unRAID Server Release 6.2.0-beta20 Available

[DMcK]

Hello,

The upgrade from 6.1.9 to 6.2 beta 20 was painless.  It works great except...

 

When using Kodi, I can no longer see subfolders using NFS.  I have a Videos share that has the usual Movies and TV shows (these work fine). But I also have my home videos in simple folder structure which is no longer accessible. So it looks like this:

 

Videos

-Movies (only Movies already in my database can be played. No new content can be seen/added.)

-TV Shows (only TV Shows already in my database can be played. No new content can be seen/added.)

-Home Videos (I can see this level and the files in it)

--2016 (can't see this anymore)

--2015 (can't see this anymore)

etc.

 

I've looked around for a change in NFS settings, but I can't see anything related.

 

diagnostics.zip attached.

 

Thanks,

Doug

 

Edit: Old content in my databases can be seen and played. New content can't be seen or added.

mediaserver-diagnostics-20160328-1039.zip

[RobJ]

Upgraded to 6.2.0-beta20 and array would not start due to four disks missing. I have a diagnostics file from before and after the upgrade.

 

My understanding from reading through the previous release posts is that this could be related to the marvel SATA controller that is integrated into this motherboard. Four drives are connected to the Marvel controllers. Prior to the upgrade all disks have functioned properly and this has never been a problem in the past. I checked my Bios firmware and I believe that I am on the most recent version for the GA-990FXA-UD5.

 

I have an LSI SATA SAS2008 RAID controller card plugged into the motherboard that has been flashed into IT mode that currently does not have any drives connected to it.

 

Should I power down and connect the drives currently attached to the Marvel SATA controllers to the SAS2008? (I have not tried this card - just flashed it recently and then installed it)

 

Or should I downgrade the software and then look at switching those drives over to the SAS2008 card?

 

Is there some way of upgrading the driver for the Marvel SATA controller? I would still like to be able to use those ports and they have always worked well in the past.

Really appreciate having both diagnostics, one from 6.1.9 and one from 6.2-beta20.  It so often makes it easier to see the issues in the differences.  Thank you!

 

Well, it's not the normal Marvell issue with virtualization, but it *IS* related.  You have a pair of Marvell 88SE9172 cards (on 03:00.0 and 09:00.0), 2 SATA ports each.  In BOTH versions, they are recognized by the kernel and set up, 4 SCSI channels are set up for the ports, 4 ATA channels are set up, and all 4 establish the lower level SATA link to their attached drive.  But in 6.2-beta20, the following events occur, and none of the 4 drives can respond to identity requests.

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a60440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a60450 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80450 flags=0x0070]

...and...

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9ae0440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9ae0450 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9b00440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9b00450 flags=0x0070]

There's a continuous stream of those IO_PAGE_FAULTs afterward, possibly harming overall performance.  The kernel continues trying to talk to the drives, but unsuccessfully.  A typical sequence -

Mar 28 08:41:02 Tower kernel: ata8: SATA link up 6.0 Gbps (SStatus 133 SControl 300)

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [

Mar 28 08:41:02 Tower kernel: ata8.00: failed to IDENTIFY (INIT_DEV_PARAMS failed, err_mask=0x80)

Mar 28 08:41:02 Tower kernel: IO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80420 flags=0x0070]

 

Perhaps Tom will see something else in the AMD virtualization setup.  All I can suggest is check for firmware updates for the Marvell cards, and check for a BIOS update for the motherboard.  And you could use the other disk controller card, but it would be nice to know what is wrong here because it is obviously so close to working!  It's not at all a surprise to see Marvell having virtualization issues unfortunately.

[bigjme]

Ok so i just had 2 crashes today, the last crash i was able to use the terminal in the unraid gui so tried to capture diagnostics and.... nothing

 

I left the gui log running which was still picking up data from the tail of syslog, this can be seen here:

18:23 onwards are the issues (glad i took a picture rather then rely on the log)

 

Now, in the terminal i tried the following:

• Running "diagnostic" which sat at gathering diagnostic information and did nothing
  
• Running the manual cp /var/... to copy the syslog to the usb, no file was created but the cp ran without saying any issues
  
• Running the powerdown module - again sat at gathering diagnostic information and did nothing further
  

 

It seems more and more like an issue with my USB since nothing is actually able to write to it in this state. Pressing a key on my vm keyboard was the thing that finally locked it up to being unusable

 

I had again, no ssh, telnet, web gui or anything else. The syslogs did not save so i wasn't able to get that either

 

I've attached a few diagnostics now but i have included another again for anyone needing it. Right now i was told beta19 should help, then beta 20 should help

 

I have removed my 750ti in case that was causing weird issues. I have disabled my onboard marvel sata controller in case that was causing issues, i have checked for bios updates and it is all fine

 

Anymore ideas?

archangel-diagnostics-20160328-1835.zip

[RobJ]

Ok so i just had 2 crashes today, the last crash i was able to use the terminal in the unraid gui so tried to capture diagnostics and.... nothing

 

I left the gui log running which was still picking up data from the tail of syslog, this can be seen here:

...[snipped]...

 

Now, in the terminal i tried the following:

• Running "diagnostic" which sat at gathering diagnostic information and did nothing
  
• Running the manual cp /var/... to copy the syslog to the usb, no file was created but the cp ran without saying any issues
  
• Running the powerdown module - again sat at gathering diagnostic information and did nothing further
  

 

It seems more and more like an issue with my USB since nothing is actually able to write to it in this state. Pressing a key on my vm keyboard was the thing that finally locked it up to being unusable

 

I had again, no ssh, telnet, web gui or anything else. The syslogs did not save so i wasn't able to get that either

 

I've attached a few diagnostics now but i have included another again for anyone needing it. Right now i was told beta19 should help, then beta 20 should help

 

I have removed my 750ti in case that was causing weird issues. I have disabled my onboard marvel sata controller in case that was causing issues, i have checked for bios updates and it is all fine

 

Please excuse what may be an annoying question, redundant, but I can't remember if I or another asked if you had exhaustively tested the RAM?  I'm thinking something like 24 hours of Memtest.

 

Although there have been reports of issues with certain USB 3.0 drivers, what I see above looks too low level to be USB related.  Looks more like RAM or timers or CPU or VM/driver race condition or the like.

 

Edit: It would be interesting to know if anyone else has the same motherboard and BIOS and CPU etc, and what issues they are having?

[jude]

So it turns out that there was an updated Bios available which I have now installed.

 

The result is still the same in terms of the 4 missing disks.

 

I don't know if I can (or how to) update the firmware for the Marvell controllers that are built into the motherboard.

 

New diagnostic attached

 

 

 

Upgraded to 6.2.0-beta20 and array would not start due to four disks missing. I have a diagnostics file from before and after the upgrade.

 

My understanding from reading through the previous release posts is that this could be related to the marvel SATA controller that is integrated into this motherboard. Four drives are connected to the Marvel controllers. Prior to the upgrade all disks have functioned properly and this has never been a problem in the past. I checked my Bios firmware and I believe that I am on the most recent version for the GA-990FXA-UD5.

 

I have an LSI SATA SAS2008 RAID controller card plugged into the motherboard that has been flashed into IT mode that currently does not have any drives connected to it.

 

Should I power down and connect the drives currently attached to the Marvel SATA controllers to the SAS2008? (I have not tried this card - just flashed it recently and then installed it)

 

Or should I downgrade the software and then look at switching those drives over to the SAS2008 card?

 

Is there some way of upgrading the driver for the Marvel SATA controller? I would still like to be able to use those ports and they have always worked well in the past.

Really appreciate having both diagnostics, one from 6.1.9 and one from 6.2-beta20.  It so often makes it easier to see the issues in the differences.  Thank you!

 

Well, it's not the normal Marvell issue with virtualization, but it *IS* related.  You have a pair of Marvell 88SE9172 cards (on 03:00.0 and 09:00.0), 2 SATA ports each.  In BOTH versions, they are recognized by the kernel and set up, 4 SCSI channels are set up for the ports, 4 ATA channels are set up, and all 4 establish the lower level SATA link to their attached drive.  But in 6.2-beta20, the following events occur, and none of the 4 drives can respond to identity requests.

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a60440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a60450 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80450 flags=0x0070]

...and...

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9ae0440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9ae0450 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9b00440 flags=0x0070]

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [iO_PAGE_FAULT device=09:00.1 domain=0x0000 address=0x00000000a9b00450 flags=0x0070]

There's a continuous stream of those IO_PAGE_FAULTs afterward, possibly harming overall performance.  The kernel continues trying to talk to the drives, but unsuccessfully.  A typical sequence -

Mar 28 08:41:02 Tower kernel: ata8: SATA link up 6.0 Gbps (SStatus 133 SControl 300)

Mar 28 08:41:02 Tower kernel: AMD-Vi: Event logged [

Mar 28 08:41:02 Tower kernel: ata8.00: failed to IDENTIFY (INIT_DEV_PARAMS failed, err_mask=0x80)

Mar 28 08:41:02 Tower kernel: IO_PAGE_FAULT device=03:00.1 domain=0x0000 address=0x00000000a9a80420 flags=0x0070]

 

Perhaps Tom will see something else in the AMD virtualization setup.  All I can suggest is check for firmware updates for the Marvell cards, and check for a BIOS update for the motherboard.  And you could use the other disk controller card, but it would be nice to know what is wrong here because it is obviously so close to working!  It's not at all a surprise to see Marvell having virtualization issues unfortunately.

tower-diagnostics-20160328-1432.zip

[saarg]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

The license check is only for the beta and RC releases and Trial keys as stated here.

So for the stable releases it's not included.

[bigjme]

Please excuse what may be an annoying question, redundant, but I can't remember if I or another asked if you had exhaustively tested the RAM?  I'm thinking something like 24 hours of Memtest.

 

Although there have been reports of issues with certain USB 3.0 drivers, what I see above looks too low level to be USB related.  Looks more like RAM or timers or CPU or VM/driver race condition or the like.

 

Edit: It would be interesting to know if anyone else has the same motherboard and BIOS and CPU etc, and what issues they are having?

 

Thanks for the reply Rob. I have done some 24/7 mem tests with no issues. I haven't done one recently though, worth another set of runs do we think?

 

Edit:

I have started a memtest so let's see what that does

[thestewman]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

 

[Frank1940]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

Try going to 'Settings' >>  'Network Settings' and change 'Obtain DNS server address automatically:' setting to "No".  The use the settings that your IPS recommends or use Google DNS servers at  8.8.8.8    and    8.8.4.4   

 

 

[captain134]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

You are using beta software. Did you really bother support with an issue about a beta? Just stick to retail unraid. I'm sure lime-tech would rather have no feedback from you about the beta then have to listen to your complaints about a function that has been explained several times.

[bennymundz]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

You are using beta software. Did you really bother support with an issue about a beta? Just stick to retail unraid. I'm sure lime-tech would rather have no feedback from you about the beta then have to listen to your complaints about a function that has been explained several times.

 

Do you honestly believe that the key check "function" is not a precursor for a future move into production. Honestly, LT say it's for only for BETA (maybe for this release) but don't kid yourself if you think it's not a dry run for a move for future production releases?

 

I have never had an issue before and have run many public release beta since early versions of 6.x and now magically there is a key check "feature" to help with tracking debugging issues in beta. I think it's completely relevant to bring up that it's problematic and impacts users. Given there is no real workaround from LT I think bringing up the issue again given the beta20 release obviously was attempting to provide some sort of resolution to this issue with the 45 second wait implemented.

[bennymundz]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

That is what i need to do, set a router up with the same IP as my PFsense box allow unraid to vaildate and load then plug my WAN connection back into my unraid box so PFsense can bring up my net connection properly.

[captain134]

 

- When necessary to query keyserver, poll up to 45 seconds for a connection.

 

 

Does this mean that when booting unraid, that the key check now allows 45seconds for the array to come online and VM's to start. IE virtualised pfsense before the keycheck disables the system for an invalid key ?

 

Yup the license check is still a piece of crap. Whatever was done in this release is even worse then the previous two. Previously in 18 and 19 i could reach the unraid web gui and would be told invalid license. Now in 20 the web gui won't load at all until i get a WAN connection, this means i have to manually go patch in a router to get internet access for the license check. It is as retarded as it sounds !!

 

I really don't know why more focus isn't placed on resolving this rather poorly thought out license validation and wasting effort on throw away inclusions like updating firefox in the console gui who no one asked for / wants / even needs.

 

+1

 

My system refuse to validate and its the first time i ever had any interaction with LT support and they have been zero help.

Waited most of the day for them to verify it will not validat and they the took the rest of the day to tell me I need a reliable internet connection.

 

How did you do this ?

"manually go patch in a router to get internet access for the license check"

 

You are using beta software. Did you really bother support with an issue about a beta? Just stick to retail unraid. I'm sure lime-tech would rather have no feedback from you about the beta then have to listen to your complaints about a function that has been explained several times.

 

Do you honestly believe that the key check "function" is not a precursor for a future move into production. Honestly, LT say it's for only for BETA (maybe for this release) but don't kid yourself if you think it's not a dry run for a move for future production releases?

 

 

 

Yes that is what I believe.

 

If it makes it into a retail release, then you may have a legitimate complaint but even at that point, you are not forced to use that update.

[bennymundz]

If it makes it into a retail release, then you may have a legitimate complaint but even at that point, you are not forced to use that update.

 

Are you going to continue trolling this tread for every comment that is not specifically related to a bug in the beta release ?

[RobJ]

So it turns out that there was an updated Bios available which I have now installed.

 

The result is still the same in terms of the 4 missing disks.

 

I don't know if I can (or how to) update the firmware for the Marvell controllers that are built into the motherboard.

For onboard controllers, the firmwares are included in the BIOS, so you've done all you can.  As you found, syslog too shows identical behavior, no improvement at all.  Sorry.

[RobJ]

I realize some here don't have a long experience with LimeTech, but it still saddens me to see the tone of some posts, as well as the tone after the release of 6.2-beta18, when the licensing change was first revealed.  I don't like unpleasantness of any kind, so I'll just make a statement and leave it there.

 

Those of us who have been around LimeTech for a long time have NEVER seen Tom or any other LimeTech associated people do *anything* with a greedy or user-hostile intent.  In fact, I (and I'm sure many others) can't count how many times we have seen Tom bend over backwards to help users, and if a software or policy change resulted in complaints, he has *always* (as far as I can remember) adjusted in favor of users.  So it really hurts to see users discover a change and immediately assume the worst, when there is NO previous evidence that Tom would do something like that.  Certainly at times, he has had to make policy changes to ensure the future of unRAID, but he has always been fair in every way, usually given fair warning, and always been open to frank discussion.

[BRiT]

@RobJ, There's always a first.

 

Although I take different meaning from their posts. I merely read it as consumers being very passionate about unRAID. I don't read any hostility in it, just a lot of passion that consumers are having trouble in expressing in the prim and proper and well to do way. Sometimes when something is an obviously stupid scheme, regardless of the intent of it, it needs to be called out for what it is, calling a spade a spade.

 

With all that said... I do agree that the tone needs to be taken down a notch to become civil and respectful. You know something is up when I'm calling for people to be civil, regardless of however much entertainment value it provides. 

[jude]

So it turns out that there was an updated Bios available which I have now installed.

 

The result is still the same in terms of the 4 missing disks.

 

I don't know if I can (or how to) update the firmware for the Marvell controllers that are built into the motherboard.

For onboard controllers, the firmwares are included in the BIOS, so you've done all you can.  As you found, syslog too shows identical behavior, no improvement at all.  Sorry.

 

No worries, I switched all four disks over from the onboard Marvell SATA controller to the SAS2008 PCIe controller. The array came online nicely after a reboot.

 

Thanks for your help.

[jude]

Docker updates worked perfectly but now I am having some issues getting my Win10 VM updated.

 

I went into the VM tab and attempted to "edit" the VM (Win10ProIsolCPUs) so that it would update to the new settings. I may have done something wrong during that process because now the VM wont start and their are errors displayed on the VM tab. See screen capture.

 

I have copies of my old XML file and copies of the Win10 disk image. Should I attempt to fix this current "edit" or would it be better to use a template and import the existing Win10 disk image and then make changes to the generated XML if needed?

 

edit

 

I went into terminal and Virsh to see if I could start the VM from there. It reported that the VM started and when I turned on my TV the VM was being passed through Audio and Video as well as USB controller and attached devices. See screen shots.

 

The VM tab is still showing errors and the dashboard is no longer showing my working Dockers or VM's

 

Diagnostics attached

[jbartlett]

@RobJ, There's always a first.

 

That is a lousy thing to say. There isn't always a first. People go their entire lives without doing some things, so saying that there is always a first is unfair and setting up the expectation that people will fail and will fail their customers.

[BRiT]

@RobJ, There's always a first.

 

That is a lousy thing to say.

 

It's as perfectly fine a thing to say as is posting about the history. History does not always dictate the future. That's all I was conveying.

[jude]

Docker updates worked perfectly but now I am having some issues getting my Win10 VM updated.

 

I went into the VM tab and attempted to "edit" the VM (Win10ProIsolCPUs) so that it would update to the new settings. I may have done something wrong during that process because now the VM wont start and their are errors displayed on the VM tab. See screen capture.

 

I have copies of my old XML file and copies of the Win10 disk image. Should I attempt to fix this current "edit" or would it be better to use a template and import the existing Win10 disk image and then make changes to the generated XML if needed?

 

edit

 

I went into terminal and Virsh to see if I could start the VM from there. It reported that the VM started and when I turned on my TV the VM was being passed through Audio and Video as well as USB controller and attached devices. See screen shots.

 

The VM tab is still showing errors and the dashboard is no longer showing my working Dockers or VM's

 

Diagnostics attached

 

Couldn't add the diagnostics file to the previous post so here it is

tower-diagnostics-20160328-2110.zip

[BobPhoenix]

Couldn't add the diagnostics file to the previous post so here it is

Here is how to add a file to an existing post:

[*]Upper right hand corner click on modify

[*]Type in any changes to the text that you need if any

[*]Click on the plus graphic next to "Attachments and other options"

[*]Where it says Attach: click on "Choose File" button and navigate to your file and select it

[*]When you see the file name of the file in the box click on the "Save" button on bottom right to save all change and file to post.

If you tried this and it didn't work not sure why as it has always worked for me as long as you have not added a file that is too big itself or too big when combined with other files already attached to the post.

[jude]

Couldn't add the diagnostics file to the previous post so here it is

Here is how to add a file to an existing post:

[*]Upper right hand corner click on modify

[*]Type in any changes to the text that you need if any

[*]Click on the plus graphic next to "Attachments and other options"

[*]Where it says Attach: click on "Choose File" button and navigate to your file and select it

[*]When you see the file name of the file in the box click on the "Save" button on bottom right to save all change and file to post.

If you tried this and it didn't work not sure why as it has always worked for me as long as you have not added a file that is too big itself or too big when combined with other files already attached to the post.

 

I don't think that it liked the combined size of the attachments, funny cause they are all small files 

[BobPhoenix]

I don't think that it liked the combined size of the attachments, funny cause they are all small files 

That would be my guess then.  Just wanted to make sure you or anyone else for that matter knew how to attach a file to an existing post - mostly the same as new post anyway.