unRAID Server Release 6.0-beta13-x86_64 Available


limetech

240 posts in this topic Last Reply

Recommended Posts

Unfortunately I've restarted (multiple times) / reverted back to beta 12 (Still didn't fix the issue) a

 

I am in the exact same position. I was wondering if downgrading to beta12 would get me working again. Thanks for trying it first and posting the results. I won't bother. I used the gui in b12 to format the cache drives to btrfs. I am very eager to get working again as my Dockers are on there as well as my KVMs.

 

thanks for any guidance

Link to post
  • Replies 239
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

Here is mine

 

btrfs filesystem show

 

warning devid 2 not found already

Label: none  uuid: d864df1d-d16c-4df8-9aa3-d6eee075de76

        Total devices 2 FS bytes used 158.83GiB

        devid    1 size 223.57GiB used 223.57GiB path /dev/sdb1

        *** Some devices missing

 

Btrfs v3.18.2

 

----

 

file -s /dev/sdb1

/dev/sdb1: BTRFS Filesystem sectorsize 4096, nodesize 16384, leafsize 16384)

 

file -s /dev/sdc1

/dev/sdc1: data

 

 

 

Link to post

I note that, since installing this update, the space used in my docker.img has dropped from 12.07GB to  7.04GB.  Is this expected or might it signify that there is a problem somewhere?

Label: none  uuid: 969cdbc6-06ca-42df-a8f5-60ee6bad29df
Total devices 1 FS bytes used 3.85GiB
devid    1 size 15.00GiB used 7.04GiB path /dev/loop0

Btrfs v3.18.2

 

And, unrelated, I'm sure that the system info displayed at the top-right of the web gui used to inform what licence type was in place (eg: Plus).  This info seems to have gone - is this deliberate?

Ooops - I have a terrible memory! ... but I was really happy to see that the system info had been made so much easier to read!

Link to post

And, unrelated, I'm sure that the system info displayed at the top-right of the web gui used to inform what licence type was in place (eg: Plus).  This info seems to have gone - is this deliberate?

On my system it displays this under the logo on the top left.  As far as I know this has been the case for the last few betas at least.

Link to post

And, unrelated, I'm sure that the system info displayed at the top-right of the web gui used to inform what licence type was in place (eg: Plus).  This info seems to have gone - is this deliberate?

On my system it displays this under the logo on the top left.  As far as I know this has been the case for the last few betas at least.

Hasn't changed ...

Link to post

how would i proceed to upgrade my second server ?

 

cache drive is partitioned with fdisk for sure

 

sdg1 is btrfs cache drive

sdg2 is swap file as btrfs doesn't support swap files

 

my first server i reformatted the drive as i couldn't wait any longer for a solution... most things work again but rescanning plex will take days .....  50.000 + tv episodes

Link to post

I have a BTRFS Cache pool with 2 SSD drives that's now coming back as unformatted when I start the array after updating to beta 13.

Unfortunately I've restarted (multiple times) / reverted back to beta 12 (Still didn't fix the issue) and then reinstalled beta 13 already, so I'm not sure how useful the syslog will be.  I don't believe I've run a system check, and I haven't reformatted or done a scrub yet.  The drives were setup through the GUI in a previous beta, possibly 9 or 10a.

 

file -s /dev/sdd1

/dev/sdd1: BTRFS Filesystem sectorsize 4096, nodesize 16384, leafsize 16384)

 

file -s /dev/sde1

/dev/sde1: data

 

As seen in the following screenshot the 2nd cache drive is showing a blue icon indicating a new drive, which isn't the case.

http://gyazo.com/50ab47397cabe74688ddd112b281c201

 

After starting the array:

http://gyazo.com/50c2910c498dc48056f6c93f035745a9

 

root@unRAID:/var/log# btrfs filesystem show

warning devid 2 not found already

Label: none  uuid: 4b26bea4-4d80-4dcf-a8ed-d839b0d999b0

Total devices 2 FS bytes used 56.04GiB

devid    1 size 232.89GiB used 232.89GiB path /dev/sdd1

*** Some devices missing

 

Btrfs v3.18.2

 

Attached syslog

 

Edit: I am recalling having the /dev/sde drive setup as just an ext3 (i think) partition that stored my Xen VMs and wasn't part of the array prior to getting the 2nd SSD and converting it to a btrfs cache pool through the UI.  That was several beta versions ago though.

 

What version did you upgrade from?  Was this from 12-13 or directly from 10 to 13?

Link to post

After upgrading to -beta13 if your cache disk now appears 'unformatted', please do not run file system check or scrub, or reformat.  Instead please capture the system log without rebooting and post here.  Probably what's happening is this: those who experience this issue have probably manually partitioned their cache disk, or perhaps partitioned and built file system on some other linux distro.  I need to know if this is the case before implementing a fix.  thx

I have not tried this release yet but I created a btrfs cache pool following these instructions posted by jonp. Those instructions specifically told us to manually partition and format the other disks for the btrfs cache pool.

 

Is this what is causing the problem?

Link to post

After upgrading to -beta13 if your cache disk now appears 'unformatted', please do not run file system check or scrub, or reformat.  Instead please capture the system log without rebooting and post here.  Probably what's happening is this: those who experience this issue have probably manually partitioned their cache disk, or perhaps partitioned and built file system on some other linux distro.  I need to know if this is the case before implementing a fix.  thx

I have not tried this release yet but I created a btrfs cache pool following these instructions posted by jonp. Those instructions specifically told us to manually partition and format the other disks for the btrfs cache pool.

 

Is this what is causing the problem?

 

This was how it was done before the webUI supported it directly.  This has nothing to do with the issue some folks are experiencing.  It's more related to folks that manually created multiple partitions on their cache device.  I can't say for certain that's 100% what it is, but the unformatted issue is not the result of how a btrfs raid group was created from beta 6.

Link to post

[sigh] Just so I understand, you're telling me that it is a "theory" to tell people, "if you experience any problem with drives showing as unformatted, DO NOT TR Y TO FIX IT, and send us a syslog"

I see what you're getting at. Will address when I get to the office.

 

Hey Jonp

 

I see you added this to the OP, but don't see any reference to the apcupsd piece (you need to remove from 12 before upgrading to 13). This should likely be noted as well to save people headaches.

Link to post

[sigh] Just so I understand, you're telling me that it is a "theory" to tell people, "if you experience any problem with drives showing as unformatted, DO NOT TR Y TO FIX IT, and send us a syslog"

I see what you're getting at. Will address when I get to the office.

 

Hey Jonp

 

I see you added this to the OP, but don't see any reference to the apcupsd piece (you need to remove from 12 before upgrading to 13). This should likely be noted as well to save people headaches.

 

Probably not the right place for it.  We do not officially support this feature just yet.  We are getting ready to, but anyone that's getting ahead of us and using community work here is really on their own.  This isn't an issue of us not caring, it's us not being able to keep track of each community work that gets affected through each release in our beta.  If we do it for APCUPSd, why wouldn't we have to do this for other plugins then?

Link to post

Well apcupsd is right now the only one for which you've added the package to the default install. That is what makes it "special" and why there is a reasonable argument to be made for a note saying, "if you don't listen to us about removing all your other plugins before upgrading, you should listen to us about this."

 

This isn't official support, this is noting a problem you KNOW people will have because, supported or not, people are using apcupsd and a change you made is now "known" to cause issues and you know how to prevent it. That is distinctly different than asking you to make a note about every possible unofficial plug-in that isn't essential to the safe operation of a unRaid ... something I don't think you can successfully argue about apcupsd.

 

But I could also see your argument that people should be removing them all, period, no quarter given :)

 

Just another thought for the little it is worth.

Link to post

Just a FYI if you're running Sonaar and updating it internally, if you're using the develop branch.  If you use the unraid internal mech to update the docker (to get rid of the red text on your docker page about the update status not available), it could possibly break your sonaar install.  Basically wiped my configuration back to the defaults, database and settings, everything.  I do have a backup, so I'm trying to get it back up and running now.

 

**EDIT**  Gotten much worse, the unraid UI is almost completely unresponsive.  I've had to shutdown from commandline twice now.  The UI came up once after the first shutdown, but you couldn't do anything, trying another shutdown now..

 

**EDIT 2** After the 3rd reboot, I was able to get the UI stable again, still haven't gotten Sonaar working.  For whatever reason the repo I was using won't auto-update anymore to the latest version, so my database version for the software is ahead of the program version, causing issues...  I've tried using the needo version from his repo, it'll update, but for whatever reason the database won't load correctly, and the docker shows its running, but is actually not running (ghost image).  Another reboot shortly hopefully will fix that issue.

 

 

Link to post

Updated to b13 via webui. Rebooted.

 

Now my cache drive shows up as unformatted (it's a 250gb ssd with xfs). It has all my xen vms on there so am a bit stuck.

 

Any thoughts?

 

Thanks.

 

**********************************************************

Just noticed this is the second similar report already - suggest holding off on applying this update until we see some kind of limetech response.....

***********************************************************

 

Need a system log please.

 

Here is my system log. I'm not sure if this will be any use as I've restarted a few times subsequently and also suffered a red ball disk in the process so I'm running a rebuild on that now.

 

I do have my VMs backed up but I'd much prefer get my cache restored rather than having to reformat and transfer everything from a state a few weeks ago.

 

I would add that the theory of the cache disk being formatted elsewhere is likely valid as I was using this SSD a few weeks ago to test booting into various linux OSes, only transferring it into unRAID and formatting it XFS via unRAID UI.

syslog.zip

Link to post

Well apcupsd is right now the only one for which you've added the package to the default install. That is what makes it "special" and why there is a reasonable argument to be made for a note saying, "if you don't listen to us about removing all your other plugins before upgrading, you should listen to us about this."

 

This isn't official support, this is noting a problem you KNOW people will have because, supported or not, people are using apcupsd and a change you made is now "known" to cause issues and you know how to prevent it. That is distinctly different than asking you to make a note about every possible unofficial plug-in that isn't essential to the safe operation of a unRaid ... something I don't think you can successfully argue about apcupsd.

 

But I could also see your argument that people should be removing them all, period, no quarter given :)

 

Just another thought for the little it is worth.

 

This makes perfect sense to me, no one is asking you to support the plugin, just give people a heads up of a know issue, to save them from a know issue during the upgrade. Show your customers you care by saving them some hassle whenever you can.

Link to post

I have not tried this release yet but I created a btrfs cache pool following these instructions posted by jonp. Those instructions specifically told us to manually partition and format the other disks for the btrfs cache pool.

 

Is this what is causing the problem?

 

Yes, that is one way to cause this problem.

 

We took down -beta13 and will have a -beta14 posted up asap that addresses this issue.

Link to post

Well apcupsd is right now the only one for which you've added the package to the default install. That is what makes it "special" and why there is a reasonable argument to be made for a note saying, "if you don't listen to us about removing all your other plugins before upgrading, you should listen to us about this."

 

This isn't official support, this is noting a problem you KNOW people will have because, supported or not, people are using apcupsd and a change you made is now "known" to cause issues and you know how to prevent it. That is distinctly different than asking you to make a note about every possible unofficial plug-in that isn't essential to the safe operation of a unRaid ... something I don't think you can successfully argue about apcupsd.

 

Absolutely. This is a very reasonable plan since all release notes provide a Known Issues section. This is clearly an item to be designated in the Known Issues section. If this doesn't fit there, then nothing will ever fit in that section.

 

Link to post

ooops...., this has never happened before..

 

iYHfAGD.jpg

 

 

Completely dead, locked up, can't telnet or ssh in.

 

no response over IPMI or at keyboard on server itself.

 

 

Please try this.  Move your flash device to a PC and edit the file 'syslinux/syslinux.cfg'.

Assuming you are booting non-xen, non-safe-mode, change:

 

default /syslinux/menu.c32
menu title Lime Technology
prompt 0
timeout 50
label unRAID OS
  menu default
  kernel /bzimage
  append initrd=/bzroot
label unRAID OS Safe Mode (no plugins)
  kernel /bzimage
  append initrd=/bzroot unraidsafemode
label Memtest86+
  kernel /memtest
label Xen/unRAID OS
   kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot
label Xen/unRAID OS Safe Mode (no plugins)
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot unraidsafemode

 

to

 

default /syslinux/menu.c32
menu title Lime Technology
prompt 0
timeout 50
label unRAID OS
  menu default
  kernel /bzimage
  append [b]intel_pstate=disable[/b] initrd=/bzroot
label unRAID OS Safe Mode (no plugins)
  kernel /bzimage
  append initrd=/bzroot unraidsafemode
label Memtest86+
  kernel /memtest
label Xen/unRAID OS
   kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot
label Xen/unRAID OS Safe Mode (no plugins)
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot unraidsafemode

 

(All you're doing is insert "intel_pstate=disable" on the first 'append' statement.)

 

There was a patch to the intel_pstate driver to fix frequency scaling issue with Haswell CPU's.  Could be your CPU doesn't like this patch, so above line disables the driver (it will fall back to default cpufreq scaler)  :P

 

What CPU are you running with?

Link to post

Well apcupsd is right now the only one for which you've added the package to the default install. That is what makes it "special" and why there is a reasonable argument to be made for a note saying, "if you don't listen to us about removing all your other plugins before upgrading, you should listen to us about this."

 

This isn't official support, this is noting a problem you KNOW people will have because, supported or not, people are using apcupsd and a change you made is now "known" to cause issues and you know how to prevent it. That is distinctly different than asking you to make a note about every possible unofficial plug-in that isn't essential to the safe operation of a unRaid ... something I don't think you can successfully argue about apcupsd.

 

But I could also see your argument that people should be removing them all, period, no quarter given :)

 

Just another thought for the little it is worth.

 

The intent was for apcupsd webGui plugin to be included in the build, but we removed because we haven't completed it's testing and we didn't want to delay this release even more.  Right we didn't explicitly test the public plugin, thinking it should have worked - all that should have happened was the plugin noticed the package was already installed, and just skip that and move on.  But I forgot we rebuilt apcupsd package to be more secure by making a slight change in /etc/apc*.conf to have it listen only on localhost instead of all interfaces.  So yeah, this one slipped through, but that's why there are betas right?  Pretty sure we're going to move to closed betas for 6.1.

Link to post

ooops...., this has never happened before..

 

iYHfAGD.jpg

 

 

Completely dead, locked up, can't telnet or ssh in.

 

no response over IPMI or at keyboard on server itself.

 

 

Please try this.  Move your flash device to a PC and edit the file 'syslinux/syslinux.cfg'.

Assuming you are booting non-xen, non-safe-mode, change:

 

default /syslinux/menu.c32
menu title Lime Technology
prompt 0
timeout 50
label unRAID OS
  menu default
  kernel /bzimage
  append initrd=/bzroot
label unRAID OS Safe Mode (no plugins)
  kernel /bzimage
  append initrd=/bzroot unraidsafemode
label Memtest86+
  kernel /memtest
label Xen/unRAID OS
   kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot
label Xen/unRAID OS Safe Mode (no plugins)
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot unraidsafemode

 

to

 

default /syslinux/menu.c32
menu title Lime Technology
prompt 0
timeout 50
label unRAID OS
  menu default
  kernel /bzimage
  append [b]intel_pstate=disable[/b] initrd=/bzroot
label unRAID OS Safe Mode (no plugins)
  kernel /bzimage
  append initrd=/bzroot unraidsafemode
label Memtest86+
  kernel /memtest
label Xen/unRAID OS
   kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot
label Xen/unRAID OS Safe Mode (no plugins)
  kernel /syslinux/mboot.c32
  append /xen --- /bzimage --- /bzroot unraidsafemode

 

(All you're doing is insert "intel_pstate=disable" on the first 'append' statement.)

 

There was a patch to the intel_pstate driver to fix frequency scaling issue with Haswell CPU's.  Could be your CPU doesn't like this patch, so above line disables the driver (it will fall back to default cpufreq scaler)  :P

 

What CPU are you running with?

 

 

Intel Xeon E3 1240V3 which is haswell.

Link to post

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.