unRAID Server Release 5.0-beta6a Available

[Whaler_99]

FYI - just went back to 4.7 and everything comes up fine.

 

The parity disk, may have been used in a 5bx version - as I was testing these before I went with 4.7 for my final production build. I do believe I even had 5b6 on it before going to 4.7.  Not sure if that makes any difference.

 

Let me know if there is any info or commands you guys would like run on either version to get more info...

 

Shawn

 

EDIT - just noticed, after boot up, it automatically started a parity check...

[BRiT]

The last 6 bytes before the partition table is no mystery.

 

Bytes 440-443 of the MBR is reserved for the optional Disk Signature. Win OS and LILO and some other boot loaders use this to identify the disks should BIOS ever switch up their order. This is how LILO is able to correctly identify the proper root and boot disks.

 

Bytes 444-445 is typically 00s but can be anything so some boot loaders such as LILO use it as a magic checksum flag.

 

Bytes 446-509 is reserved for the partition table.

 

Bytes 510-511 is required to be 0xAA55 as it specifies the MBR signature.

 

Reference: http://en.wikipedia.org/wiki/Master_boot_record

 

[PeterB]

The last 6 bytes before the partition table is no mystery.

 

Ah, okay - thanks for pointing that out.  The mbr definition I found just said that the first 446 bytes was for the bootloader.

 

So, who is allowed to set a disk id?  In beelyte's case it would seem most likely that it is the motherboard BIOS, since there is no sign of a bootloader, or any other user of the mbr.

 

So we have no real mystery - just (at least three) other uses of the mbr which unRAID is not allowing for because it sets all of these other locations to zero.

 

In light of all this, I do agree with you that Tom will have to seriously consider revising the checks which unRAID applies to the mbr - the most compelling argument is that the BIOS (mobo, or hba) can make changes to the mbr as soon as unRAID has created it. As a minimum, bytes 440-445 will have to be ignored but, I guess that it is still up to Tom to decide whether he wants to allow non-unRAID partitions and a bootloader - to some extent, these two items go hand-in-hand.

[BRiT]

I'm fine with the strict MBR checks for drives new to the array, but believe they must be relaxed for drives already in the array.

 

What might make this a bit of a non-issue or at least change the checks entirely is the need to deal with GUID Partition Tables (GPT) required to support drives larger than 2.2 TB. Things get kind of funky there due to protective measures for a redundant GPT at the very end of the drive as well.

[Gizmotoy]

When diagnosing a problem with the pre_clear script, we came across something Joe L mentioned I should post here as a potential bug.  It appears that my disk.cfg file is out of sync with the status of the server.

 

This is my disk.cfg file

# Disk configuration
spindownDelay=3
queueDepth=1
spinupGroups=yes
defaultFormat=2
md_num_stripes_default=1280
md_write_limit_default=768
md_sync_window_default=288
parity=pci-0000:01:00.0-scsi-1:0:0:0
disk1=pci-0000:02:00.0-scsi-0:0:0:0
disk2=pci-0000:01:00.0-scsi-0:0:0:0
diskSpindownDelay.0=-1
diskSpindownDelay.1=-1
diskSpindownDelay.2=-1
diskSpinupGroup.0=host3
diskSpinupGroup.1=host1
diskSpinupGroup.2=host2

 

and this is the output of the command ls --time-style='+%Y-%m-%d %I:%M%p' -ld /dev/disk/by-path/*

lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1d.0-usb-0:1.5:1.0-scsi-0:0:0:0 -> ../../sdb
lrwxrwxrwx 1 root root 10 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1d.0-usb-0:1.5:1.0-scsi-0:0:0:0-part1 -> ../../sdb1
lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-0:0:0:0 -> ../../sdd
lrwxrwxrwx 1 root root 10 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-0:0:0:0-part1 -> ../../sdd1
lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-1:0:0:0 -> ../../sde
lrwxrwxrwx 1 root root 10 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-1:0:0:0-part1 -> ../../sde1
lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-4:0:0:0 -> ../../sdf
lrwxrwxrwx 1 root root 10 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:00:1f.2-scsi-4:0:0:0-part1 -> ../../sdf1
lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:02:00.0-scsi-0:0:0:0 -> ../../sda
lrwxrwxrwx 1 root root  9 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:02:00.0-scsi-1:0:0:0 -> ../../sdc
lrwxrwxrwx 1 root root 10 2011-04-08 12:07AM /dev/disk/by-path/pci-0000:02:00.0-scsi-1:0:0:0-part1 -> ../../sdc1

 

Here you see that disk2 and parity from disk.cfg are not even valid devices.  Disk1 is a valid device, but it points to device sda, which is not actually in the array.  My array devices are sdd, sde, and sdf.

 

But, the server runs perfectly fine, so it appears that it is ignoring the incorrect parameters in this file somehow.

[WingmanNZ]

I have a question, if I am building a new UNRAID rig from scratch. Can i go right ahead and use 5.0B6a? and use the preclear script? will i run into any issues with MBR's? whats the best way to super-clean the MBR etc so its perfect for UNRAID. I am wanting to have 10 1.5TB WDEARS HDDs in one pool with two shares passed through to my Xeon xserve using NFS. its my hope to have my xserve share out the NFS shares over SMB and AFP, i can also use OS X to manage the ACL's

 

Is all that possible with this beta?

 

I will be using 6 onboard SATA and 4 from an IBM Br10i

 

cheers

[Joe L.]

I have a question, if I am building a new UNRAID rig from scratch. Can i go right ahead and use 5.0B6a? and use the preclear script? will i run into any issues with MBR's? whats the best way to super-clean the MBR etc so its perfect for UNRAID. I am wanting to have 10 1.5TB WDEARS HDDs in one pool with two shares passed through to my Xeon xserve using NFS. its my hope to have my xserve share out the NFS shares over SMB and AFP, i can also use OS X to manage the ACL's

 

Is all that possible with this beta?

 

I will be using 6 onboard SATA and 4 from an IBM Br10i

 

cheers

From scratch with a brand new array you should have no issues at all.  Yes, use the most recent version of the preclear_disk.sh script.  It is the best way to prepare the MBR.

 

Joe L.

[SSD]

And you must use 5.0b6a with the BR10i. Support for spindown is not working, but there is a workarounds that works for some. Make sure you read this announcements thread.

[lewcass]

I gave this release a try with basic three drive unRAID. Followed Tom's installation instructions to the letter, except for not disabling all user authored utilities, as it seems there haven't been any issues reported with any of them. Also did not create any additional users, as I wasn't using AFP. Just left root as with earlier versions.

 

It looked like things were working fine from the Web Interface and unMENU. But soon found I wasn't able to access one of the two disk shares, or any of my user shares, or the flash drive, from a Windows 7 64 client. Only one disk was accessible. The others all showed up, but acted as though I didn't have access permission. I double-checked that all that Samba sharing/exporting was set up for all drives with workgroup security. (Also enabled NFS export - but didn't get a chance to test that.)

 

Still no success. So I tried rebooting unRAID and also the Windows client machine. This made the problem much worse. Now //tower is not accessible at all via Windows Explorer, web interface or telnet.

 

Thought I had better share my syslogs and ask for guidance, before trying anything else. I am attaching two logs. Though, since I see they're over the 192KB attachment limit,  on this post is the initial boot of 5.0-b6a with completion of configuration. I'll followup with a second post with the log of the reboot.

 

I'm looking for guidance as to what went wrong, and what I can do about it. Also want to know if it is safe to reboot with my flash backup of 4.6rc3, which was my previous working configuration. (And, yes, I did install 4.7 and verify it was working, as instructed, before installing 5.0.6a.)

 

syslog-20110411-121648.txt

[lewcass]

Here is the 1st part of the second log (the entire log is over the 192KB) limit.

syslog-20110411-121656_part_1.txt

[lewcass]

Log 2, part  2:

syslog-20110411-121656_part_2.txt

[bcbgboy13]

to lewcass - you have permission issue and I believe under Utilities you have to go into running a "permission" script - you have to click on the check mark and let Unraid change the permissions/ownership of all the files and folders (it may take a while if you have many disks with multiple shares) After that everything will be OK with your Win7 access.

[lewcass]

to lewcass - you have permission issue and I believe under Utilities you have to go into running a "permission" script - you have to click on the check mark and let Unraid change the permissions/ownership of all the files and folders (it may take a while if you have many disks with multiple shares) After that everything will be OK with your Win7 access.

I already did that when I first booted with 5.0-b6a. As I said, I followed the instructions. The script completed for both data drives. Of course now, if I wanted to attempt to run it again, I'm not even able to access the web menu.

 

I can access the terminal via a monitor connected directly to the server, and could attempt to troubleshoot from there, with a little guidance...

[WingmanNZ]

I have a question, if I am building a new UNRAID rig from scratch. Can i go right ahead and use 5.0B6a? and use the preclear script? will i run into any issues with MBR's? whats the best way to super-clean the MBR etc so its perfect for UNRAID. I am wanting to have 10 1.5TB WDEARS HDDs in one pool with two shares passed through to my Xeon xserve using NFS. its my hope to have my xserve share out the NFS shares over SMB and AFP, i can also use OS X to manage the ACL's

 

Is all that possible with this beta?

 

I will be using 6 onboard SATA and 4 from an IBM Br10i

 

cheers

From scratch with a brand new array you should have no issues at all.  Yes, use the most recent version of the preclear_disk.sh script.  It is the best way to prepare the MBR.

 

Joe L.

 

One other thing, if i dont want Parity (I will have 1:1 backups of each drive by mirroring from the SMB drive share to an external HDD of the same size) can i just not add a parity drive to the array, that way i get storage pooling and the performance?

 

also if I use the "fillup" method, what happens if i have a TV show stored on one drvie and upon flowing over to the second drive will in move the entire show folder? or just create another folder and start adding the files to that?

 

sorry i know this doesnt really go here but i just wondered if anything had been changed regarding the storing of Data in 5.0b6a

 

cheers and thanks to all. Im hoping to start this build this weekend!

[Rajahal]

 

One other thing, if i dont want Parity (I will have 1:1 backups of each drive by mirroring from the SMB drive share to an external HDD of the same size) can i just not add a parity drive to the array, that way i get storage pooling and the performance?

 

Yes, this will work.  The only issue I see with it is that there's the potential for your pooled storage to become larger than you can fit on a single external drive, so you'll just have to watch out for that.  If you are trying to make use of a bunch of smaller drives and a few larger externals then this approach makes sense.

 

 

also if I use the "fillup" method, what happens if i have a TV show stored on one drvie and upon flowing over to the second drive will in move the entire show folder? or just create another folder and start adding the files to that?

 

unRAID will create a new folder on the second drive.  The original folder will not be affected.  Through disk shares you'll see both folders, through user shares you'll see one big folder or share that contains the contents of both.

 

sorry i know this doesnt really go here but i just wondered if anything had been changed regarding the storing of Data in 5.0b6a

 

No, I don't believe it has.  The ways in which you can access said data has changed, however.  There are many more options now (AFP, via Web GUI, etc.).

[lewcass]

I'm looking for guidance as to what went wrong, and what I can do about it. Also want to know if it is safe to reboot with my flash backup of 4.6rc3, which was my previous working configuration. (And, yes, I did install 4.7 and verify it was working, as instructed, before installing 5.0.6a.)

 

I was able to successfully revert to 4.6rc3. For whatever reason it did initiate a parity check when I booted but everything appears to be OK.

 

I don't have time now to further pursue what all went wrong with 5.0.6a, but it seems almost certain that there were one or more issues with the utilities I have been running. One of those was SNAP, which I see from more thorough reading here is not yet compatible with release 5. I had misinterpreted Joe L's post about there being few compatibility issues. Should have done more of my own homework.

 

Probably the way to proceed now would be for me to try 5.0.6a again with my go script commented out, then gradually enable unMENU, packages, s3.sh, modprobe powernow-k8.  Not going to bother though until SNAP is ready for unRAID 5. Lesson learned for me was to proceed more carefully in the future.

[Joe L.]

I'm looking for guidance as to what went wrong, and what I can do about it. Also want to know if it is safe to reboot with my flash backup of 4.6rc3, which was my previous working configuration. (And, yes, I did install 4.7 and verify it was working, as instructed, before installing 5.0.6a.)

 

I was able to successfully revert to 4.6rc3. For whatever reason it did initiate a parity check when I booted but everything appears to be OK.

 

I don't have time now to further pursue what all went wrong with 5.0.6a, but it seems almost certain that there were one or more issues with the utilities I have been running. One of those was SNAP, which I see from more thorough reading here is not yet compatible with release 5. I had misinterpreted Joe L's post about there being few compatibility issues. Should have done more of my own homework.

 

Probably the way to proceed now would be for me to try 5.0.6a again with my go script commented out, then gradually enable unMENU, packages, s3.sh, modprobe powernow-k8.  Not going to bother though until SNAP is ready for unRAID 5. Lesson learned for me was to proceed more carefully in the future.

You had asked about packages installed from unMENU.  SNAP is not installed from unMENU. 

 

I don't know of issues of packages installed from unMENU.  (but then I don't use all of them either, but nobody has let me know of them)

 

I had no knowledge if SNAP is compatible or not... I do not use it. (and apparently, it is not at this time) 

 

Joe L.

[lewcass]

You had asked about packages installed from unMENU.  SNAP is not installed from unMENU. 

 

I don't know of issues of packages installed from unMENU.  (but then I don't use all of them either, but nobody has let me know of them)

 

I had no knowledge if SNAP is compatible or not... I do not use it. (and apparently, it is not at this time) 

 

Actually I wasn't the one who asked. But moot point, it's all on me for not reading carefully and jumping to the wrong conclusion.

[Gizmotoy]

I'm looking for guidance as to what went wrong, and what I can do about it. Also want to know if it is safe to reboot with my flash backup of 4.6rc3, which was my previous working configuration. (And, yes, I did install 4.7 and verify it was working, as instructed, before installing 5.0.6a.)

 

I was able to successfully revert to 4.6rc3. For whatever reason it did initiate a parity check when I booted but everything appears to be OK.

 

I don't have time now to further pursue what all went wrong with 5.0.6a, but it seems almost certain that there were one or more issues with the utilities I have been running. One of those was SNAP, which I see from more thorough reading here is not yet compatible with release 5. I had misinterpreted Joe L's post about there being few compatibility issues. Should have done more of my own homework.

 

Probably the way to proceed now would be for me to try 5.0.6a again with my go script commented out, then gradually enable unMENU, packages, s3.sh, modprobe powernow-k8.  Not going to bother though until SNAP is ready for unRAID 5. Lesson learned for me was to proceed more carefully in the future.

You had asked about packages installed from unMENU.  SNAP is not installed from unMENU.  

 

I don't know of issues of packages installed from unMENU.  (but then I don't use all of them either, but nobody has let me know of them)

 

I had no knowledge if SNAP is compatible or not... I do not use it. (and apparently, it is not at this time)  

 

Joe L.

 

SNAP is not currently compatible with 5.0b6a.  I have been helping queeg test out some code over the past few days, and it seems like he's got it sorted out.  I have a modified version of the script running fine on my server at this time.  I can't speak for him, but I would expect support should come along shortly.

[mrcsx]

I just upgraded both my UnRaid servers to 5.0b6a and they run great with one minor problem.  When I mount both servers via AFP, I only get the shares of the server which mounted first, when I click on the second server, it shows the shares of the first.

 

My servers are named Kirk and Spock.  Kirk has a share called TV Shows, Spock has a share called Movies.

 

If I click on Kirk and mount the TV shows folder, when I click on Spock on the sidebar on my iMac, it shows the TV Shows folder.

 

Any suggestions?  There must be something simple I'm missing!!

 

Thanks

[johnodon]

3 - There is a double scrollbar on the far right (see attached screenshot)

4 - The banner image is too narrow - when it repeats on the top it leaves an unattractive vertical line (see attached screenshot)

 

I have always had these two issues...back to 5b2 but it only happens on my workstation at home that is running 1920x1080.  The odd thing is that my work laptop runs the same rez and is fine.    All other rez's I have tried seem to scale properly.

 

You will also notice that there is a scrollbar at teh bottom.

 

John

 

Quick update...

 

Both issues decided to go away when I updated my browser to IE9.

 

John

[prostuff1]

Finally had a chance to upgrade from 5.0b2 and attached are two files as explained in the General forum.

 

I did not see any of the problems outlined in the thread and none of my drives poped up with MBR unknown.

disk_mbr_debug50b2.txt

disk_mbr_debug50b6a.txt

[lostwebb]

I might have missed it but where is the GUID in this version? 

[Joe L.]

I might have missed it but where is the GUID in this version? 

Click on the word "flash" on the left side of the main page in the management interface.  You'll find it there.

[lostwebb]

Thanks alot Joe! Silly me thought I had pressed everything!