Tried to add two drives to my array, problems[SOLVED]

Hi community,

 

Am using unraid pro lincense with version 5 RC8a version. Was running out of spave so I bought two extra wd20EARX drives. Tried to preclear them first one succeeded. The other I am not sure about. Added one precleared drive to the array.

Started the array and wanted to move stuff from my laptop to the array. Get an error message on my system that there is not enough space left (See screenshot) According to me there shpuld be enough space. Looked at the shares and according to me the setting are okay, all disks of the array should be included.

 

Can anyone help with this. Next challenge is the second new disk, also a WD20EARX, tried to preclear it several times but it stops, making it almost impossible to use the web interface. Via putty I can acces the system but cannot do more than a clean powerdown. Tried again this evening and saved the syslog. Can anyone with some knowledge of syslogs share some ideas, is there something wrong with this disk? It is a brand new one, if there is something wrong I can return it to the shop telling it is a doa.

 

Third question, when I look into my shares I see folders that I cannoth delete, the are empty according to me but a delete attempt doe not work. Not even in midnight commander.

 

Hope you guys can help, looked in the forums but cannot find specific info or similr info to my probs, or I must have overlooked them. Attached screenshots and a syslog.

syslog-2012-11-12.zip

The deletion issue is a bug in SAMBA apparently. Check the release thread for information on upgrading SAMBA, I'm running into the same deletion issue myself and will be upgrading my code ASAP.

Upgrade your motherboard BIOS to the latest one available - you are four revisions in the past and then try again.

 

With RC8a- you will have to manually install the SAMBA - more info is hidden in the release thread

Will try updating the bios. will report back.

Anyone have some info about the drive, is there something wrong with that?

On samba 3.6.8 now and bios updated to the latest version. Still no joy. Cannot put files on my server even though there is plenty of room.

 

Second problem, one of the two WD20EARX drives I added to the system cannot be precleared. Reason unknown to me.

 

Anyone?

On samba 3.6.8 now and bios updated to the latest version. Still no joy. Cannot put files on my server even though there is plenty of room.

 

Second problem, one of the two WD20EARX drives I added to the system cannot be precleared. Reason unknown to me.

 

Anyone?

 

You have two separate issues going on here.  Let us look at them one at a time.  Begin by ignoring the preclearing problem with that one particular  drive. 

 

Go to the Settings page on the web interface, and click on the SMB icon in the Network Service block.  Your settings for the Enable SMB:  should be "Yes (Workgroup)".  The setting for Workgroup: should be the name of your workgroup on your Windows machines.  (I would be careful that you are  not using non English characters in the workgroup name--- for example, Ä.  )  Set the Local Master to "Yes"

 

Be sure to click on Apply buttons for both sections. 

 

With Windows Explorer, click on the "Network" and you should find your server's name, Tower,  listed.  Now click on the Tower icon and your share "MEDIA" on the unRAID server should show up. 

 

Now try to copy a file to the server.  Post back your results and any error messages on your windows machine-- preferably in English. 

Post a new syslog and the smart status for the HD in question.

Dear Frank1940,

 

Here is a screenshot from the smb settings on my system.

 

Windows being Dutch it is hard to make it in english.

Translated by me the error message is something like this:

 

There is not enough space on Media (//tower) availiable.

 

9.83 GB is needed to copy this item. Delete or move files to obtain this space.

 

Make space availiable and try again.

 

So no joy here.

 

Hope this helps.

 

Dear bcbgboy13,

 

Attached is the latest syslog. Can add preclear reports if you like.

 

Hope you can help.

 

syslog-2012-11-15.zip

It appears that this 2TB HD is bad

 

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: SMART Self-test log structure revision number 1

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: # 1  Extended offline    Completed: read failure      70%        92        388562688

 

If you want you can try to connect it to a motherboard port and try to run both SMART tests (from Unmenu) to see it it will pass.

If it cannot past these two tests you should RMA the disk.

Your drives are beginning to fill up and you may have a couple of issues that are associated with what happens as that occurs.  Try copy a very small file (<25KB) to the MEDIA share and see what happens.  If it copies, then you have run into the issue of almost full drives. 

 

Read this section of the unofficial Wiki:

 

    http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Min._Free_Space

 

You need to set a Min. Free Space on you MEDIA Share page.  (You posted this page in Reply # 4 of this thread.)

 

Also read the thread on Split level:

 

    http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Split_level

 

The one on split levels will probably confuse you on the first reading.  (Most native English speaking people have a problem with it--- so don't feel bad!)  The split level is used to force all files in a directory onto a single disk.  Unfortunately, it can cause a copy failure if the drive (that a file has to be copied to) does not have enough space on it.  This type of failure will occur even though the Share has plenty on room on it.  Your current Split Level is set to 3. 

Frank,

 

Tried to copy a very small file but that didn't work either. First put the shares on 10GB minimal space and splitlevel 2.

All shares are like that. All disks included, none excluded.

 

I'm lost. Still but still tank you for trying to help.

It appears that this 2TB HD is bad

 

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: SMART Self-test log structure revision number 1

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: # 1  Extended offline    Completed: read failure      70%        92        388562688

 

If you want you can try to connect it to a motherboard port and try to run both SMART tests (from Unmenu) to see it it will pass.

If it cannot past these two tests you should RMA the disk.

That line simply says there was one un-readable sector on the disk, and the long-test was terminated. (It always terminates if is finds a failure)  There are several thousand spare sectors available for re-allocation on today's larger disks. 

Frank,

 

Tried to copy a very small file but that didn't work either. First put the shares on 10GB minimal space and splitlevel 2.

All shares are like that. All disks included, none excluded.

 

I'm lost. Still but still tank you for trying to help.

 

IF the split level is causing a problem, you went the wrong way.  Try changing the split level to 200.  (I know this is very large but this value is only for a test.)

There are global disk include and exclude settings. I beliee they are on the settings tab under the shares icon. Make sure they aren't exluding the new disk.

 

The min free space setting will not help if the split level does not allow the directory split you are trying to create. The main use for the min free space setting is to prevent the fill-up allocation method from attempting to "over" fill a disk and giving an error. I've run without any min free space settings for years without an issue. I only use the min free space setting for the cache disk.

settings are like this. see atteachment

for the test I put the split level to 200, restarted the server and still the same error.

Anyone?

 

Anyone?

Remove all add-ons and see if the problem persists.

No add ons present, except snap. Before i installed snap it did not work. Hope this helps.

It appears that this 2TB HD is bad

 

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: SMART Self-test log structure revision number 1

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error

Nov 14 20:37:15 Tower preclear_disk-diff[30189]: # 1  Extended offline    Completed: read failure      70%        92        388562688

 

If you want you can try to connect it to a motherboard port and try to run both SMART tests (from Unmenu) to see it it will pass.

If it cannot past these two tests you should RMA the disk.

That line simply says there was one un-readable sector on the disk, and the long-test was terminated. (It always terminates if is finds a failure)  There are several thousand spare sectors available for re-allocation on today's larger disks.

 

This drive has failed testing and there is no indication of a reallocated or pending sector. If this devise can not pass testing is should be removed.

Are the shares you're trying to copy to allowed to use disks with space? If you look at the shares page what do the settings look like for the share you're copying to? Have you tried copying to a disks share?

The share in the attachment is the one I want to write to. What draws my attention now is that it looks the same as a cache drive that died on me and made the sytem unresponsive. Could it have something to do with a mover action gone bad?

The share in the attachment is the one I want to write to. What draws my attention now is that it looks the same as a cache drive that died on me and made the sytem unresponsive. Could it have something to do with a mover action gone bad?

 

You have a cache drive?  Is it full?  Your problem could be that the mover program is not running for some reason.

John,

 

I used to have a cache disk, it broke, do not know if the problems occured after that. It is in the trash now as I took it out and ran some tests on it but it did not even spin up anymore. It is in the eternal hunting fields now.

Do you know a way to tell the system that the move action was complete? Or should I try and take disk6 out as I am pretty sure unraid was writing to this drive due to all others being quite full?