So thanks to this i finally managed to start building my 2 gaming vm's setup with the current hardware.

Asus Crosshair VI Hero with latest bios
32gb 3200mhz Corsair

2 GTX 1060.

So far it's working great!

On 10/30/2017 at 4:57 PM, zarroc24 said:

Would it be possible to patch the kernel to allow the Elgato HD60 pro capture card to properly work when being passed to a windows VM. Here is the patch needed https://gist.github.com/numinit/40fc5ad96fd0990b0a63/ , to provide full support for the card, otherwise we need to rebuild the kernel ourselves and at the moment there are no instructions on how to build the latest stable release. 

This would be greatly appreciate and i believe a lot of users would be grateful about the added support. 

 

Hi Zarroc,

 

Not really sure what you're requesting here.  If this capture card is a PCI based device, nothing should need to be done to the kernel specific to the device to enable pass through.  We do not need to load device-specific drivers when we're doing pass through as the underlying guest VM will supply that for us.  If this is a USB-based device and you're attempting to use USB assignment and that's not working, you might want to toggle the USB mode (edit your VM and turn on advanced view and you can toggle between USB 3 and USB 2 emulation).

24 minutes ago, jonp said:

 

Hi Zarroc,

 

Not really sure what you're requesting here.  If this capture card is a PCI based device, nothing should need to be done to the kernel specific to the device to enable pass through.  We do not need to load device-specific drivers when we're doing pass through as the underlying guest VM will supply that for us.  If this is a USB-based device and you're attempting to use USB assignment and that's not working, you might want to toggle the USB mode (edit your VM and turn on advanced view and you can toggle between USB 3 and USB 2 emulation).

 

From the gist it looks like it's for fixing issues when reseting the card. Probably something along how some AMD cards needs a fix. 

I am able to pass the card to a VM but only the audio part works, from what i have been able to find the patch i included fixes to enabled both quirk_no_pm_reset and quirk_no_bus_reset for this device. This will allow the device to be seen on the VM.

On 3/12/2016 at 9:18 PM, tinimun said:

Progress. Enabled both quirk_no_pm_reset and quirk_no_bus_reset for this device, and it enumerates in the Windows guest.

 

Here is a very high-resolution picture of the board after removing the cover. The actual SoC used (chip on the right) is the Mozart 395S combo PCIe controller/H.264 encoder. Here is the specsheet.

 

http://i.imgur.com/JMdBp2h.jpg

 

Finally, a kernel patch: https://gist.github.com/numinit/40fc5ad96fd0990b0a63/

 

I will keep this thread updated. Somehow managed to screw up my Windows user account during this process.

 

Just joined to post about this: I have been using unraid for the last couple of months and did a build based around a Ryzen 1700x and an Asus X370 pro mobo. I have not been using any virtualization at all since I have hard there has been a lot of stability issues with that. Up until updating to the rc10b build I have been running a very stable system on the rc9b release and when I saw there were ryzen fixes on the new update I jumped on it two days ago....

 

Since updating to rc10b my system has needed a hard reset twice and does not stay up and running for more than 12 hours at a time at most. I went through this thread and either didn't see anyone else really experiencing the same things as me. is there a way to roll back to the previous build or is there something I can do to add a bit of stability to this? I am excited to use VMs but I don't mind rolling back and waiting if I cant keep my system stable for even one full day  .... or maybe I am just too new to unraid an missing something! Thanks in advance for the help, and if there is anything I can provide log wise please let me know!

37 minutes ago, gcalaway said:

Thanks in advance for the help, and if there is anything I can provide log wise please let me know!

 

See this thread:

 

 

I see that unraid has samba 4.6.8, any chance to get that upgraded to 4.6.9 or 4.7.1?

 

# sudo smbstatus -V
Version 4.6.8

 

Thanks @johnnie.black for the link, had not found that. I have added the line to my Syslinux Configuration and we will see how it goes!

Well, its been up for 3 days and everything seems fine after I made the edit Johnnie had me do. I'm sitting watching Plex and it dies. I go to the Control Panel. "Array Undefined"

 

 

tower-diagnostics-20171104-2232.zip

Edited November 5, 20178 yr by RonUSMC

On 10/30/2017 at 7:44 AM, bonienl said:

You can change to time to wait (default 90 sec). See Settings -> Disk Settings -> Shutdown time-out

 

I know that, but are you sure that it will help? My impression is that it will just extend the time unRAID will wait for processes to gracefully terminate. But the array will not stop as long as a ssh session is in a directory below /mnt and the ssh session will not gracefully close either. Only once the grace period is over, the processes will get killed, again resulting in an unclean array shut down, independently of how long you set the time to wait.

5 hours ago, tstor said:

But the array will not stop as long as a ssh session is in a directory below /mnt and the ssh session will not gracefully close either.

 

Yeah this is known behavior. Currently the only way to circumvent this, is either closing the ssh session yourself or move to a directory outside /mnt.

 

42 minutes ago, bonienl said:

 

Yeah this is known behavior. Currently the only way to circumvent this, is either closing the ssh session yourself or move to a directory outside /mnt.

 

 

Over my years using unRAID, this must attribute to 99% of the reasons why my array won't stop. I remember many times in my early years get frustrated and hard rebooting the server cause I was stuck at this same issue. I just didn't know any better.

 

I wonder if this is that common among others.  If that's the case, it makes me thing that a setting, enabled by default, that kills all ssh processes when 'stop array' is pressed would help more than it would hurt. Or perhaps a better middle ground is simple to 'ps' at time of shutdown and report to the GUI that there is a ssh terminal open and to please close it.

It would be nice, when requesting to stop array, that unRaid would bring up a summary panel of things impacting stopping the array. I easy spend 3-5 minutes tracking down ssh sessions, screen sessions, and open files on my rare stop array occasions. If there was a way unRaid could offer to kill these processes (at user request to avoids killing something that could cause data loss), like the powerdown feature, that would be very convenient!

i have a HP N40L and the CPU load on the dashboard is not working (empty, showing no progress) with 6.4.-rcX
on 6.3.5 it was fine. 

 

maybe because of this? 

"webGui: utilize pubsub/cpuload nchan websocket endpoint to update cpu load info"

52 minutes ago, phenomeus said:

i have a HP N40L and the CPU load on the dashboard is not working (empty, showing no progress) with 6.4.-rcX
on 6.3.5 it was fine. 

 

maybe because of this? 

"webGui: utilize pubsub/cpuload nchan websocket endpoint to update cpu load info"

 

Known issue with websockets and Safari. Workaround: use a different browser, e.g. Chrome.

3 hours ago, SSD said:

It would be nice, when requesting to stop array, that unRaid would bring up a summary panel of things impacting stopping the array. I easy spend 3-5 minutes tracking down ssh sessions, screen sessions, and open files on my rare stop array occasions. If there was a way unRaid could offer to kill these processes (at user request to avoids killing something that could cause data loss), like the powerdown feature, that would be very convenient!

I am considering adding a feature to the Tips and Tweaks plugin that would let you choose to shutdown ssh and bash sessions.  I'm thinking it would be best to let the user specify the processes to stop.  i.e. enter the processes to stop.

 

Shut down processes on array stop:   bash,ssh

 

An event would then be created to stop the processes whenever the array is stopped so the disk drives will unmount.  Before I undertake this endeavor, I would appreciate some feedback.

Just a minor little thing.

 

Daylight savings time ended at 2am today.  Neither of my servers (6.3.5 and 6.4) picked up the change.  I use pool.ntp.org, and I know that it works, as I could change my time zone from EST to any other and it would change the time appropriately.  However when changing it back to EST, the time was still out by 1 hour.

2 minutes ago, Squid said:

Just a minor little thing.

 

Daylight savings time ended at 2am today.  Neither of my servers (6.3.5 and 6.4) picked up the change.  I use pool.ntp.org, and I know that it works, as I could change my time zone from EST to any other and it would change the time appropriately.  However when changing it back to EST, the time was still out by 1 hour.

Works for me on both my servers on EST time.

5 minutes ago, dlandon said:

Works for me on both my servers on EST time.

 

Strange.  Must be gremlins

 

Edited November 5, 20178 yr by Squid

Plugin page not allowing to check for updates and not being able to shutdown cleanly is still present in this build. I checked my flash drive it had no errors and I changed the port it was using. I dunno what changed in about rc3/rc4 but it broke these things. I submitted a bug report and hopefully it will get looked at before it goes GA as it would suck for a lot of people to have this problem.

 

This is literally always present right before things to go hell. I am guessing it has something to do with it.

Nov 3 21:38:10 Tower kernel: traps: emhttpd[11870] trap divide error ip:419be9 sp:1491121d3e00 error:0 in emhttpd[400000+25000]

 

Edit:

I've decided to just downgrade back to 6.3.5 since I'm tired of dealing with it. 

Edited November 5, 20178 yr by Jerky_san

2 hours ago, dlandon said:

Works for me on both my servers on EST time.

 

Mine also adjusted.

1 hour ago, SSD said:

 

Mine also adjusted.

 

As did mine.

 

# date
Sun Nov  5 13:15:35 EST 2017
 

Fine... My system is possessed  

 

EDIT: Although @BRiT  If you are truly in EST, then the time of your posting is 2:15 pm, but your date shows 1:15 pm

Edited November 5, 20178 yr by Squid

12 minutes ago, Squid said:

Fine... My system is possessed  

 

EDIT: Although @BRiT  If you are truly in EST, then the time of your posting is 2:15 pm, but your date shows 1:15 pm

Uhm, no. We switched to EST, before we were in EDT (Eastern Daylight Time). Now Daylight Time is off. 

 

# date
Sun Nov  5 13:41:29 EST 2017
 

 

---

Eastern Daylight Time, or EDT, starts on the second Sunday in March at 2:00 AM when the time moves an hour ahead. The time changes back to Eastern Standard Time on the first Sunday in November.

Edited November 5, 20178 yr by BRiT

Never how I've thought of it.  (Nor is it how Windows as an example treats Eastern Standard Time)