plupien79

PM Sent...

Simple... Windows sees a kilobyte as 1024bytes Linux sees a kilobyte as 1000bytes So a gigabyte for each system looks a bit different: 1000*1000*1000=1,000,000,000bytes 1024*1024*1024=1,073,741,824bytes So... 44,000,000,000/1,000,000,000=44 gigabytes 44,000,000,000,1,073,741,824=40.978 gigabytes. Tada

Depending on what you need CPU wise. The 1037u Celeron is a great option. A board with that CPU is under $100 easy. My Network Setup... 8 Disk Unraid - 1037u Kodi Box - 1037u Router 847 HP Laptop - Domain Controller and PBX Plus switch and modem PoE for Voip Phone and WAP draws ~150w It all depends on if that can do the CPU lifting you need.

Is it possible to keep your VMs up and running regardless of the state of the protected array? Assuming the VM is stored on the cache drive or a non-array drive. Here's my Use-Case Argument. A user is running a PBX as a VM. They want to put the array into maintenance mode to scan disks. However, they still want to be able to have their phone system up and running. Is this possible now, is it planned, if not is it worth suggesting?

I have notice it seems to be a very similar situation. Would love to know why, though.

It seems that converting my data disks to XFS has solved the issues. I'm on my 14th day up without issue or interruption. Anyone have advise on how to swap the cache drive from Reiser without messing up my dockers?

Are you disks on ReiserFS?

From the Internets...

So this is writing directly to the protected array. 80-90MBs is quite normal performance. This is due to the slower disks and the parity calculations. Hence the reason many people opt for a cache drive.

Is the slower folder being watched by EMBY? Or anything else for that matter. If they are competing for resources that could account for the drop in speed.

And a discrete video card to pass through for Gaming on the windows 10 box. A GTX 950 would be more than enough for some gaming. They can be found under $150 usually.

I switched all my data drives to XFS over the last week. However it is far too soon to tell if this resolves the issue. Same hardware had 163 days of uptime before switching to version 6. Current record with version 6 is 14 days.

This sounds a lot like the issues that I and some others are having. An attempt is made to write to the array, then for some currently unexplained reason, the write operation doesn't work, but also doesn't fail or create any errors. Then the SHFS process involved just sits using as much CPU as it can raising the load on the server until it becomes unresponsive. You then hard reset (a software reboot fails) do a parity check and find no errors. Are there any messages on the physical console?

This Works... Thank You

Ok... I've switched all my data drives from ReiserFS to XFS. This MAY have solved some things. Not 100% sure yet as only time will tell. I did see an odd Reiser error on my terminal when I had a crash: "REISERFS error (device MD2): vs-4010 is_reusable block number is out of range" but running the scan in Maintenance mode found nothing, so I moved all my data disk by disk and changed to XFS. Ran a parity check and now everyone is humming along.

http://www.amazon.com/Supermicro-Motherboard-Proprietary-DDR3-A1SA7-2750F-O/dp/B00KDOA00I/ref=pd_sim_sbs_147_9?ie=UTF8&dpID=314-qNXGRCL&dpSrc=sims&preST=_AC_UL160_SR160%2C160_&refRID=11J53NB74V527ZTE74B4 This has 17 SATA3 ports onboard, a 20w 8 core CPU. Anyone taken a look at this thing for unraid?

There are several Dynamic DNS Dockers and Plugins. As for support packages, I don't thing there is one for Unraid yet, but a docker could be ported quite easily.

https://blog.docker.com/2015/11/docker-1-9-production-ready-swarm-multi-host-networking/ The Swarm seems quite nice, ESP if a PBX docker is ever released, or anything else that isn't bound to the Array.

I never used any of the beta's. I suspect it's something funky with the NM70 Chipset and the Linux kernel. I've had issues in the past with OpenELEC, but this is a different issue. But since nothing seems to show up with the logs and it seems that I'm the only person using this hardware it seems like a big hill to climb.

Think I'm going to throw in the towel. I'm going to roll this box back to version 5. The goodies are nice, but I need a reliable NAS more.

root@InfoSphere:/usr/local/emhttp# lsof -p 21149 COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME shfs 21149 root cwd DIR 0,2 380 2 / shfs 21149 root rtd DIR 0,2 380 2 / shfs 21149 root txt REG 0,2 83216 4568 /usr/local/sbin/shfs shfs 21149 root mem REG 0,2 171470 6323 /lib64/ld-2.17.so shfs 21149 root mem REG 0,2 134893 6252 /lib64/libpthread-2.17.so shfs 21149 root mem REG 0,2 48928 6341 /lib64/libcrypt-2.17.so shfs 21149 root mem REG 0,2 1944360 6269 /lib64/libcrypto.so.1.0.0 shfs 21149 root mem REG 0,2 242840 6305 /lib64/libfuse.so.2.9.3 shfs 21149 root mem REG 0,2 2102965 6246 /lib64/libc-2.17.so shfs 21149 root mem REG 0,2 18988 6296 /lib64/libdl-2.17.so shfs 21149 root 0u CHR 1,3 0t0 1029 /dev/null shfs 21149 root 1u CHR 1,3 0t0 1029 /dev/null shfs 21149 root 2u CHR 1,3 0t0 1029 /dev/null shfs 21149 root 3u CHR 10,229 0t0 5361 /dev/fuse shfs 21149 root 6r REG 0,2 984 4588 /usr/local/emhttp/update.htm

Did it again... after 11 days this time. infosphere-diagnostics-20151024-1838.zip

I can... After looking at the ARK site for Intel it looks like all the Skylake CPUs support VT-d out of the box. Even the LOW LOW end chips.

Any other troubleshooting? Is there anything in the logs that looks funny... Cause I'm not seeing anything?

This has been done. Several Times in fact.