unRAID Server Release 6.0-beta14b-x86_64 Available

[jonp]

I successfully used the gui to update from 14a to 14b, but it wasn't easy...

 

After applying the update I wasn't able stop Docker - when I changed "enable docker" to "no" and hit apply, nothing happened.

 

So I used the gui to stop the array, hoping it would stop the dockers on its way down, but the gui just hung while trying to unmount the drives.

 

So I SSH'd to the server and ran "docker ps" to list running dockers and then "docker stop [dockername]" to shut them off.

 

Then I ran "powerdown -r" to reboot.

 

It came back up and appears to be fine.

 

 

This experience makes me wonder... since an OS update requires an immediate reboot, should we require the array to be offline before the OS update can be applied?

The update doesn't do anything to the running state of the system.  The reboot is required to load the new bzroot / bzimage file that was downloaded to your flash device. You could download an update to the OS and wait weeks before rebooting. You'd be fine.

 

Correction, this applies to the server OS, but not the webgui.

 

[sacretagent]

Can Limetech please explain me if i can upgrade my server where i used fdisk to partition 1 big part of my cache disk in btrfs and 1 very  small part in raw to be used for swap file ?

i spend 5 days on my other server to retrieve my complete tv database in plex...

this server had 6000 movies in plex... which i would hate to rescan all

 

so how do i go about it ?

[limetech]

Can Limetech please explain me if i can upgrade my server where i used fdisk to partition 1 big part of my cache disk in btrfs and 1 very  small part in raw to be used for swap file ?

i spend 5 days on my other server to retrieve my complete tv database in plex...

this server had 6000 movies in plex... which i would hate to rescan all

 

so how do i go about it ?

 

With -beta14 or higher we don't mess with the partition structure of a single cache disk if there already exists a mountable partition 1, so I think you should be good.  To be safe you could use 'dd' command to save your current MBR record to a file:

 

dd if=/dev/sdX of=/boot/mbr.backup count=1

 

(Replace 'sdX' with device name of your specific cache device.)

 

After you upgrade and start array, if your cache device comes up 'unmountable' then do not click Format - instead send me email: [email protected].  But this shouldn't happen

[Fireball3]

With an RC on the horizon, I want to set up my backup server on v6.

Is there an official limetech recommendation with respect to the filesystem that should be used?

Thanks

[limetech]

With an RC on the horizon, I want to set up my backup server on v6.

Is there an official limetech recommendation with respect to the filesystem that should be used?

Thanks

 

For the array: XFS

For the cache: btrfs

 

This is also how the defaults are set up for unRaid-6.

[chickensoup]

With an RC on the horizon, I want to set up my backup server on v6.

Is there an official limetech recommendation with respect to the filesystem that should be used?

Thanks

 

For the array: XFS

For the cache: btrfs

 

This is also how the defaults are set up for unRaid-6.

 

Do you still recommend btrfs for the cache drive if you are not cache pooling? Due to some of the ongoing btrfs related issues I've noticed on the forum I was planning on using xfs for my cache drive when I upgrade from v5.

[dalben]

Do you still recommend btrfs for the cache drive if you are not cache pooling? Due to some of the ongoing btrfs related issues I've noticed on the forum I was planning on using xfs for my cache drive when I upgrade from v5.

 

I was going to ask the same us the undercurrent in the forum seemed to suggest XFS for the cahe drive if there is not intent tio cache pool.

[PeterB]

Have I forgotten to set something up?  With the apcupsd plugin I was getting email notifications of power events.  I haven't had any such notification since February 21 - the day I installed b14.

 

Using the smtp test in notifications settings I still receive an email.

[dlandon]

Do you get notifications on the webgui in the upper right corner of the screen?  Are the notifications listed on the Archived Notifications page?

 

I believe you need to specify that you receive emails on the notification web page for events.  You might just apply the settings again.  I believe I read where the notifications and parity schedule need to be re-applied.

[BobPhoenix]

• Device spinning status is now maintained internally and the actual devices are no longer interrogated to get their spinning state.  This means that if you manually use

  hdparm -y

  to spin down devices, or use a 3rd party plugin that does so, the webGui will likely get out of sync with the actual device state.  So don't do that.
  

You should be able to get them back into sync by clicking the SpinUp or SpinDown button on the main page though - right?

[PeterB]

Do you get notifications on the webgui in the upper right corner of the screen?  Are the notifications listed on the Archived Notifications page?

 

The only notifications I have seen are the notices telling me that the array is up with seven drives.  I have emails turned off for notices, so don't receive emails for that 'array up' event.

 

Do archived notifications survive a reboot?  Again, the only archived events I have seen are 'array up' events.

 

I believe you need to specify that you receive emails on the notification web page for events.  You might just apply the settings again.  I believe I read where the notifications and parity schedule need to be re-applied.

 

Yes, I recall reading that, so I had already ensured that the settings are according to my preference and the 'Apply' button clicked.

[dlandon]

Let me confirm on my test server that indeed notifications are emailed.

[PeterB]

Let me confirm on my test server that indeed notifications are emailed.

 

It's not just the email - I don't get the red box at top right of the web gui to notify of the power failure.  Sorry, I don't think that I had made that clear! 

[dlandon]

I have received the notifications when the communications is lost.

[Squid]

Actually, all releases pre-dynamix, used 'hdparm' to only get spinning state in background.  Temperatures and other SMART data was only read "on-demand" as a result user navigating webGui.  In analyzing where time is spent in each page refresh, this operation accounted for most time, especially on larger arrays.  Go back and boot unRaid-5 and you'll see the page navigation is not as "snappy" - this is entirely due to reading SMART.

Me bad - I mixed up the old poll_spindown setting and the new poll_attributes settings.

[dlandon]

Do archived notifications survive a reboot?  Again, the only archived events I have seen are 'array up' events.

 

No.  They are lost on reboot.

[WeeboTech]

Do archived notifications survive a reboot?  Again, the only archived events I have seen are 'array up' events.

 

No.  They are lost on reboot.

 

 

I think if you set the notification directory to some persistent device, they will continue to exist.

[jonp]

Do you still recommend btrfs for the cache drive if you are not cache pooling? Due to some of the ongoing btrfs related issues I've noticed on the forum I was planning on using xfs for my cache drive when I upgrade from v5.

 

I was going to ask the same us the undercurrent in the forum seemed to suggest XFS for the cahe drive if there is not intent tio cache pool.

If you never ever ever plan on pooling cache devices, you can choose between xfs and btrfs.  The issues some folks have experienced with btrfs I have not. I have multiple test systems that use a mix of filesystems on the array and all use btrfs for the cache.

 

Never had a system crash with btrfs, never lost data, and never had corruption.  Some others have reported otherwise with their setups, but some were in older versions and some may have had hardware related issues.  Btrfs is the most robust filesystem type we've included. It natively supports features like copy on write, which then adds checksum support.

[dlandon]

Let me confirm on my test server that indeed notifications are emailed.

 

It's not just the email - I don't get the red box at top right of the web gui to notify of the power failure.  Sorry, I don't think that I had made that clear! 

 

I believe the server is shutting down before the notifications are posted and the email sent.  Just as a quick test, unplug your UPS communications cable and see if you get the notification and email.  It may take a bit of time to get the email.  Give it a few minutes.

[mr-hexen]

Poll attributes on spinup, clear temperature on spindown.  I've adjusted my settings down to 300 (5 minutes)

 

What if device spins up due to I/O?  Maybe that's not so bad because it would just add a bit more delay after spinup... lemme think about that.

 

Something useful would be to see what impact, if any, a "high" SMART polling rate has on media streams.  Different players may have different results.

 

Please do not append SMART polling after I/O induced spin up. My popcorn hour already times out sometimes due to this, requiring a full-reboot. Delaying the data stream any longer might cause other media players to do the same.

 

Instead, trigger a SMART poll xx seconds AFTER I/O starts. most media players should have buffered the data by then, no?

[PeterB]

I believe the server is shutting down before the notifications are posted and the email sent.  Just as a quick test, unplug your UPS communications cable and see if you get the notification and email.  It may take a bit of time to get the email.  Give it a few minutes.

 

I have already tried that - I waited until the 'Power failure on UPS Tower. Running on batteries.' message appears on the ssh session on my desktop machine (slaved from apcupsd running on Tower).  There is no 'red box' displayed, and no email received.

[dlandon]

I believe the server is shutting down before the notifications are posted and the email sent.  Just as a quick test, unplug your UPS communications cable and see if you get the notification and email.  It may take a bit of time to get the email.  Give it a few minutes.

 

I have already tried that - I waited until the 'Power failure on UPS Tower. Running on batteries.' message appears on the ssh session on my desktop machine (slaved from apcupsd running on Tower).  There is no 'red box' displayed, and no email received.

 

Ok.  I found the problem.  The apcupsd.notify script file that sends the notifications is not set to executable.

 

I'll PM Jonp and have LT change that.

[Woodpusherghd]

• We are still investigating cpu scaling issue with Haswell cpu's, and have removed a 'patch' that seems to have been responsible for some users reporting kernel crashes.

 

This may have no relation to the above, but I have a  2.6 GHz AMD Athlon II X4 CPU in a  ASUS M4N68T-M-V2 motherboard. Dashboard shows my CPU always at 2600 MHz. CPU utilization flucuates as expected. I checked the bios and "Cool and Quiet "is enabled.    Any suggestions?

[GHunter]

beta14b install we well and seems to be working good. My spin down issues have been fixed in this version too. Thanks. I know this problem was a tough one to find.

 

Gary

[archedraft]

Just updated to 14b and I have noticed that every time I restart unRAID I get a notification that one of my hard-drives has 1 relocated sector... IMO I don't need to be notified every time I reboot about the relocated sector. I personally only would like to know if that 1 relocated sector turns to 2... Is there a way for unRAID to keep track of the smart values and only let me know if the count increases?