unRAID Server release 4.3-beta6 available


Recommended Posts

Download.

 

More changes to fix user-reported problems.

 

unRAID Server 4.3-beta6 Release Notes
=====================================

Changes from 4.3-beta5 to 4.3-beta6
-----------------------------------

Improvement: update to latest Intel network driver (7.6.15.4).

Bug Fix: Correct problem enabling SMART before reading disk temperatures.

Bug Fix: Fix crash that may occur if there is a large number of user shares.


Changes from 4.3-beta4 to 4.3-beta5
-----------------------------------

Bug Fix: Fix 'hang' that may occur starting syslogd.

Improvement: Add 'sync' as last step in 'mover' script


Changes from 4.3-beta3 to 4.3-beta4
-----------------------------------

New feature: cache disk support.

Improvement: enable SMART before reading disk temperature.

Improvement: upgrade from linux kernel 2.6.24.3 to 2.6.24.4 (refer to http://lwn.net/Articles/274741).

Improvement: upgrade from Samba 3.0.28 to Samba 3.0.28a (addresses some Vista issues, refer to http://us1.samba.org/samba/history/samba-3.0.28a.html).

Improvement: added back a few more missing libraries needed for certain user customizations.

Bug Fix: Support normal expansion of array when Parity is not installed.


Changes from 4.3-beta2 to 4.3-beta3
-----------------------------------

Bug fix: Turn off user share filesystem debugging output accidentally left on - doh!


Changes from 4.3-beta1 to 4.3-beta2
-----------------------------------

Bug fix: Restore missing 'installpkg' script.

Bug fix: Fixed possible user share crash if top-level disk directories contain ordinary files.

Bug fix: Fixed warning message generated by 'find' when searching user shares.

Bug fix: Fixed problem reporting correct hostname to DHCP server.


Changes from 4.2.3 to 4.3-beta1
-------------------------------

Improvement: With user shares enabled, directories/files created directly in a disk share will show up immediately in the correct user share.

Improvement: Included 'vsftpd' server.

Improvement: Upgrade to linux kernel 2.6.24.3.

Improvement: Enable kernel support of up to 4GB of memory.


Upgrade Instructions (Please Read Carefully)
============================================

If you are currently running unRAID Server 4.2-beta1 or higher (including 4.2.x 'final'), please copy the following files from the new release to the root of your Flash device:
    bzimage
    bzroot

If you are currently running unRAID server 4.0 or 4.1, please copy the following files from the new release to the root of your Flash device:
    bzimage
    bzroot
    syslinux.cfg
    menu.c32
    memtest

This can be done either by plugging the Flash into your PC or, by copying the files to the 'flash' share on your running server.  The server must then be rebooted.

If you are currently running unRAID Server 3.0-beta1 or higher, please follow these steps to upgrade:

1. Referring to the System Management Utility 'Main' page, make a note of each disks's model/serial number; you will need this information later.

2. Shut down your server, remove the Flash and plug it into your PC.

3. Right-click your Flash device listed under My Computer and select Properties.  Make sure the volume label is set to "UNRAID" (without the quotes) and click OK.  You do NOT need to format the Flash.

4. Copy the files from the new release to the root of your Flash device.

5. Right-click your Flash device listed under My Computer and select Eject.  Remove the Flash, install in your server and power-up.

6. After your server has booted up, the System Management Utility 'Main' page will probably show no devices; this is OK, navigate to the 'Devices' page. Using the model/serial number information gathered in step 1, assign each of your hard drives to the correct disk slot.

7. Go back to the 'Main' page and your devices should appear correctly.  You may now Start the array.


If you are installing this release to a new Flash, please refer to instructions on our website at:

http://www.lime-technology.com/wordpress/?page_id=19

Link to comment

yup, expanded fine as well (i hope my earlier troubles with the zipped bzroot file didn't get anyone paranoid...it was probably just something with my system).

 

i somehow lost parity when i copied the 4.3-beta5 over the 4.3-beta2...i did that, then hit "stop" to get ready to reboot (probably should have done that first), but all of a sudden all drives showed as "unformatted"...made my heart sink for a second, but i realized that must be a glitch...the array wouldn't let me stop it, so i had to do a dirty shutdown...still rebuilding parity (9hrs down, 6hrs to go).

 

i'm really looking forward to try the new cache feature...and getting my temps back with this new build...thanks for keeping on top of it, always, Tom! (just turned a friend in Manhattan - also named Tom - on to your technology, so another 1500-LL/Pro order will be coming in any day now.)

 

Link to comment

I don't know if this is an issue with beta6.  I believe it is an issue related to the new Cache feature.

 

I currently have rsync running to on my unraid server to 'pull' a backup from another machine on my network once per day.  It's been working great for months.  Since rsync is executing on the unraid side I have the rsync destination configured as "/mnt/user/backup/fs1".

 

'Backup' is a user share and I have the cache disk enabled for this share.

 

Rsync appears to be copying the files correctly from the source to the destination on my unraid server, but at the end of the rsync operation I get hundred of errors, one for each file and directory, that look like this:

 

rsync:  chown  "/mnt/user/backup/fs1/temp" failed: no such file or directory

 

I never received these errors before and I'm assuming it has something to do with a change in how permissions (or some other thing) are handled with the /mnt/user/ directory now that we have a /mnt/user0/ directory.

 

What are your thoughts.  Should I change my rsync destination from "/mnt/user/backup/fs1" to "/mnt/user0/backup/fs1" ?

 

I thought maybe If I changed my rsync destination to "/mnt/config/backup/fs1" that it might speed the copy process up a bit but after further thought my guess is that once the mover had moved the cached files to the array the next day when rsync ran rsync would think that the destination is empty and would attempt to back up every file from the source side.  This would not be good.  In my mind I can't see how the new cache feature can benefit my rsync process unless I changed to a 'push' model where files are pushed to my unraid server.

 

Any thoughts?

Link to comment

I just maybe did a bad thing.  I swapped out an Intel Pro 10/100 NIC, swapped in a NetGear GA311 (Realtek 8169), and upgraded from Beta3 to Beta6 at the same time.  The server seemed to boot up fine --the NIC and switch (NetGear GS108) both lit up at 1000Mbps, but I couldn't access it on the network.

 

It's normally headless, so I had to haul a monitor and keyboard in there, and here is the last few lines of the boot-up:

 

etc/rc.d/rc.inet1.conf: line 18: /boot/config/network.cfg: no such file or directory

etc/rc.d/rc.inet1.conf: line 19: /var/temp/network.cfg: no such file or directory

 

Welcome to 2.6.24.4-unRAID

 

Tower Login:

 

There is a network.cfg in the config folder on the USB flash drive.  And I'm a linux noob, but even I can tell that the problem is with that config file.  Do I need a different network.cfg for this NIC?  There is still only one NIC in the machine, in the same PCI slot, and the on-board NIC is still disabled.  I don't recall changing anything when I switched from the onboard NIC to the Intel NIC, so I was hoping it would load the correct drivers automatically.

 

On a separate and much less important note, I would like to enable jumbo frames if I can.  I have a couple 100Mbps machines still on the network, but I have one machine with another GA311 (support up to 7500 bytes) and another machine with built-in Marvel Yukon that supports up to 9K (as does the switch).  If I turn on jumbo frames on the PCs, do I have to do anything on the server?  And if I do all that, will my "regular" PCs still work?

 

 

Link to comment

When you upgraded, did you reformat the USB flash drive, or just copy the new bzroot and bzimage into place.

 

If you did reformat and reload the USB drive, the error you are getting seems to indicate the flash drive did not get mounted at /boot, probably since it does not have an UNRAID volume name.

Link to comment

no reformat--just copied the 2 files (bzroot and bzimage) onto the flash, overwriting the existing ones--just like I did to get to beta 3.  It has been working fine on beta 3 since it's release.  Just thought I'd go to the latest beta since I was taking it down to upgrade the NIC anyway...

Link to comment

If you did not reformat, the odds are it is as you suspected.  The alternate network card does not have a driver in the latest beta.

 

To be certain, type

ls -l /dev/disk/by-label

 

The output should look something like this:

root@Tower:/boot# ls -l /dev/disk/by-label

total 0

lrwxrwxrwx 1 root root 10 Apr  5 09:18 UNRAID -> ../../sda1

 

Joe L.

Link to comment

cohiba-

Since you have a monitor/keyboard, after system boots, log in via console and please copy the system log to the the Flash:

 

cp  /var/log/syslog /boot/syslog.txt

 

Then power down, plug Flash into PC and post the syslog.  While you're at it, also upgrade to 2.4-beta6.

Link to comment

will do that tonight.  I figured Tom would want the syslog, but I forgot the syntax.  So thanks.

 

Tom, I ALREADY upgraded to beta6.  In fact, I REDOWNLOADED beta6 and RECOPIED the bzroot and bzimage files to the root of my flash drive just to make sure I didn't screw it up the first time.  :)

 

So, Tom, beta6 does have the driver and SHOULD be working with my NetGear NIC?

Link to comment

cohiba-

I don't think the problem you're seeing is due to the driver not getting loaded.  Seeing the syslog will be a big help.

 

I'm not completely sure about the state of Realtec support.  This is because we don't have any Realtec-based NIC's to test with, but looking at the linux change log, there have been updates to the built-in driver.

 

BTW, the "built-in" driver is different than the driver supplied on the Realtec site.  We tried downloading and building the latest Realtec driver, but it does not compile with our kernel (2.6.24.4) - gets numerous compile errors.  Realtec is very slow to update their drivers and typically lag pretty far behind linux kernel development.  This is probably why everyone uses the 'built-in' driver, which is maintained independently from Realtec (probably kernel developers get tired of dealing with Realtec lag time to update their drivers - just a guess).

 

Here is latest compile error message:

make[2]: Entering directory `/usr/src/linux-2.6.24.4'
  CC [M]  /usr/src/r8169/src/r8169_n.o
/usr/src/r8169/src/r8169_n.c: In function 'rtl8169_init_one':
/usr/src/r8169/src/r8169_n.c:2431: error: implicit declaration of function 'SET_MODULE_OWNER'
/usr/src/r8169/src/r8169_n.c:2594: error: 'struct net_device' has no member named 'poll'
/usr/src/r8169/src/r8169_n.c:2595: error: 'struct net_device' has no member named 'weight'
/usr/src/r8169/src/r8169_n.c: In function 'rtl8169_rx_interrupt':
/usr/src/r8169/src/r8169_n.c:3526: error: 'struct net_device' has no member named 'quota'
/usr/src/r8169/src/r8169_n.c:3526: warning: type defaults to 'int' in declaration of '_y'
/usr/src/r8169/src/r8169_n.c:3526: error: 'struct net_device' has no member named 'quota'
/usr/src/r8169/src/r8169_n.c:3526: warning: comparison of distinct pointer types lacks a cast
/usr/src/r8169/src/r8169_n.c: In function 'rtl8169_interrupt':
/usr/src/r8169/src/r8169_n.c:3697: error: too few arguments to function 'netif_rx_schedule_prep'
/usr/src/r8169/src/r8169_n.c:3698: error: too few arguments to function '__netif_rx_schedule'
/usr/src/r8169/src/r8169_n.c: In function 'rtl8169_poll':
/usr/src/r8169/src/r8169_n.c:3741: error: 'struct net_device' has no member named 'quota'
/usr/src/r8169/src/r8169_n.c:3741: warning: type defaults to 'int' in declaration of '_y'
/usr/src/r8169/src/r8169_n.c:3741: error: 'struct net_device' has no member named 'quota'
/usr/src/r8169/src/r8169_n.c:3749: error: 'struct net_device' has no member named 'quota'
/usr/src/r8169/src/r8169_n.c:3752: error: too few arguments to function 'netif_rx_complete'
make[3]: *** [/usr/src/r8169/src/r8169_n.o] Error 1
make[2]: *** [_module_/usr/src/r8169/src] Error 2
make[2]: Leaving directory `/usr/src/linux-2.6.24.4'
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/usr/src/r8169-6.005.00/src'
make: *** [modules] Error 2

Link to comment

My programming days are far behind me, but it's almost like it's missing a header file or using an outdated one.  There are only 6 unique errors there, half regarding the struct net_device.

 

I ran the command, it did not give any feedback--just went back to the prompt, so I thought it worked.  But I brought the flash drive down to my PC and there is no syslog.txt on it.  What am I doing wrong?  Regardless, I'll go try again.

 

Ok, I just brought up my server again, and this time got a COMPLETELY different screen.  Here's a pic.  It's different from anything I've seen before, and it actually seemed to load something on eth0, and also see my server name and shares!  I can even browse to the web interface now!

 

I'm afraid to get a syslog because the console is NOT at a login prompt.  What should I do???

Link to comment

Just hit the enter key once or twice... You are at the login prompt, there was just some debugging text sent to the system console that is just confusing you a bit.

 

The "login" prompt will repeat when you press the enter key without entering anything.  When it says Login: type root

 

Joe L.

Link to comment

After you log in, type the command

mount

 

It will let you know if the flash drive got mounted at /boot.

 

Also, try

ls -l /dev/disk/by-label

 

and

 

ls /boot

 

Let us know what the output of those two commands are.

 

Oh yes, I once did exactly the same as you and posted a picture of the console... It helped Tom to recognize the problem and fix an elusive bug in an early version of unRaid... good thinking on your part to post the picture. It would be hard to describe in words.

 

Joe L.

Link to comment
I ran the command, it did not give any feedback--just went back to the prompt, so I thought it worked.  But I brought the flash drive down to my PC and there is no syslog.txt on it.  What am I doing wrong?  Regardless, I'll go try again.

 

There is no feedback if successful.  Make sure you use the corrected version of the command above.  And the syslog is flagged System and Hidden when viewed on a Windows machine, so make sure you can see System and Hidden files.

 

Clearly, Realtec has not updated their driver code.

 

The eth0 messages on your screen are normal, don't indicate whether your network driver is working or not.  Need to see the syslog.

 

Link to comment

Well, I can ping the IP, and get to the WEB console from my PC, so the network driver is definitely working--this time...   ???

 

I show hidden files, but I did have system files still hidden--that explains that.  Kind of.  I just re-ran the syslog command, only this time told it to output to syslog1.txt (attached) because I didn't want to overwrite syslog.txt from the previous unsuccessful boot.  However, there still was no syslog.txt on the flash, so I must've not done it correctly the first time--unless it is deleted when booting?

 

I am most puzzled by the fact that it booted differently this time than the previous 3 or 4 attempts, with absolutely zero change to the hardware or software on the flash?  Hopefully the syslog will still shed some light.

 

EDIT: Oh yeah, the output from the commands looks (to me) relatively "normal", but I don't think it would've looked like this on the previous boot:

 

mount

proc on /proc type proc (rw)

sysfs /sys type sysfs (rw)

usbfs on /proc/bus/usb type usbfs (rw)

/dev/sdc1 on /boot type vfat (rw)

fusectl on /sys/fs/fuse/connections type fusectl (rw)

/dev/md2 on mnt/disk2 type reiserfs (rw,noatime,nodiratime)

/dev/md1 on mnt/disk1 type reiserfs (rw,noatime,nodiratime)

shfs on /mnt/user type fuse.shfs (rw,nosud,nodev)

 

ls -l /dev/disk/by-label

total 0

lrwxrwxrwx l root toot 10 APR 8 17:49 UNRAID --> ../../sdc1

 

ls /boot

bzimage*  config/  license.txt*  menu.c32*  syslinux.cfg*

bzroot*  ldlinux.sys*  memtest*  readme.txt*

 

To my novice eye, this all looks like I would expect--it sees the usb, and my 2 hard drives.  And like I said, I could actually see my shares on the network.  Didn't dare to read/write anything...  :o

 

Also, how do I do a clean power down from the console?  I've tried "shutdown now" or "shutdown -h now" but they don't turn the power off...

Link to comment

Your syslog looks completely fine to me, you are using the r8169 driver for the Realtec chipset.  Is the problem now that you are not seeing a consistent start up?  Try a few more boots, capture the syslog each time.

 

Just a thought (can be ignored), I think it might be better in the future if support issues were not put into announcement threads, but had their own individual threads.

 

Link to comment

I am not surprised.  The server seemed fine this time, so I expected the syslog to be fine too.  Yes, my concern is definitely the problems I was having on every previous boot attempt, and specifically how this might affect reliability even if I do get lucky and get it to boot up again...

 

I kind of agree on support in announcement threads.  However, the following comes to mind:

 

-when I added my post, I don't think there was even a 4.3 forum, let alone a 4.3 beta6 support thread.

 

-once the announcement has been made (especially on something as short-lived as a beta), what else could possibly be more appropriate for follow-up posts in that thread than support issues with that beta???  It's not like Tom is needs an uncluttered "beta6 announcement" thread for something else...   ;)

 

And regardless, thanks for taking the time to help me through this.  I'll go try and start my server again.

 

EDIT: My server seemed to come back up "ok" again.  What the heck?  Nothing changed.  Oh well.  I am still a little jumpy about reliability, but I really really want to check performance with a gigE NIC.  I think I'll just have to chance it. :D

 

EDIT2: Chance it I did.  Went from ~8MBps writes on my old 100Mbps stuff to ~32MBps write just now.  This is just copying a DVD rip from my local hard drive to a (nonparity) share on the server.  Maybe not too scientific, but it's the main writing task I do, so I don't really care about anything else.  8)    Using a little app from NetGear, it showed Network Utilization around 45%.  Wonder if it will get any better when Realtek actually updates their driver????

 

But considering I got the NICs for $10ea and the 8-port switch for $35(AR), I think that is one hell of an ROI!!!  :D

 

Is there any way to enable jumbo frames on the unRAID server?  The GA311 supports up to 7500 bytes, and my 2 main PCs also support jumbo frames.  However, I also have 2 older machines that probably don't--will doing so affect them talking to the server?

Link to comment

My server with 6 year old motherboard upgraded to beta 6 no problem. It now has 11 disk including the cache disk which was an old 250MB disk I had in an external enclosure. The gigE card (Realtek based) worked flawlessly as well.

 

The write times when using the cache were about 2 1/2 (a whopping 16MB/s) times faster than writing to a non-cache disk

 

The mover service must have executed just fine since the cache disk was empty sans directories.

 

Writing files into a pre-existing directory not on the cache disk resulted in a direct write to the share, not having the file get written to the cache. In these cases I had to explicitly create the path on the cache disk first.

 

One hard thine for me to do is to give up control where files are being placed. I have always been very organized as to what goes on what disk...but I think that is from using unRAID 3.x for over two years without bothering to upgrade. The thought of taking down my zero downtime server and possibly messing it up scared me. hehe. After getting a new case and power supply and a few new disks necessitated getting current. And I'm glad I did.

 

Keep up the good work!

 

 

Link to comment

Please god can we stop doing support etc in the Announcement forum. EVERY time i see a new post in this forum i read it as it should be important i.e. an Announcement.... and proportionately none of the time it is.

 

Please no flames or smart arse replys saying Im posting in the very forum I am complaining about. I post here cause it proves a point.

 

Im not saying don't post about things just not here.

Link to comment

NASUser, RObJ posted a similar sentiment a few posts ago, and I've had some further time to reflect as well.  And right now, well, I just disagree.

 

I am not the only person to post about a "support" issue in this thread, so I am not taking this personally, but I think you and RobJ bring up a valid discussion point that I am interested in pursuing...  Hopefully Tom will chime in here with an "official" response, if one becomes appropriate. :)

 

When a new version beta is released, there is not necessarily a forum even created yet for it.  v4.3 is a perfect example.  Although there is a 4.3 forum now, I do not believe there was one when I first posted about my problem here (I remember looking for one).  What is your suggestion for this?

 

Next, what else (in your opinion) besides the first post of "beta6 available" *should* go in the "beta6 available" thread?  What type of extra announcement are you expecting to see?  Especially for something as short-lived as a beta release?   I'm not being smart, I'm honestly curious!

 

Because to me, every "support" issue brought up in this thread is a valid "announcement" about that beta, giving users potentially critical information about the current beta, what works, what doesn't, and what problems users are having with it.  I for one found it very convenient to come to this one thread and find out not only instructions on installing beta6, but also see immediately what the current issues were.  In my case, Tom's post confirms that there is indeed an issue compiling the current 8169 driver...

 

Now, if the forum exists, and especially if Tom were to provide a link in the "announcement" thread to the corresponding "support" thread, then I think we may have the best of both worlds.  But at the same time, I fail to see why the current state is so reprehensible...  ;)

 

This is seriously the first software I've purchased in years (I get most everything I need licensed from my job), and the fantastic support by both Tom and other users is a big reason why I did.  So I will certainly do my best to abide by whatever forum decorum (say that fast 5 times...) is established. 

Link to comment

I am with cohiba on this (although I hate cigars)

 

in a beta anouncement post you normaly expect feedback from people anout it

 

that is the whole point

 

many fora have a special section where only the mod(s) can post something - without replies

Tom can implement this and use it for major announcements (something that would go in the first page if the web page was formed as a portal)

 

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.