unRAID Server Release 4.5.3 Available

[markguy]

Have you tried 4.5 again?

[stoner]

Tried back both 4.5 and 4.5.1 and works perfectly. Read about the write speed improvements in 4.5.3 and disappointed cannot upgrade... 

[markguy]

"Yet." You forgot that key word, there!

 

I don't have any useful advice for you, I'm afraid. My stock response to something going wrong is to roll back, report and wait for my de-ignorance-ification.

[levster]

Are these cable too good to be true for the price? SAS 8087 to 8087.

 

http://cgi.ebay.com/1ft-Mini-SAS-36-Pin-to-Mini-SAS-36P-Data-Cable-SFF-8087_W0QQitemZ320496621061QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item4a9f165605#ht_3677wt_1167

 

[xamindar]

I just want to make a suggestion to LimeTech for the next release. Someone please correct me if this has already been implemented in this release, but I highly doubt it.

 

Sense drives are going to start moving over to 4k sectors unraid needs to start partitioning them correctly. I don't want to jumper my new drives forever. Linux DOES support these drives at 4k sectors but at the moment you need to partition them specifically, the linux tools can't figure it out for you. Hopefully that will be fixed soon.

 

Is there anywhere LimeTech posts the planned updates for the next version of unraid?

[BRiT]

Is there anywhere LimeTech posts the planned updates for the next version of unraid?

 

No.

 

[speeding_ant]

I've come across a bug where it comes up with "Unformatted" next to all drives when doing a parity upgrade. Scared the crap out of me. Cancelled parity sync, restarted and tried again. Didn't happen again.

[smino]

As of Unraid 4.5.3, can someone direct me to the list of supported raid cards?

Thank You...

[RoninTech]

I've come across a bug where it comes up with "Unformatted" next to all drives when doing a parity upgrade. Scared the crap out of me. Cancelled parity sync, restarted and tried again. Didn't happen again.

 

Yep  I recall that it gave me a bit of a scare also. 

[NAS]

4.5.3 should be removed.

 

we have already seen at least one member who pressed format by mistake. For a project thats whole point is data integrity one member is too many and the potential for more is to high a price to pay.

 

Users shouldnt be required to know not to press this button to avoid potential catastrophic date loss or rather not to press a button when a bug causes that button to appear.

[prostuff1]

While I agree with some of your sentiments, I also think that some common sense needs to be used when this happens.

 

The one thing I never quite understand is why people rush to push this button as soon as they see a drive that pops up unformatted.  I grant that the web interface is reporting it wrongly and it needs to be fixed, but I would NEVER (and have never) pushed the format button when I knew a disk had been previously shown to be good.  If it was once formatted and now magically, out of the blue, showed as unformatted, the first thing I would not do is press the format button.

 

Another issue that is similar is the "restore" button. There is a description next to the button that clear (at least to me) states exactly what it does.  If people would take the time to read the text beside the Start button and the restore button after replacing/adding a disk they should be able to figure out which one is the correct one to push.

 

 

Now, back to the 4.5.3 topic.  Tom posted in one of the threads in the 4.4 subforum that he was out of town until the 17th I think. He said he would be looking into the array appearing with all disks unformatted (even though they are fine) when he got back in.

[NAS]

Agreed you should never press the button or any button you dont know the ramifications but people do.

 

If this release had been beta tested it wouldn't have been classed as stable but it wasn't and is classed as stable.

 

I am not interested in moaning/complaining but if a stable goes out without public beta testing then this is what can happen, and quite simply it shouldn't.

 

No flame intended but to my eye this is a serious breakdown in the release cycle.

[speeding_ant]

I think its also a matter of does the user know what unformatted means? The general user doesn't make informed decisions like people who work with this stuff every day. Just think to the classic UAC dialogue box as a recent example. This update should be pulled or the bug fixed asap.

[purko]

I'll have to agree.

A unexperienced person sees a problem saying "Unformatted", and right next to that is a button saying "Format".

Without much thinking, the "remedy" to the problem seems like just one click away.   

 

[brainbone]

I think most/all unraid users understand what "unformatted" and "format" means... however, when you see "unformatted" there will be an unavoidable emotional response of "crap... I think I just lost everything".   The problem is that some may simply take the frustrated leap of clicking "format" before going through step by step to figure out what the root cause of the "unformatted" message was.  With a message like "unformatted", I think the response of clicking "format" can seem perfectly reasonable -- especially when your mind is already clouded and preoccupied taking inventory of all the data you think you just lost.

 

This whole "unformatted" thing isn't isolated to 4.5.3 though, is it?  I seem to remember it happening in the past.

 

[calvinandh0bbes]

This whole "unformatted" thing isn't isolated to 4.5.3 though, is it?  I seem to remember it happening in the past.

4.4.2 and 4.5 both have shown an "unformatted" status for a formatted drive (including those with lots of data).  Haven't updated to 4.5.3 yet.  A simple refresh usually corrects the error.  Even though I've seen it more than a few times, it always makes me jump a little.

[unraided]

Any word on when the next stable release of unraid server is going to be released? I understand limetech had to create a minor release, though it seems it has caused a issue for some with their disks appearing as unformatted. I rolled back to 4.5.1 only to get wol working again, as it stopped functioning once I upgraded to 4.5.3.

[keland44]

i've been away from unraid for a while last beta i used was 4.5 beta 3 would it be save to say 4.5.3 is the latest and greatest or is there a beta version that's newer then 4.5.3

[BRiT]

4.5 FINAL, 4.5.1, and 4.5.3 are all newer than any of the betas.

[keland44]

4.5 FINAL, 4.5.1, and 4.5.3 are all newer than any of the betas.

 

thanks ;-)

[terrastrife]

Preliminary testing... sustained 40MB/sec writing to parity protected drive over the LAN.... no spikiness.  

bubbaQ,

 

Is that with to user-shares, or disk shares? 

If user-shares, is there a cache drive configured?

 

40 MB/s sustained writing performance over the LAN is outstanding!!!!!!!!!!!!!!!!!!!

so is 40MB/sec not normal? i get this wirting to user shares, but it drops off to 30MB/sec at the last 500GB or so of the disk.

[bubbaQ]

so is 40MB/sec not normal?

 

Not writing to disk shares or user shares w/o cache.

[lionelhutz]

Tom;

 

As I went from 4.3.2 through the beta versions to 4.5.3, I changed the settings of my user shares from this;

 

Share name = Movies

Comment =

Allocation Method = High Water

Split level = 1

Include disks = disk1

exclude disks =  

 

to this

 

Share name = Movies

Comment =

Allocation Method = High Water

min. free space =                

Split level = 1

Include disks = disk1 disk2

exclude disks =  

 

and then finally to this;

 

Share name = Movies

Comment =

Allocation Method = Fill-up

min. free space = 50000000

Split level = 1

Include disks = disk1 disk2

exclude disks =  

 

and somewhee during this the shares quit filling correctly.

 

The directories in my share look like this and they were all on disk1 to start;

1 Movies

2    Old Dogs

2    Up in the Air

2    Planet51

2    James Bond Movies *

3         Die Another Day

3         Live or Let Die

2    Die Hard Movies *

3         Die Hard

3         Live Free or Die Hard **

etc

 

The directories with a * were created blank on disk2 and the directory with a ** was created with the movie contents on disk2 when I copied that move to the share even though the "Die Hard Movies" directory already existed on disk1. 50,000,000 gives around 50gig as the free space and I had something like 170gig free at the time on disk1.

 

It appears that the split level was ignored.

 

I also don't understand why the directories would be blank since the share directories are created when they are needed to hold a file.

 

If you go here on message #11 another user has appeared to have a similar problem.

 

http://lime-technology.com/forum/index.php?topic=5493.msg52761#msg52761

 

Any comments on what went wrong?

 

Peter

[lionelhutz]

With these settings;

 

Share name = Movies

Comment =

Allocation Method = Fill-up

min. free space = 50000000

Split level = 1

Include disks = disk1 disk2

exclude disks =  

 

disk1 has ~223gig free and disk2 has ~675gig free.

 

I ended up with a subdirectory under the Movies share repeated on both disk1 and disk2 with the directories and files in that directory split between the 2 drives AGAIN.

 

It only seems to do it if I've got a subdirectory with directories in it.

 

So, what is up with the split level not working? Tom, are you reading this? I never got a response to the email...

 

Peter

 

[just2bad]

Flash share problem..

 

am  I the only one who have problem to acces the flash share in write mode. The share is exported in read/write and everytime I try to copy, create or modify I'm able to do it like one time after I have access denied on eveymod I try to make until I reboot the server after the reboot I'll able to do 1-2 operation then access denied again till another reboot.  Really anoying