unRAID OS version 6.4.0-rc7a available

[r4id 3]

3 minutes ago, Squid said:

But have you actually run it?

Sent from my LG-D852 using Tapatalk
 

test now.

[Interstellar 1]

Running this since release now, no issues as far as I can tell.

 

AMD GPU/NIC Passthrough, dockers, etc all seem to be working A-OK!

[r4id 3]

10 hours ago, Squid said:

But have you actually run it?

Sent from my LG-D852 using Tapatalk
 

Memtest is ok.

Edited August 5, 2017 by r4id

[JorgeB 3491]

On 02/08/2017 at 1:59 PM, johnnie.black said:

Just had a minor issue during disk replacement, "disk2" is not aligned, same with other browser and other themes, strange thing is I replaced disk1 yesterday and pretty sure it looked normal, will replace disk3 tomorrow, so I'll check if it happens again.

 

Still happening, but only with these Toshiba disks, so I'm guessing it as something to do with their identifier length.

[zoggy 19]

just tossing this up on the radar: ryzen crashes on linux 4.12 under high load -- https://www.phoronix.com/scan.php?page=news_item&px=Ryzen-Test-Stress-Run

Edited August 5, 2017 by zoggy

[bonienl 1101]

On 7/30/2017 at 0:04 AM, saarg said:

 

I can confirm there is a bug that makes it impossible to choose cores if you have 8 or less cores when creating a new VM.

If there are more than 8 cores available the first 8 are grayed out until you choose a core above core 8. Then all are available for selection. 

 

Created a new VM on my test machine which has 4 cores. I can choose which ever core I want.

 

What exactly do you observe, maybe some screenshots may help?

 

[bonienl 1101]

20 hours ago, johnnie.black said:

 

Still happening, but only with these Toshiba disks, so I'm guessing it as something to do with their identifier length.

 

Corrected in next version.

 

[saarg 430]

4 hours ago, bonienl said:

 

Created a new VM on my test machine which has 4 cores. I can choose which ever core I want.

 

What exactly do you observe, maybe some screenshots may help?

 

 

Below you'll see that when I create a new VM on a computer with only 4 cores, I can't choose any cores or deselect core 0.

On my 8 core Xeon I only get half the cores to choose from until I select one of them, then I get the full 16.

Both me and @CHBMB see this. I used chrome and I think he used both chrome and firefox. cache cleared.

[eschultz 95]

On 8/4/2017 at 10:00 AM, r4id said:

 

Yes, my server having freezing issues, when no load. 

 

unRAID OS version 6.4.0-rc7a

M/B: ASRock - X370 Gaming K4

CPU: AMD Ryzen 7 1700X Eight-Core @ 3400

HVM: Enabled

IOMMU: Enabled

Cache: 768 kB, 4096 kB, 16384 kB

Memory: 64 GB (max. installable capacity 64 GB)

 

Same M/B and CPU here, with rc7a, it was idle and completely stable for at least a week (was rebooted by us after that for other testing).

 

There are BIOS updates (AGESA 1.0.0.6a) for both the ASRock and Gigabyte M/Bs you've tried, you may want to apply those if you haven't already.

 

It might also be helpful if you posted your diagnostics zip file (Tools --> Diagnostics from the web ui) here on the forums so we can take a look to see if there's something else that stands out.

[willbendr 0]

13 hours ago, saarg said:

 

Below you'll see that when I create a new VM on a computer with only 4 cores, I can't choose any cores or deselect core 0.

On my 8 core Xeon I only get half the cores to choose from until I select one of them, then I get the full 16.

Both me and @CHBMB see this. I used chrome and I think he used both chrome and firefox. cache cleared.

 

I have the same results on my 8-core Ryzen whether in Chrome/Firefox/IE on another device or using Firefox on the rc7a server (all cache cleared, inside or outside of safe mode). Half of the 16 threads are grayed out when creating a new VM until I select one of the available choices in addition to cpu 0.

[de0n 0]

 

I have the same results on my 8-core Ryzen whether in Chrome/Firefox/IE on another device or using Firefox on the rc7a server (all cache cleared, inside or outside of safe mode). Half of the 16 threads are grayed out when creating a new VM until I select one of the available choices in addition to cpu 0.

I am also having this issue on my intel dh77eb with an i5-3330 cpu. All 4 cores are greyed out. Only additional thing i have is a usb controller which shows up under "other pci devices" and my ups under "usb devices". I have tried using chrome and ie. Cache was cleared on ie.

 

Sent from my LG-H870 using Tapatalk

 

 

 

[Darksurf 9]

I've found something interesting. I had my parity drive fail on me. it was getting some read errors every week so I tried moving it to another slot, issues persisted, then the drive failed and was gone. since the drive is no longer registering my webUI is doing something strange.

 

Lucky for me (or is it?), the drive is still under warranty. I'll be sending it back for RMA later.

cold-rewind-diagnostics-20170807-1709.zip

[bonienl 1101]

On 8/6/2017 at 4:26 PM, saarg said:

 

Below you'll see that when I create a new VM on a computer with only 4 cores, I can't choose any cores or deselect core 0.

On my 8 core Xeon I only get half the cores to choose from until I select one of them, then I get the full 16.

Both me and @CHBMB see this. I used chrome and I think he used both chrome and firefox. cache cleared.

 

Thanks for the details, cpu selection is fixed in next release.

[bonienl 1101]

8 hours ago, Darksurf said:

...since the drive is no longer registering my webUI is doing something strange

 

Thanks for reporting, fixed in next release.

 

[clowrym 25]

Woke up this Morning to one of my docker's shut down, on trying to restart it, it returned execution error, no such container. Rebooted, removed & reinstalled, but now have no option to add an IP address unless I( turn Bonding off (Which has always been on)

tower-diagnostics-20170808-0934.zip

[zoggy 19]

firefox 55 is out, fixes all the checkbox visibility issues in unraid ui

Edited August 8, 2017 by zoggy

[Darksurf 9]

On 8/5/2017 at 4:04 PM, zoggy said:

just tossing this up on the radar: ryzen crashes on linux 4.12 under high load -- https://www.phoronix.com/scan.php?page=news_item&px=Ryzen-Test-Stress-Run

AMD just announced they have found the root cause and fix.

https://hothardware.com/news/amd-confirms-rare-ryzen-smt-bug-and-fix

 

[r4id 3]

On 6.08.2017 at 6:33 PM, eschultz said:

 

Same M/B and CPU here, with rc7a, it was idle and completely stable for at least a week (was rebooted by us after that for other testing).

 

There are BIOS updates (AGESA 1.0.0.6a) for both the ASRock and Gigabyte M/Bs you've tried, you may want to apply those if you haven't already.

 

It might also be helpful if you posted your diagnostics zip file (Tools --> Diagnostics from the web ui) here on the forums so we can take a look to see if there's something else that stands out.

 

 

I removed the usb rapoo (wireless keyboard) and the system has been running stable for 2 days

[methanoid 18]

AMD just announced they have found the root cause and fix.
https://hothardware.com/news/amd-confirms-rare-ryzen-smt-bug-and-fix
 

Except they do not really say what the fix is, just that they will work with people. Not very convincing. And no progress on NPT issue for Ryzen :-(


Sent from my iPhone using Tapatalk

[Stu 0]

Hi,

 

Not sure if I'm adding any value, but thought I would call out as having had the exact same error as r4id.  I assume it is the exact same as the last line noted as "Code" is completely identical. 

 

My own experience is that the system is stable (currently 6d+ up), until I turn on my win 10 VM which means I'll be lucky to get a day out of it before it ends up crapping out on me. The VM is not essential but would be nice to be able to get working.  i'll continue monitor this thread and check to see if it resolves with any future releases.

 

Let me know otherwise if I am able to assist in any way.

 

CPU: Ryzen 1600

Motherboard: MSI B350M Pro-VDH (Bios version is latest official - 7A38vA4 w/ AGESA 1.0.0.6)

 

[jjslegacy 1]

Not sure if I am loosing it or haven't had enough caffeine this morning.

 

I added two new disks into my array the other day and have precleared and beat them up for a bit.  This morning I stopped the array to add one of them in.

 

When I pick the disk to add to the slot I want the page just refreshes and keeps the old unassigned status.  I can't seem to get it to stick to any of the choices I pick.

 

I have tried every browser I can find and no difference.

 

Any thoughts?

 

I see this in the syslog when I select the disk:

Aug 11 09:11:32 MJMS emhttpd: req (17): slotId.9=Hitachi_HUS724030ALE641_P8HP8WVP&csrf_token=EDDA1A93016334A3

[trurl 1569]

1 hour ago, jjslegacy said:

Not sure if I am loosing it or haven't had enough caffeine this morning.

 

I added two new disks into my array the other day and have precleared and beat them up for a bit.  This morning I stopped the array to add one of them in.

 

When I pick the disk to add to the slot I want the page just refreshes and keeps the old unassigned status.  I can't seem to get it to stick to any of the choices I pick.

 

I have tried every browser I can find and no difference.

 

Any thoughts?

 

I see this in the syslog when I select the disk:

Aug 11 09:11:32 MJMS emhttpd: req (17): slotId.9=Hitachi_HUS724030ALE641_P8HP8WVP&csrf_token=EDDA1A93016334A3

Do you have an adblocker or any web security that might be interfering? Whitelist your server. You might also try clearing browser cache.

[jjslegacy 1]

No security that I know of that would be causing an issue.  I have whitelisted the server also.

 

Is it trying to pop something up?  I haven't added a drive in some months clearly I don't remember how it all goes!

Edited August 11, 2017 by jjslegacy

[bonienl 1101]

In the earlier rc releases was a possible race condition which could interfere with the disk assigments and caused slots to stay unassigned. This was fixed in rc7.

 

[jjslegacy 1]

24 minutes ago, bonienl said:

In the earlier rc releases was a possible race condition which could interfere with the disk assigments and caused slots to stay unassigned. This was fixed in rc7.

 

 

Interesting - I am on 6.4.0-rc7a

 

Anything I can do to debug this?  Would a diagnostic file help although I don't see much in the log.

 

*added just in case**

mjms-diagnostics-20170811-1311.zip

Edited August 11, 2017 by jjslegacy