accessible without going through an external controller.  Directly connected to your motherboard (with the proper kernel drivers built-in), for instance.

Link to comment

That makes a lot of sense. Thanks for your help.

 

After putting the boot drive on another controller I was able to boot and compile the necessary modules for the AOC-SASLP-MV8. Alas, it still did not work. The modules loaded correctly and drives are visible in the OS but not in unRAID for some reason. The dm-mod tweak still gives me the said error (I've checked the files, they seem correct, used wget with pastebin's Download link).

Link to comment

The dm-mod tweak still gives me the said error (I've checked the files, they seem correct, used wget with pastebin's Download link).

I'd forgotten that Makefiles can be extremely picky about using the correct whitespace (you must use tabs, rather than spaces, at the beginning of lines).  Tab characters don't seem to have been preserved in Pastebin when I copy/pasted.

 

Try downloading those two files using wget from here instead.

 

edit: also, the preclear script needs updating for Slackware 13.1.  See my post here for details.  Joe said he'd update the main script, but in the meantime I've made my locally edited version of the preclear script available here.  I make no promises whatsoever about the script, other than it worked for me :)

Link to comment

nick5429, thanks for mentioning and tracking down what the issue was. I was just starting the process of troubleshooting the same issue on my Slackware 13.2 64bit system. At first I was thinking it had something to do with byte endianess with 64bits but it seems it was the changes in 'echo'.

 

I also wanted to note that your device mapper module changes also work on the 2.6.35.x kernel series with unraid 5.0 beta 2.

Link to comment

Thanks Nick, I am now able to compile dm-mod. I put '/sbin/modprobe dm-mod' in rc.modules, putting it as you say '/sbin/modprobe' results in the modprobe usage output. Again, much appreciated.

 

I started from a mimimal install of slackware 13.1 using these tagfiles http://cybercenter.com.pt/?p=15 and then upgraded to -current. As such, I did have to install samba/nfs/ntp/fuse etc. manually. I then change their respective startup files in /etc/rc.d by chmodding a+x.

 

As for the only files I've copied into the main distro from unraid:

 

/unraid/etc/samba/smb.conf

/unraid/usr/local/sbin/emhttp
/unraid/usr/local/sbin/initconfig
/unraid/usr/local/sbin/mover
/unraid/usr/local/sbin/powerdown
/unraid/usr/local/sbin/set_ncq
/unraid/usr/local/sbin/shfs

/unraid/etc/exports-
/unraid/var/spool/cron/crontabs/root-
/unraid/lib/libvolume_id.so.1.1.0 (made the symlink to  /lib/libvolume_id.so.1)

 

I altered the emhttp binary as you mention as well. But for some reason unbeknownst to me, and maybe only to limetech, I can't see any drives connected to the AOC-SASLP-MV8 controller. I know others in the forums have suffered the same faith. Perhaps emhttp does something particular for this controller. I really wish someone running a full slackware install and this card could chime in.

 

Modules loaded are:

 

md_mod                 43389  0 
xor                    12605  1 md_mod
dm_mod                 42102  0 
mvsas                  56833  0 
libsas                 38993  1 mvsas
scst                  298301  1 mvsas
i2c_i801                5588  0 
i2c_core               11884  1 i2c_i801
scsi_transport_sas     15760  2 mvsas,libsas

 

Oh, here's some dmesg output wrt to scst:

 

[1091]: scst: sysfs_work_thread_fn:517:User interface thread started, PID 1091
[876]: scst: scst_register_device:933:Attached to scsi0, channel 0, id 0, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi0, channel 0, id 1, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi1, channel 0, id 0, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi1, channel 0, id 1, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi2, channel 0, id 0, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi3, channel 0, id 0, lun 0, type 0
[876]: scst: scst_register_device:933:Attached to scsi5, channel 0, id 0, lun 0, type 0
[1092]: scst: scst_cmd_thread:4158:Processing thread scstd0 (PID 1092) started
[1093]: scst: scst_cmd_thread:4158:Processing thread scstd1 (PID 1093) started
[1094]: scst: scst_cmd_thread:4158:Processing thread scstd2 (PID 1094) started
[1095]: scst: scst_cmd_thread:4158:Processing thread scstd3 (PID 1095) started
[1096]: scst: scst_init_thread:3913:Init thread started, PID 1096
[1097]: scst: scst_tm_thread:5663:Task management thread started, PID 1097
[876]: scst: init_scst:2351:SCST version 2.1.0-pre1 loaded successfully (max mem for commands 217MB, per device 86MB)
[876]: scst: scst_print_config:2146:Enabled features: EXTRACHECKS, DEBUG
[1098]: scst: scst_global_mgmt_thread:6443:Management thread started, PID 1098
mvsas 0000:01:00.0: mvsas: driver version 0.8.4
mvsas 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
mvsas 0000:01:00.0: setting latency timer to 64
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 124:identify failed
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 491:Init flash rom ok,flash type is 0x103.
scsi 6:0:0:0: Direct-Access     Kingston DataTraveler 102 PMAP PQ: 0 ANSI: 0 CCS
[823]: scst: scst_register_device:933:Attached to scsi6, channel 0, id 0, lun 0, type 0
sd 6:0:0:0: [sdh] 7774208 512-byte logical blocks: (3.98 GB/3.70 GiB)
usb-storage: device scan complete
sd 6:0:0:0: [sdh] Write Protect is off
sd 6:0:0:0: [sdh] Mode Sense: 23 00 00 00
sd 6:0:0:0: [sdh] Assuming drive cache: write through
sd 6:0:0:0: [sdh] Assuming drive cache: write through
sdh: sdh1
sd 6:0:0:0: [sdh] Assuming drive cache: write through
sd 6:0:0:0: [sdh] Attached SCSI removable disk
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 0 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 1 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 2 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 3 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 4 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 5 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 6 SAS ADDRESS 50030480008548a0
/usr/src/scst/trunk/mvsas_tgt/mv_spi.c 529:Phy 7 SAS ADDRESS 50030480008548a0
mvsas 0000:01:00.0: mvsas: PCI-E x4, Bandwidth Usage: 2.5 Gbps
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 0 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 0 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 1 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 1 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 2 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 2 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 3 attach dev info is 40000400
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 3 attach sas addr is 3
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 4 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 4 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 5 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 5 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 6 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 6 attach sas addr is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1246:port 7 attach dev info is 0
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1248:port 7 attach sas addr is 0
scsi7 : mvsas
[876]: scst: __scst_register_target_template:245:***WARNING***: Target driver mvst_scst doesn't support Persistent Reservations
[876]: scst: __scst_register_target_template:262:***WARNING***: Target driver mvst_scst doesn't have enable_target() and/or is_target_enabled() method(s). This is unsafe and can lead that initiators connected on the initialization time can see an unexpected set of devices or no devices at all!
[876]: scst: __scst_register_target_template:315:Target template mvst_scst registered successfully
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 384:phy 3 byte dmaded.
sas: phy-7:3 added to port-7:0, phy_mask:0x8 ( 300000000000000)
sas: DOING DISCOVERY on port 0, pid:1113
/usr/src/scst/trunk/mvsas_tgt/mv_sas.c 1406:found dev[0:5] is gone.

 

Not quite sure about what the warnings mean.

 

A snippet of my syslog. Keep in mind, I have yet to complete the guide.

 

Dec 16 19:24:54 slackware sshd[2122]: error: Could not get shadow information for root
Dec 16 19:25:12 slackware sshd[2141]: error: Could not get shadow information for root
Dec 16 19:25:17 slackware emhttp: get_config_idx: fopen /boot/config/ident.cfg: Not a directory - assigning defaults
Dec 16 19:25:17 slackware emhttp: get_config_idx: fopen /boot/config/network.cfg: Not a directory - assigning defaults
Dec 16 19:25:17 slackware emhttp: _shcmd: shcmd (2): exit status: 1
Dec 16 19:25:17 slackware emhttp: get_config_idx: fopen /boot/config/share.cfg: Not a directory - assigning defaults
Dec 16 19:25:17 slackware emhttp: read_tokens: fopen /boot/config/disk.cfg: Not a directory
Dec 16 19:25:17 slackware kernel: md: unRAID driver removed
Dec 16 19:25:17 slackware kernel: md: unRAID driver 1.1.0 installed
Dec 16 19:25:17 slackware kernel: read_file: error 20 opening /boot/config/super.dat
Dec 16 19:25:17 slackware kernel: md: could not read superblock from /boot/config/super.dat
Dec 16 19:25:17 slackware kernel: md: initializing superblock
Dec 16 19:25:17 slackware kernel: mdcmd (1): set md_num_stripes 1280
Dec 16 19:25:17 slackware kernel: mdcmd (2): set md_write_limit 768
Dec 16 19:25:17 slackware kernel: mdcmd (3): set md_sync_window 288

Link to comment

Has the step by step guide been updated for the latest unraid and slackware 64 bit?

Seriously thinking about it. It would eliminate one computer here, the Multimedia machine!

 

Has anyone tried running unraid on ubunutu or Linux Mint 10? Is it possible if all your replacing is the md files?

I just like the fact the everything works right out of the live dvd install on Linux Mint. Is there a similar build for slackware?

IE, Slackware Mint... grin...

 

I love hacking, just not when I do not have the time to do it anymore.

Link to comment
  • 3 weeks later...
  • 1 month later...

Hi,

 

I just finished building unRAID4.7 over Slackware 12.2 but cannot manage to have my licence working... in the webUI the GUID of my flash drive begins with a bunch of strange character, so it don't match the one I registered.

 

I'm not a linux-king, could somebody let me know if the procedure I followed ( http://www.lime-technology.com/wiki/index.php?title=Installing_unRAID_on_a_full_Slackware_distro ) should allow unRaid to correctly read the GUID ?

 

Thanks

Link to comment
  • 2 weeks later...

Hi,

 

I just finished building unRAID4.7 over Slackware 12.2 but cannot manage to have my licence working... in the webUI the GUID of my flash drive begins with a bunch of strange character, so it don't match the one I registered.

 

I'm not a linux-king, could somebody let me know if the procedure I followed ( http://www.lime-technology.com/wiki/index.php?title=Installing_unRAID_on_a_full_Slackware_distro ) should allow unRaid to correctly read the GUID ?

 

Thanks

 

I will be doing the same here shortly.  Just got Slackware 12.2 installed.  I am having some lilo/mbr issues with my clean install.  Once I get those resolved, I will report back on if I am having issues with my USB stick/GUID.

Link to comment
  • 2 weeks later...

I will be doing the same here shortly.   Just got Slackware 12.2 installed.   I am having some lilo/mbr issues with my clean install.  Once I get those resolved, I will report back on if I am having issues with my USB stick/GUID.

 

I just reinstalled everything but still have the following GUID : -?X?-???-c?L?0000000

 

My stick is registered with the following GUID : XXXX-XXXX-0000-00000000001C (it'is correctly displayed in Slackware shell using lsusb)

 

do you had a chance to run a test on your new installation ?

 

Thanks for your feedback

Link to comment

Your USB GUID issue may be related to the rather ancient version of the UDEV subsystem that Slackware 12.2 runs. The UDEV package that unRAID 4.5/4.6/4.7 uses is newer than what Slackware 12.2 but older than what Slackware 13.1 utilizes. You could try updating the UDEV package to see if it alleviates your issue. You could also try installing unRAID on top of a Slackware 13.0 system.

Link to comment
  • 2 months later...

Hi everyone,

 

I'm currently trying to get unRAID 4.7 up and running on a Slackware 13.1 installation and am running into a small snag. I followed the Installing unRAID on a full Slackware distro instructions as best I could and took into account the changes suggested here and here. At this point I have a booting, fully functional Slackware system. I was able to build and load the md-mod kernel module and successfully start the emhttp web server. To keep things simple and (hopefully) eliminate any unnecessary errors, I choose to run the 2.32.9 kernel (the version used by unRAID 4.7).

 

The problem I'm having is this: I can't bring up the array. When I got to the step where I reattached all my array drives and created the symlinks to the flash drive, unRAID successfully identified all disks and correctly assigned them to their proper place on the devices page in the GUI (drive order, parity, and cache were all correct). However, the start button on the GUI main page is disabled and there is the following message:

 

No array disk devices have been assigned!

 

I cat'd /proc/mdcmd and noticed the following line:

 

mdState=ERROR:NO_RAID_DISKS

 

The same error can also be found in a hidden form element on the main GUI page:

 

<input type="hidden" name="startState" value="ERROR:NO_RAID_DISKS">

 

There are no other errors (that I've seen), and otherwise things seem to be working properly. I can make changes through the GUI and unRAID is properly identifying my license on the flash drive. Any ideas what's going on here? I'm currently running unRAID 4.5 on the flash drive and have installed 4.7 on the Slackware drive, could that somehow be causing me problems?

 

Any help at all would be great!

 

Thanks,

Kierse

Link to comment

You did not follow the right instructions.  The Wiki article is NOT for unRAID 4.7 and NOT for Slackware 13.x.

 

Thanks for the reply bubbaQ. As I mentioned, I followed the wiki instructions as best I could where I could. The other two links I referenced, Full Slackware install trouble and and earlier post in Running unRAID with a full Slackware distro provide additional detail and come from other users who were able to get things going with Slackware 13.x and versions of unRAID newer than 4.5.

 

Kierse

Link to comment

You likely don't have the proper drovers loaded.

 

Can you hdparm your drives?  What does lsscsi show?

 

I suspect it is something simple like not having a driver loaded. I can run hdparm on all drives, including my boot drive. Running lssci shows all connected drives, including the boot drive and unRAID thumb drive. Going over syslog I noticed the following:

 

May 16 12:30:46 Tower kernel: md: unRAID driver 1.1.1 installed
May 16 12:30:46 Tower kernel: md: could not read superblock from <NULL>
May 16 12:30:46 Tower kernel: md: initializing superblock

 

Kierse

Link to comment

Did you label the flash, and add the flash to fstab?

 

What do you mean by 'label'? I mounted the flash drive (by name) to /flash and symlinked /boot/config to /flash/config. I also made the appropriate entry in /etc/fstab so that it is automatically mounted at boot.

 

What does mount show? 

 

Nothing, the only drives it shows as being mounted are the boot drive and the unRAID flash.

Link to comment
  • 2 weeks later...

So... I don't suppose there's a version of unRAID that works with 64-bit Slackware 13.37, eh?  I dropped the updated md files from my 4.7 USBkey into the kernel and ran into build problems compiling md.c and when I did a diff, I saw that there was quite a bit more than just a few changes... I was checking out the wiki (the article is somewhat old, so I figured I'd try a newer version of Slackware) and I don't know that I'm ready to update to 5.x yet, but I could if necessary, or if there's another version of 64-bit Slackware I should use instead to get my unRAID4.7, I'd be happy to give that a shot... The important part is that I need to get this running on 64-bit...

Link to comment

There is. See my other thread about getting unRAID to work on the Slackware 64bit distros [ http://lime-technology.com/forum/index.php?topic=5639.0 ]

 

When I first did it, I used Slackware 13.0 and unRAID 4.5.x which used an older kernel and did not need unraid kernel source code modifications.

 

The basic gist is you install Slackware 13.37 64bit, then you install the Alien MultiLib packages [ http://alien.slackbook.org/dokuwiki/doku.php?id=slackware:multilib ], then you follow the unRAID 5.x on Slackware installation instructions, then you copy over what libraries (from /usr/lib/ and possibly /lib) be needed from the 32bit unRAID 5.x series as determined by using "ldd" on /usr/local/sbin/emhttp and /usr/local/sbin/shfs, then you'll need to deal with the custom kernel which will require modifications to the unraid.c and md.c source code files to be usable on the linux kernel  2.6.37.x series (changes and removal of barriers), then you compile and install your new kernel. You're now good to go.  8)

 

I have unRAID 5.06a (and 5.06d) working perfectly fine on a Slackware Current 64bit distro which is basically 13.37 plus newer changes since release.

Link to comment

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...