Abnorm

I'd take a look at Ubiquity equipment too. When I had 500mbit internet I used a ubiquity edgerouter lite, also had a few static VPN links up and the router never broke a sweat I also generated a huge amount of traffic locally and externally pretty much 24/7, but my setup wasn't 20 connected devices though. The lite router might be a bit too small for your needs. Unifi security gateway or bigger variants of edgerouter could also be an alternative depending on your needs, price is quite good vs business grade "higher end" gear (i.e cisco, juniper etc). Management is superb too.

cache'd somewhere i'd reckon or committed / reserved for OS stuff. totally normal, exactly the same on my system (which is 16gb as well).

You'd need a parity drive that is the same size or bigger than your datadrives (individually not combined). You cannot use a parity disk that is smaller than your data disks. If you've got lets say 2x 4TB drives, the parity-disk will need to be 4TB or bigger (lets say 6TB if you plan on using bigger datadisks in the future as the price point will be more economically viable). You can also upgrade your parity drive any time. If you've got different sized datadisks, parity will need to be the same (or bigger) than your largest drive in the array.

Hey! Sorry but I did not have the time to test a new upgrade until today, but I'm happy to report everything is working perfectly now! I believe the underlying issue was misconfigured MTU sizing, booted just fine after I set this back to 1500 in network.cfg Thanks for all the tips guys!

Thanks again for taking the time, really appreciated Allright, for some reason this is passed through to a linux VM. I've gone through my VM configs now to make sure nothing is passed through. I do not use this as this guest vm only does rar2fs. About the mainboard, it's worth mentioning that it is a sabertooth 990fx R2.0, not sure if that's obvious in the logs. It was required to install a bios upgrade when I first set up this machine, it was mentioned in a hardware compatibility chart. The reason for the upgrade had to to with IOMMU-issues as far as I can recall. There's a newer BIOS version 2901 so I'll upgrade to this if another 6.3.2 upgrade still has the same issues. The MTU stuff is most likely relevant, as it seems to me packets transactions seems to stop on 6.3.2 (explains ssh text output to just stop halfway for instance). This was defined in my network.cfg file on the flash, I've changed "MTU=9000" to "MTU=1500" now, I believe the reason I've changed this was that on my 6.0.x I had some strange issues with transferspeeds in my local network. One tip was to increase MTU to "jumboframes" on the server and also on the virtio networkdevices on my VM's. I had a transferspeed of about 100MB/s to 4-5MB/s randomly on my local network, whilst copying from shares to a seperate machine for example. I'll give 6.3.2 a go later this evening and report my findings, attached is a new log from my reboot earlier with the newest modifications. Thanks again everyone blackbox-diagnostics-20170329-1541.zip

Right, that explains it thanks!

Hey, both folders exist on the flash and both folders are empty since my cleanup

Thanks for taking the time to check it out first of all The crond from syslog i'm not really sure about either, but it is commented as "Generated system data collection schedule" , but I've uncommented it in /etc/cron.d/root now. Exit status 127 seems to be "command not found" by the way. Something old I reckon. Yes there has been alot of different packages that are crazy old, kinda explains my long boot-time but I was pretty much aware it needed to be cleaned up at some point. These has been installed at a really early stage when I was testing out different things. I'll remove the host, unmenu and nameserver lines as well now. Attached is the latest diagnostics from 6.1.9 blackbox-diagnostics-20170328-1644.zip

Nope, didn't help much. I've got an old 2-port nic installed, this has been removed and I'm reusing my onboard realtek nic for now, thought it might be an issue. Same thing happens now, ssh doesn't seem to work very well, stops while outputting text halfway. Doing it locally on the server works fine. This is in safe-mode as well, just to be clear. Webinterface still doesn't respond at all, tried sftp-ing to get the diagnostics logs and it times out when i try changing the directory chkdsk on the usb-stick gives no indications of an error either pretty much at a loss here, but rolling back to 6.1.9 again attaching latest log with onboard nic blackbox-diagnostics-20170326-1738.zip

True, I'll give it a go, thanks!

yeah thats what i did.

Hey, so, I though trying to upgrade for once, didn't really go as I expected I updated from 6.1.9 to 6.3.2 via the webinterface, booted the server and stuff seems to not working at all. I could after about 5 minutes log on via ssh, and to try to figure out why my webinterface didn't work i did a dmesg, it writes 5-6 lines and stops. Waited for about 10 minutes, it would not go any further. trying to terminate dmesg with ctrl+z did nothing. I tried opening a new ssh-session, same deal. Tried again with top to see if the server was busy for some reason, got a few lines, and it stops. Shares are not available, webinterface is unresponsive. Have tried a few reboots, nothing seems to work. Attached diagnostics. I don't seem to be able to upgrade to any higher version than 6.1.9, 6.2 upgrade also gives me wierd issues. HW compatibility issue ? Rolling back to 6.1.9 now, not sure if it'll work but I'll edit the post with results. Edit: booted up normally. blackbox-diagnostics-20170326-1251.zip blackbox-diagnostics-20170326-1256.zip

Had the same problem with windows 10, did a weekend of troubleshooting, no go. SMB seems somewhat screwed, resetting credentials is a no go, even created a specific user without any luck. Rolled back to 6.19, all good now.

Hey binhex, wonderful docker, works very very well. One issue though, there doesn't seem to be any support for FLAC (it recognizes the files and adds to db, but playback fails, missing transcodersupport i bet). Any way for you to look into this ? Found some info here; https://wiki.archlinux.org/index.php/Subsonic#Install_transcoders I'm not sure how to implement it in the docker but I think this would be a nice feature to have on your container either way.

Amazing, Lime Tech to the rescue, again ! You guys rock

Hi, it would be superb if you could also show us the actual kodi version number of the docker, as for my part I'm running Kodi 14.2 on one linux htpc, two rpi's (running openelec w/14.2) and kodi 14.1 as a docker for library updates. As with every new kodi release there are changes to the database and stuff stops getting library updates because stuff isn't compatible any more. Planning upgrades is a bit easier that way, for me at least. anyhow, fantastic work, I really appreciate your dedication in all aspects unraid, best choice I ever made

lol, this was kinda neat; https://github.com/docker/dockercraft

I guess you can connect two gpu's to a KVM switch, but the keyboard/mouse/devices part is a bit more tricky as we passthrough usb devices to a specific VM, a VM cannot share the same physical USB port as another as you're likely aware of. Although.. if you get a KVM switch with USB and you get a secondary PCI USB controller, it should in theory work fine. Check out the linus techtips video, works fine there, as in theory is the exact same setup as he uses, except the extra monitor and keyboard/mouse. You could prolly get away with a VPN site-to-site connection from home -> work -> customer . Should be acceptable from your workplace/IT dept. That depends on your employer of course. Unraid is awesome as a home storage solution, and if you're a geek (like me) and want all sorts of dockers, VMs, services and cool stuff to show of to your geek-friends... it f-ing rocks. I cannot praise it enough

I've had heaps of problems reusing old sata cables, replaced them all, never had a problem since. It doesn't seem to be an issue with your drives as far as Smart report passes, have you double checked the power cabling ? could be a bad connection or something, considering you've just rebuilt your system basic cabling errors are easily missed

Hi, you could download the txz file directly from the slack repository and put it in your \Flash\packages folder on the memory stick. It'll install the tcpdump package at boot, i use this on a few packages myself (net-snmp for example) http://pkgs.org/slackware-14.1/slackware-x86_64/tcpdump-4.4.0-x86_64-1.txz/download/ there's also a wiki article about this here; https://lime-technology.com/wiki/index.php/Methods_of_user_customization#Packages_added_dynamically_on_each_boot hope you get it sorted

Hi, I'm using Observium docker and it works great http://lime-technology.com/forum/index.php?topic=38410.0

link to .plg doesn't work though. Good idea making this a plugin though! It works, Thanks for the effort coppit! Can you connect if you open a VPN tunnel to the remote site? Yes, You'd need static VPN tunnels. It would be a WAN pretty much. We use this at work, we got locations all over the world and we monitor traffic/load with cacti via snmp.

'lo sparkly, Regarding the kodi headless docker (not coma) I'm just wondering, getting a lot of; "WARNING: Create - Compiled without non-free, rar support is disabled" in kodi.log when performing a scan. Is it possible to compile the docker with unrar? Only thing I know of is Debian removes rar-support in terms of their license; http://www.auqitainegroup.com/showthread.php?tid=205399&action=nextnewest I'm using the headless version for updating my library via cron, doesn't seem to be able to add some of my media, most likely its because some archives are compressed, which Kodi doesn't support if it's compiled on Debian. Scans on my other Kodi clients works though (kodi on ubuntu, openelec on rpi). My entire archive is rar-ed - no go extracting it all. I can live without it, but it kinda removes the purpose of running a headless version for my intended usage

Great! exactly what I was looking for, thanks a lot jonp

Hello! I've got a dual-nic pciex card and I was wondering if it's possible to dedicate ports to different VMs ? I'd like to have only unraid use the mainboard port