dadarara

deleted the old Plugin and installed new. now its not working. have old UPS. the driver I used before is upbhid-ups no IPs, no user names . it just worked. Scanning for UPS...Scanning USB bus.Scanning NUT bus (avahi method).[nutdev1]driver = "usbhid-ups"port = "auto"vendorid = "0D9F"productid = "0001"bus = "002" so now what ?

but should I : remove dirty bit ? Copy original to backup or Copy backup to original

attached the latest one tower-diagnostics-20170919-1017.zip I think there is still the error In the syslog also, what is the suggested actions running the fsck on the USB stick ? I run the bellow: root@Tower:~# parted /dev/sdg 'print' Model: Storage Xtreamer (scsi) Disk /dev/sdg: 8099MB Sector size (logical/physical): 512B/512B Partition Table: msdos Disk Flags: Number Start End Size Type File system Flags 1 4194kB 8099MB 8095MB primary fat32 boot root@Tower:~# fsck /dev/sdg1 fsck from util-linux 2.28.2 fsck.fat 3.0.28 (2015-05-16) 0x41: Dirty bit is set. Fs was not properly unmounted and some data may be corrupt. 1) Remove dirty bit 2) No action ? 2 There are differences between boot sector and its backup. This is mostly harmless. Differences: (offset:original/backup) 65:01/00 1) Copy original to backup 2) Copy backup to original 3) No action ? 3 /dev/sdg1: 3359 files, 135803/246912 clusters root@Tower:~#

connected to MAC. checked the stick with DISK utility with firstAid. nothing. dont have windows..

one more symptom after reboot, I have my system TIME wrong. I need to play with the NTP time definitions to make it right. I add one more server to the list and apply. after that the time is synced. something is really not right with my system. I dont know what should I do, without total reinstall or messing with the file systems on my disks. please help

I am reviving this topic I "kind of" bypassed the issue by moving the APPSDATA folder to the array disks away from the cach. but after reading about it and deciding that I should keep APPSDATA on the cache disk I am back to the situation where the CACHE disk gets HOT. (60c) HOW can I find out which application/plugin is the one that causes the heavy use of the drive ? I tried to use the "file activity" plugin but its not monitoring the APPDATA folder. I must point out that I have an open "on the wall" system design. https://forums.lime-technology.com/topic/56589-my-wall-mounted-server/ so the disks are totally open and there are no air flow as such. at normal operation the disks (array and the cache) are hovering around 35-45c tower-diagnostics-20170918-1009.zip

common problems didnt show anything until yesterday. found call traces. if anyone can check the diagnostics please? whats the best way to check and fix the USB stick ? tower-diagnostics-20170918-1009.zip

?

I get the users deleted every time I make a reboot. also the SSH access fails and I need to delete the known_hosts file in the remote computer. anyone can direct me to how to investigate it ?

wanted to post my solution for passing through the NvidiA GTX 1070 to windows 10 while having the Error 43 in windows. I had it working for a short time last year. Than it so happened I burned the card and sent it for RMA. after installing the new one, for whatever reason it didnt work. started to play around with all the options and fixes. obviously needed to get the ROM file as the version of the new one was different. in my case I succeeded to offload it with GPUz from techpowerup site. The windows was booting allright but with error 43 for the Nvidia1070. (BTW, trying to install the windows from scratch I got into the OVMF bios loop. for me the problem was bad ISO file. so had to download it new.) So only after I used this patch on the ROM file I succeeded. https://github.com/Matoking/NVIDIA-vBIOS-VFIO-Patcher hope this will help people

anyone?

was playing with UPS setup and had system crash while the Array was up and some VMs and DOCKER apps as well. after reboot I wasnt able to login to SMB /AFP shares. looked at the USERS tab and there was only the root user defined. with SSH access I see that indeed no users are defined in the linux system. I have defined some users AGAIN via WEB UI. But after reboot the list got empty again. It doesnt keep any users I define permanent. the shares are intact as far as I can see. I can start VMS and Docker apps. Diagnostics attached. (not that I know what it can help...) tower-diagnostics-20170911-0437.zip

WTF...... I had only one miserable VM with a space in the name and I didnt even use it. This is great, absolutely great.

whats the solution for the 500 error ?

had to reboot 2 times to make it work. there was no access to web ui. tower-diagnostics-20170521-1002.zip

is this normal? I know that 70c would be a disaster. but 56c sounds hot. normal temps are 40-45c. how can I see who is the heavy user of the disk ? which process, docker, plugin ... ?

Hi all thanks for the comments. the board is about 2m above my desk. so no one is touching. in principle there is no problem to touch any of the electronic parts as they are coated with protective lacquer (usually). the problem might be "the fingers in the fans" though and touching the pci connector pins . anyhow, the biggest problem is to blow air on it once a day or two to clean the dust away. truthfully I forget sometimes. But to compare to a closed case, there is actually less dust. In the closed case the fans suck it in and it accumulates there. on an open board like mine, it doesnt. Regarding the GTX1070 I have. well its RMA. had to send it to California USA for replacement. But it turned even to a longer story when the UPS lost the newly sent card. So now I am in negotiations with Gigabyte ...

Thanks appreciate your comment

bought it on digikey.com look for 3M riser PCI cable. 3M are the best. shielded. no interference. no need to lower the PCI speed like other cables. dont even think about anything else.

downloaded the iso from . . look for windows 10 ultra

Thanks, unfortunately, I didnt take the picture of the back of the board. will do that the next time I will take it off. but it is a mess. reasonably controlled mess. but a mess it is. the temperature units are Hot Digital LCD Thermometer Meter Tester Temperature Gauge PC Car Mod C/F they measure the temp NEAR the CPUs, GPUs, Hard disks and the water in the future etc. not something to rely on or have an alarm. it just looks nice.

My Wall Mounted Server

Once upon a time, in a land far far away, lived a prince who was strong, moderately nice looking and quite stupid….. That’s me. Well, not a prince, as far as I know. And frankly not that strong either. Nice looking? Hmmmm What made me think that this will be an easy project?! I like to build computers. Built two or three for the family use. Always dreamed to build a water cooled rig. April 2016 my son complained he needed a faster gaming machine. Right about the same time I thought I could use a desktop of my own. Up until that time a MacBook Air was enough for my needs. Fell in love with MacOS in 2010. Since then I will never go back to Windows OS as a primary computer. I also had a mini server bellow the TV in the living room with 2Tb raid 0 external storage for movies pictures etc, backed up to CrashPlan cloud. Crashplan actually works very nicely. Don’t know why some people complain. Amazingly I even succeeded to restore lots of photos from it after accidental deletion. Support was great, very helpful. So I recommend. Also thought about replacing my daughter’s desktop as well. Anyways, why not building ONE central server for all my needs, make it a water cooled wall mounted board, have fun building it and live happily ever after?! I even though that the cost will be lower than separate systems, which actually I think it is. Didn’t make the exact calculations though… So I embarked on a long journey somewhere in August 2016. First things first, needed to do a lot of research on the web. How to’s, parts, tools what other people made. Etc… Stumbled on a nice build log of a wall mounted system and though it was amazing and I was hooked. http://www.overclock.net/t/1424387/gallery-build-log-ultimate-wall-mount-rig-maxxplanck-v2-completed In parallel, looked for the best options for virtualization. ESXi, Proxmox, unRAID some others. Very quickly decided that unRAID is the best fit for me. Allowed me to have virtualized desktops, use applications as Dockers, have extendable NAS with redundancy that kept the data in its native format and didn’t demand too much hands on and linux knowledge. Started also designing the board on the Google SketchUP. BTW amazing program. While roaming the WWW, I found out that there are lots of really cheap XEON CPUs on sale. And that was my first purchase. I bought 2x XEON E5-2670 2.6Ghz for about 150$ on eBay. Everything else followed and complemented those two CPUs. The motherboard, I was contemplating some Gigabyte and Asus dual CPU boards but after reading lots and lots of recommendations, I simply got tired of all the different options out there and went for the Supermicro X9Dai board. Why? I think I made a small mistake by choosing a board without a graphic chip. Otherwise it had all the main features I needed and it had a good reputation for stability. Moreover, I’ve seen some successful Hackintosh builds with this board. For whatever reason I thought it was important to have a MOBO that I could use for OSX. With unRAID its not very important though, as it masks the HW in most part from the OS. The board was not cheap. 450$. Also, people, remember that I live in the promised land of Israel. We have a beautiful country. But we are still looking for the milk and honey, as all our neighbors have the oil. AND to ship anything to Israel add up approximately 10-15% to the total cost of anything you want to buy from outside. RAM, I was really lucky finding a very reasonably priced 64G Samsung ECC Registered. 145$ Why 64G ? you have to have enough !!! it’s better to pay a little extra but not finding yourself juggling with the configuration transferring memory between the VMs all the time. And believe me I realized that, when I had to add 12GB more RAM to the GAMING VM so that the “Battlefield 1” will play right. 1-2-3 and its done. No hassle. I had enough spare and then some. And thank you unRAID team to make is so easy to do. After making some budget calculations I have decided to postpone the water cooling part of my build to a later stage. According to my calculation the total cost would be 1000$ for the parts related to the water cooling (CPU GPU blocks, fittings, tubes radiators, reservoirs, shipment etc) And also, let’s face it, it’s my first wall mounted rig and to add on top of it my first water cooling setup would be asking for trouble. Slowly but steadily I continued to buy stuff on eBay, Aliexpress, Amazon and other places. Cannibalizing my old systems, I had: -GTX560ti -2x WD 2TB GREEN -1x WD 1TB Green and that’s about it. All I could really use in my new system. So I bought - Gigabyte RX480 Gaming for my son. Coupled with LG 34UC98: 34 Class 21:9 UltraWide® WQHD IPS that one was really extra. Part of a deal with my son. Don’t ask…. - GTX1070 for me (made a mistake as its not working yet with OSX) - two additional WD 4TB RED HDs - 1x Samsung 1TB SSD EVO750 (chipper then EVO850, but in retrospect I think should have went for the 850, don’t know really, time will tell. So far the slower speeds are ok for my needs) - PSU, EVGA 1600W Titanium, the best there is (I think). Do not save on power supply. To have problems related to power supply is a real headache. You never really know why things fail. A system hanging or the data corruption or any other random occurrences, all can come from a faulty PSU. And good luck spending time looking for WHY… In my case, I have 3 GPUs, 2 CPUs and 6 hard disks and a future water cooling pumps. And I want it to be silent. My EVGA DOES NOT TURN THE FAN ON. NEVER. So far.. I have bought a power measuring device on the main line (220v). I can tell you that the average consumption is about 200W. since December 2016 I have a total 550KW. Which is about 80$ of electricity till today 13th April 2017. These are the main parts. There are lots of others complementary stuff. (planning to create a parts list on https://pcpartpicker.com/ The most important ones where the PCI extension cables for the video cards as they are not positioned on the MOBO PCI slots. And I needed 3 of them. You know the idioms - “buy cheap, buy twice” or “cheap at twice the price” ? Well…. That’s me. I thought I could save some money and went and bought some modDIY riser shielded cables 30cm and 19cm to daisy-chain them together. For 75$ a pair. To make the story short, if anyone will need to buy raiser PCI cables , go for the 3M brand. Nothing else works. 100$ each cable. But it just works. I thought I would have problems with the HDMI long 10m and 15m cables. And the USB extension cables. But the Cabernet Ultra CL2 Active High Speed HDMI Cables work great. 3440x1440 resolution with all gaming settings set to high works great. The mouse and keyboard control without lagging with a simple MADE in CHINA USB over RJ45 over cat6 cables. My current unRAID configuration has: - OSX Sierra VM for my daily use. Utilizing the GTX560Ti for now. Thanks to gridrunner for his really excellent guides !!! And what a nice and wonderful thing it is when I can simply add 8Gb more RAM to the OSX so that the VMware Fusion running nested windows VM will feel nice and worm. Plus some extra CPU cores to have a complete Nirvana. - Windows VM for my son’s Gaming using the pass-through RX480 GPU. And he can switch it on if its down via WOL add-on using his mobile app. Fantastic. - Windows VM for desktop use for my daughter. She is supposed to use the GTX1070. But at this time the board is RMA in USA . my fault. Never play with molex power cables while its still “hot” plugged, powered and working. Well now its not. Waiting for the card to be sent back. - Windows Ultra-Light install VM (1.5GB disk space windows 10 install for the PRTG networking monitoring setup. - Ubuntu VM to have the Virt-Manager running. Makes the configuration of passing through devices to VMs really easy. And also its VNC client works when the noVNC in unRAID doesn’t. I am really would love to see a Docker with the virtManager working. Does NOT work for me. Yet. - Docker with CrashPlan - Docker with Emby Media Server - Docker with Krusader file manager, although I am using MC via the terminal ssh access. - Docker with NextCloud cloud database. Something new. Want to make all the family mobile phones backups. - Docker with PrinterCUPS. Great stuff. Print server for every computer in the house being able to use one printer located in my small office room. Didnt find how to use the HP printer SCANNER in the same networked setup yet. - Docker with delugeVPN torrent downloader. - Dynamix SSD TRIM Plugin - Dynamix System information Plugin - Dynamix System statistics Plugin - Fix Common Problems Plugin - Libvirt Hotplug USB Plugin - Nerd Tools Plugin - Preclear Disks Plugin - Speedtest command Line Tool Plugin - Tips and Tweaks Plugin - Unassigned Devices Plugin - unRaid DVB Edition Plugin (needed it to play with passing through firewire on board device) - unRAID Server OS Plugin (duhhh… ) - User Scripts Plugin (AutoVMsBackup, icon download and sync, USBportscripts, vm settings backup) - Virtual Machine Wake On Lan Plugin some pictures They were watching too...

Thanks guys. but this forum is not the right place for posting pictures and builds (I think) I dont even know where would be the right place for it, which forum...

Hi guys I am doing trial and error here. gathering stuff from other people's successes. I am not really clearly understand it all. and might be I am explaining this wrongly but lets try for the sake of other people: 1) pcie_acs_override=1033:0194,1b73:1100 : ACS override patches, thats for the 2 identical USB hubs I have. (and the USB hub PCI extension card for which I am not sure I need to do it though) it supposed to override the IOMMU assignments and separate them into two distinct groups. you need to specify the device id. Again, not sure this is the exact explanation, but it works IOMMU group 21 [1033:0194] 08:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host Controller (rev 04) IOMMU group 22 [1033:0194] 09:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host Controller (rev 04) 2) vfio_iommu_type1.allow_unsafe_interrupts=1 : REALLY dont understand what exactly it does. I do know what interrupt is, I dont know what is unsafe one. seen it a lot so decided to use it as well... 3) isolcpus=2-13,18-29 thats for the isolation of the CPU cores I want to use for the unraid, dockers and stuff 4) vfio-pci.ids=8086:244e and the modprobe.blacklist=i2c_i801,i2c_smbus here it starts to be tricky . at first I was trying to assign both the PCI bridge and the Firewire device to the vfio , but after some trial left only the bridge in the command. regarding the blacklisting, I saw in the /var/log/syslog the following lines: Mar 20 19:16:09 Tower kernel: vfio-pci 0000:0a:02.0: enabling device (0000 -> 0002) Mar 20 19:16:09 Tower kernel: genirq: Flags mismatch irq 18. 00000000 (vfio-intx(0000:0a:02.0)) vs. 00000080 (i801_smbus) I got the idea that the smbus of my motherboard gets in the way of the IRQ assignments and I have stumbled upon some entry that was dealing with the exact description of my smbus. I didnt know how to define the smbus in the blacklisting until I saw that post (somewhere...) another step was to ONLY pass through the firewire device and NOT the PCI bridge. passing both I get the error that the bridge couldn't be reseted. and indeed it does not have this feature. I would really like someone to confirm that I understood the meaning of ">Reset- " as CAN NOT accept reset. or I am totally wrong here ... see bellow. Also, I didnt understand, why the system couldnt send a reset to the whole PCI bridge and everything behind it. but it doesnt matter now, as it works. root@Tower:~# lspci -vvvs 0:1e.0 00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev a6) (prog-if 01 [Subtractive decode]) Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Latency: 0 NUMA node: 0 Bus: primary=00, secondary=0a, subordinate=0a, sec-latency=64 I/O behind bridge: 0000f000-00000fff Memory behind bridge: df900000-df9fffff Prefetchable memory behind bridge: 00000000fff00000-00000000000fffff Secondary status: 66MHz- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort+ <SERR- <PERR- BridgeCtl: Parity+ SERR+ NoISA- VGA- MAbort- >Reset- FastB2B- PriDiscTmr- SecDiscTmr- DiscTmrStat- DiscTmrSERREn- Capabilities: [50] Subsystem: Super Micro Computer Inc 82801 PCI Bridge so now the FireWire works in the windows VM and I can capture the DV camera. (you will not FU#$@#ing believe what I was trying to do to make it work. as I have 3 video cards in my rig and I have some physical problem to put an addon card, I had to make it work. so I tryed to install TVheadend and SageTV dockers and make them work with the FW0/FW1 devices the unraid recognized the firewire and the camera. and install some dvgrab utility to grab the video. I almost succeeded. I could control the camera but no capture of video stream. BUT what the hell. I learned a lot. and now it works I want to thank all the guys here in the forum for the help. this unRAID is really something. I have now replaced 5 computers. my son is really happy with his Gaming capabilities. fast games and excellent video quality. and the whole rig takes 200-250W . this was almost the power usage of only one computer before.