unRAID Server Release 5.0.4-i386 Available

[limetech]

Download

 

Please see

readme.txt

in the release zip file for installation/upgrade instructions.

 

Explanation of changes:

 

- slack: added curl

 

This was per customer request.

 

- samba: upgrade to 3.6.21-p (patched with share point mtime fix)

 

This is to address this issue:

http://lime-technology.com/forum/index.php?topic=30551.0

 

Please report in that thread if the problem is solved.

 

- linux: added additional netfilter options:

 

This was per customer request.

 

- mover: removed -O switch from rsync command, wasn't doing what I thought

 

The process of moving files from the cache disk to the array will modify containing directory's mtime.  This has always been true and is unavoidable using the 'mover' script as-is.  There's a way to fix this but no time to do so at present.  Is it really an issue?

 

- zip release file: upgraded memtest to version 5.01

 

See here for details:

http://www.memtest.org/

 

- zip release file: added make_bootable_mac (thanks to Kyle at Greenleaf, aka, prostuff1)

 

Thanks to Kyle at Greenleaf for doing the ground work in putting together a 'make_bootable_mac' script.  I took his script and modified, while at same time reorganizing the top-level files a bit.  Specifically, created a 'syslinux' directory to gather all the syslinux-specific files.

 

If you are running 5.0.x, it's only necessary to copy bzimage/bzroot/readme.txt to the flash as before.  You can also copy over memtest if you want the new version.  If you want to make the flash look just like the zip file, then you should make a backup of your config directory, plug flash into your PC (or Mac), reformat, re-install, (don't forget to click make_bootable (PC) or make_bootable_mac), then restore your config backup.

 

The 'make_bootable_mac' script seems to work, please follow the directions carefully.  I'm of course open to any ideas for enhancement, etc.

 

Sorry development of 'webGui on github' has taken a back seat in order to solve some other pressing issues.  I'll be able to get back to that now & that should be the next thing released, barring any huge issue with 5.0.4.  After 'webGui on github' has been pushed, next will be release 5.1 which will incorporate whatever current state of webGui is at that point.  Very soon thereafter will be release 6.1, which is just like 5.1 only 64-bit kernel.  Thatz the planz anywayz.

 

One more thing: I haven't completely solved the download disconnection issues, so bear with me on that please.

 

 

[graywolf]

Hope to get a chance to play with it this weekend, but not looking good. On-Call and a major application implementation is having tons of issues.  Wished they would do better QA before rolling into production

[jumperalex]

FWI, downloaded in <10s on FiOS in virginia.  Will install after watching Hell Boy II

[nacat78]

Just wanted to say thanks for the hard work and will be installing 5.0.4 now...

[dalben]

Mine downloaded so quickly that I thought it had failed again.  I was surprised to find it had come down complete and unzipped nicely.

[squirrellydw]

Thanks Tom, so many updates.  As soon as I get around to updating you release another one.  I just installed 5.0.3 a few days ago and now 5.0.4 is out.  Great job and thank you very much.  Look forward to testing this out in a few days.

 

Thanks again and keep up the great work

[needo]

Amazing work. I am looking forward to unRAID 64 bit!

[jbartlett]

Performed the upgrade on a VirtualBox instance running UNRAID 5.0.0 and it kernel panics every boot.

 

EDIT: For those who were concerned, this should be an issue related only to the VM.

EDIT: I installed 5.0.4 on my main UNRAID server (non-virtualized) and no issue

[garycase]

<Sigh>  Just FYI, the download issue is still there ... at least for the one I just did.

 

It DID download okay when I clicked on Retry ... although not really fast (didn't time it, but probably ~ 20 seconds or so).

 

[limetech]

<Sigh>  Just FYI, the download issue is still there ... at least for the one I just did.

 

It DID download okay when I clicked on Retry ... although not really fast (didn't time it, but probably ~ 20 seconds or so).

 

Give it another try.  If you have the download page already in browser, refresh first.  BTW 20 sec is not bad.

[garycase]

Give it another try.  If you have the download page already in browser, refresh first.  BTW 20 sec is not bad.

 

Closed the browser; went to Limetech site; refreshed; clicked on the download; selected "Save As" and named it "... 2nd try" ==> exact same results.

 

Clicked on Retry and as before it downloads just fine (timed it this time -- 23 seconds)

 

Used 7-Zip to "Test archive" for both downloads and they test fine, so I'm sure they are.

 

[limetech]

I changed the mime type of the zip file from "application/zip" to "application/octet-stream" and it seemed to fix it for IE when I tried it.

Just a hunch, gleaned from:

http://blog.akumars.esoftera.in/post/2013/03/28/Download-Issue-in-Internet-Explorer%28IE%29-10.aspx

and

http://blogs.msdn.com/b/ieinternals/archive/2012/07/16/content-length-and-transfer-encoding-validation-in-ie10-download-manager-couldnt-be-downloaded-retry-cancel.aspx

 

Probably the solution is to not use IE 

[mr-hexen]

Got a syslinux menu configuration error on first try of 5.0.4.

 

with regards to the syslinux folder, should menu.c32 be in there or at root??

 

Going to move it to root and try again.

 

UPDATE: didnt work, thinking this has something to do with my new usb device.

 

UPDATE2: old device works fine but have unrelated issues now.

[patm95]

I've downloaded a couple of times now and it states that the file is invalid or corrupt when I try to unzip it.  I tried with the native windows unzip and winrar with same results.  Anyone else have this issue?

[nacat78]

I've downloaded a couple of times now and it states that the file is invalid or corrupt when I try to unzip it.  I tried with the native windows unzip and winrar with same results.  Anyone else have this issue?

 

It depends on your ISP provider dropping connection/timeout with the download server. Its been a problem for some folks lately - PM me is you like to be e-mailed the updated 5.0.4 package - also look at member nars signature - he has a mirror of the downloads avail - here http://lime-technology.com/forum/index.php?topic=30503.30

[lewcass]

I've never had a problem downloading a release until today. (Thought everyone else was just unlucky.  )

 

But I just tried four times and every time the download failed the md5 check sum test. Each file was a different size too, so I suspect none of them were complete.

 

FWIW I am using Firefox 25.0.1.

[patm95]

I've downloaded a couple of times now and it states that the file is invalid or corrupt when I try to unzip it.  I tried with the native windows unzip and winrar with same results.  Anyone else have this issue?

 

It depends on your ISP provider dropping connection/timeout with the download server. Its been a problem for some folks lately - PM me is you like to be e-mailed the updated 5.0.4 package - also look at member nars signature - he has a mirror of the downloads avail - here http://lime-technology.com/forum/index.php?topic=30503.30

 

Thanks was able to download from Nars sig!

[madburg]

Hey Tom,

 

1) First time the download has ever failed for me, it brought down about 14mb (did not fail, Chrome browser, FIOS in the states). Second attempt a few minutes later succeeded.

 

2) smartmontools drivedb update fails in this release, assumption it has something to do with the addition "- slack: added curl"

 

root@Tower:~# /usr/sbin/update-smart-drivedb
curl: error while loading shared libraries: libcurl.so.4: cannot open shared object file: No such file or directory
curl: error while loading shared libraries: libcurl.so.4: cannot open shared object file: No such file or directory
/usr/sbin/update-smart-drivedb: download from trunk failed (HTTP error)

 

3) For "- mover: removed -O switch from rsync command, wasn't doing what I thought"

This spins-up Drive #1 upon execution of mover (just tested under 5.0.4 since this change was mentioned), we worked through this before so you are fully aware as to why this occurs and you added the 'O' as 'the fix' previously.

 

[madburg]

In Addition too #3

 

For an example I created a directory on the cache drive "TV Shows" which is a unRAID share (SMB/AFP), the share consists of disk 12-18.

 

I created a 0kb flat file test.txt in the TV Shows folder on the cache drive.

All drives spun down except for the cache drive.

Execute mover, Disk #1 spun up and disk 12 AND Disk 13 spun up.

 

The file was copied to Disk #13 (which is wrong based on my share configuration most free, should have been disk 18, let forget this for a moment)

 

Disk#1 spun up because of mover (you can explain that one better than I)

Disk#12 spun up because it timestamped JUST the TV Shows directory

and

Disk #13 spun up because the file was copied to it AND timestamp the TV Shows directory on this drive as well.

 

Whats wrong with this picture? If is going to timestamp just because that directory exists then it should have done so on disk 12-18, I don't believe this is the intended behavior it should have only spun up and worked with disk#13 (actual disk #18 but that a whole other issue)

 

Simple test anyone will a fairly large array with shares and a cache drive can try.

 

 

[lewcass]

Hey Tom,

 

1) First time the download has ever failed for me, it brought down about 14mb (did not fail, Chrome browser, FIOS in the states). Second attempt a few minutes later succeeded.

 

Similar experience with Chrome though via Bright House Roadrunner. First download failed check sum. Second attempt OK. Both downloads were the same size, so apparently the first download had been corrupted.

[garycase]

I was looking at my calendar trying to assess just when the download issues started; and while there may be a better record of that buried in the posts here, as near as I can tell it was around mid-October.  The first time I had an issue was at a friend's house on 28 October.    I had never had a problem downloading prior to that ... although admittedly downloading UnRAID isn't something I do every day 

But in the September - October timeframe there was enough activity that frequent downloads were indeed common ... and I don't recall ANY issue with this until late October.

 

Not sure if that helps isolate what may have changed ... but clearly there IS a problem.  Not insurmountable (i.e. with IE if you simply click "Retry" it generally works) ... but the posts above show issues with IE, Firefox, and Chrome => and a download should just work !!

[squirrellydw]

maybe I'm wrong but can we keep this about 5.0.4.  The download issue Tom is aware about and is workig on it.  JMO

[garycase]

maybe I'm wrong but can we keep this about 5.0.4.  The download issue Tom is aware about and is workig on it.  JMO

 

You're right => Any other download comments should be posted to:

http://lime-technology.com/forum/index.php?topic=30517.0

[BobPhoenix]

Barzija did, and posted it on June 09, 2013.

I tried to get to Barzija's version and got this:

Same for me.

[hawihoney]

This 5.0.4 is the first release I can't copy to the flashs of my two machines as usual.

 

I usually copy the complete root folder from the ZIP. That worked for ages. Now I saw that there's a new directory called syslinux. Whenever I try to create/copy that directory on my Windows 8.1 boxes I get errors. So I tried to reformat, make_booatable, ... the sticks. Again whenever I try to create/copy the syslinux folder it is refused.

 

So I'm back to 5.0.3.

 

Can please somebdy explain why the folder named syslinux is refused?

 

Thanks a lot.