Leaderboard

The attachment in this post is a joint effort between @Batter Pudding and myself. @Batter Pudding supplied much of the technical part of the Attached Document and I provide most of the background information. What we are attempting to do is to show that it is easy to actually use Unraid with all of the security features that Microsoft has incorporated into Windows 10. What many of us have been doing (myself included) is to reverse those enhancements to security and use our Unraid network in what is basically a 2010 security environment. @limetechhas announced in the release thread for version 6.9.2 that they are about to increase security on Unraid in future releases. Unfortunately, this list is going to impact a lot of current Unraid users as many have setup their Unraid servers and networking to use these very features. Each user will have two choices. Either embrace security or spend time to undo each new security addition that either LimeTech or MS adds in their updates. If you decide to continue to bypass security, just realize that the number of folks prepared to assist you with any problems doing this will probably decline as more folks adopt increased security as a necessity. In some cases, this is going to present some difficult decisions. For example, I have an old Netgear NTV-550 set top media player (last firmware/software update was in early 2011) that only supports SMBv1 or NFS. Do I open up a security hole to use a well-functioning piece of equipment or do I replace it? (The choice, obviously, is one that only I can make...) Two Important things! Do not post up any problems that you have with networking between Windows 10 and Unraid in this thread! Start a new thread in the General Support forum. Please don’t tell us that there is another way to do something and that we should change our recommendation to employ that method. If you feel you have a better way, you are encouraged to write it up in detail and post it in this thread pointing out the advantages of your way. (One well regarded Windows 10 networking book has over 400 pages in it. Our document is 16 pages long…) EDIT: November 30, 2021. Recently, something has come to my attention about Unraid and SMB. There have been incidences where access to Unraid shares is restricted or blocked completely from users who should have access to it. What has been found in these cases is that a feature, has been enable on the Unraid side, called Access Control Lists (ACL for short). This will show up as an ‘+’ at the end of the Linux permissions. See the screen capture below: Note that the ‘+’ is also on the file as well as the share/directory. ACL changes the way that Linux is going to control access to these resources. After some research, I found out that Windows has used ACL for a long time. The SAMBA group has added ACL into its version of SMB. Unraid does not use ACL in its security scheme. At the present time, I can think of only one way that a ACL could be found on any Unraid server. It was done by a Windows user who was trying to change how SMB worked by applying Windows security features to an Unraid share by changing the default Security settings. (Basically, right-clicking on the Share in Windows Explorer, selecting ‘Properties’, then the ‘Security’ tab and working from there.) The point I am making is that you can’t fix a share access problem by trying to change a Unraid share security using Windows security tools on that share. If you try, you will probably make things worst! (Unless you are a Windows SMB Networking Guru…) It is important to realize that if you are denied permission to an Unraid share resource, the problem can only be fixed on the Unraid side using the Tools in the Unraid GUI (or via the command line for specific problems). If you are having an access problem to a Unraid share and can’t solve it with the tools in the GUI, start a thread in the General Support sub-forum and let the community help you fix it. EDIT: February 25, 2024. It has come to my attention that there is one more setting that was not previously covered. The Network type should be set as 'Private'. For Windows 10 Open up Settings Pick Network & Internet In right panel, click on Properties button In the Network Properties, select the 'Private' radio button. For Windows 11 Open up Settings In left panel, pick Network & Internet In the Right Panel at the top, click on Properties Under Network Profile type, click on the 'Private network' radio button EDIT: March 15, 2024 It has come to my attention that there can be problems in linking the Network Neighborhood folder into Windows File Explorer in Windows 11. While there is a solution to that problem, it has some other side effects and the full scope of those has not been evaluated. In addition, I have become aware of another way to integrate access of servers into Windows File Explorer that works quite well if you have only a few servers. (Things start to look messy if one adds more than two or three servers but that is a matter of each individual’s perception of “messy”.) So if you have having any problems with implementing "Network Neighborhood", try this new approach! This new method is actually quite simple to set up. This method is described in the attached PDF file named: “An Alternative Method to Network Neighborhood.PDF” Unraid & Windows 10 SMB Setup.pdf An Alternative Method to Network Neighborhood.pdf

At the moment the Edit option for a VM does not allow you to specify the port to be used VNC, but always sets it to 'auto'. If you manually use the Edit XML option to set the port to a specific value then when you save the settings the GUI is quite happy to correctly display the port you have just set. The problem is that if you now use the Edit option then it is reset back to the 'auto' setting. What I would like to see is the VM Edit option bringing forward the current setting, and allow you to set it to either auto or to a specific port.

Well, it's probably a mode=fullnode so it's doing everything except plotting. Try a restart of the container and see if it drops. I restarted yesterday and am at ~3GB. You may want to pull the latest test version of Machinaris which includes Chia 1.2.1 (soon 1.2.2) as its my understanding that Chia 1.2.0 was a bit of a resource hog.

Ja

Thank you very much!

Sieht richtig aus. Wenn du stattdessen die Parity als neue Disk dem Array hinzufügst, könntest du den Server inkl Docker und VM normal weiterlaufen lassen. Falls du während dem Sync noch Datenänderungen durch Docker / VM erwartest, könntest du nach dem Sync alles stoppen und kurz einen zweiten Sync drüber laufen lassen. Der bräuchte dann vermutlich nur ein paar Sekunden. Bei der Variante wäre also die Ausfallzeit von Docker / VM deutlich kürzer. Wenn das aber egal ist, kannst du es natürlich auch wie bei deiner Idee machen.

thanks for your help, I check it now and all keys are the same in both in windows and in the unraid docker container

Wow! I had a bunch for P1-DIMMC1 so I moved some RAM around and took out two sticks. I'll watch it for a while. Thank you so much for your help!!!

Perfect. Thank you so much for your help.

Yeah @Meannekes i totally agree and get your point, i should have gone with my 2 other highspeed "SPCC_M.2_PCIe_SSD" Silicon Power SSDs. I have 2 Unraid systems and the one with 2 of thoose disks in it, isnt mission critical, its more like hold some Docker Settings for plex or so and while its set to cache only, the settings are not that important. It runs raid1 anyway and its been barley used as a file cache. So failure rate should be low to extremly low. Next time the better ones

Okay ich habe mich danach für das beQuiet Pure Power 11 CM entschieden, da ich jetzt kurzentschlossen die Hardware komplett bestellt habe und nicht mehr warten wollte. Das Cache-Konzept habe ich nach deinem Post auch nochmal im Detail in der unRAID Doku recherchiert und soweit verstanden. An dieser Stelle nochmal vielen Dank @mgutt für deinen Einsatz hier und in den anderen großen Foren - ich habe da sehr viele hilfreiche Infos draus gezogen (Stromsparen, andere Tutorials und unzählige relevante und sachliche Posts.

There's a RAM problem: Jul 14 05:53:52 Tower kernel: mce: [Hardware Error]: Machine check events logged Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: HANDLING MCE MEMORY ERROR Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 7: 8c00004000010092 Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: TSC 26fa9a273d20a Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: ADDR 11da50f40 Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: MISC 4405c1a86 Jul 14 05:53:52 Tower kernel: EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1626263632 SOCKET 0 APIC 2 Check the board's system/ipmi event log, it should have more info on the affected DIMM, then remove or replace and see if the errors go away.

OpenVPN-AS was deprecated by linuxserver. If you've already had it installed, you can find it in Previous Apps. If you've never had it installed and want to run it knowing that they don't support it, then in Apps - Settings, enable show Deprecated Apps They are suggesting that you use their Wireguard App (or the built-in one in the OS)

thats the reason i gave my mx500 back and bought an Samsung 870 evo plus. i dont like to deactivate SMART checks.. yes it is a false alarm on this SSD normaly, but what is if there is a real error of this kind on an MX500, you wouldnt notice it. just my 2 cents.

I put this in the site configuration for the Unraid admin site - but it depends on how your Nginx reverse proxy is setup. For my setup, it was /mnt/user/appdata/nginx/nginx/site-confs/default

Update to 6.9.2

What version of Unraid? This is a known issue with 6.8.3 because of changes at Docker hub end although there is a workaround you can install.

I ended up shutting the system down from vacation. When I came back I re-seated all cables to SAS-expanders and booted again. Parity was marked as invalid. I did a read-check of the array which went without errors. Then I stopped the array, unassigned the parity drives, started the array, stopped the array, assigned parity drives again. It re-built both parity drives without errors, and is seemingly running well now. I'll pay attention to the RAM usage going forward. Thanks.

Binhex never-mind I realized the moment I posted that I had to set to yes to have folders created. I did and sure enough the folders appered and added files and seems to be working perfect! Thanks

Thank you, Terraria is now working. I appreciate the update!

Weiß gar nicht ob es wen interessiert, aber ich poste hier mal Update. Nach langer Suche hab ich bei Galaxus nen 9100 für 115€ bekommen. Welch Ironie, dass ich im Winter zu geizig war 110€ auszugeben und auf eine Senkung zum Sommer spekuliert habe Natürlich musste zwischenzeitlich der RAM in die Höhe schießen...aber die Pille habe ich jetzt geschluckt und heute sollte auch der 16GB Kingston Riegel kommen, zusammen mit nem Y-Kabel für die Gehäuselüfter im Node 304. Hatte nicht auf dem Schirm das das Gigabyte Board nur einen Lüfteranschluss hat. Bis auf den RAM ist alles montiert und als mein erstes Intel System, auch etwas seltsam die Pins im Sockel zu haben. Weckt irgendwie nicht so viel Vertrauen. Zusammengefasst sieht es derzeit so aus i3-9100 Gigabyte C246 ITX 16GB Kingston 3200 Corsair RM 550x 2021 500GB SN550 NVME Fractal Design Node 304 4TB Skyhawk + 3 TB WD Blue (hatte ich noch) + 1 TB SSD Frage Bin zwar noch nicht soweit, aber wenn ich Medien der Smartphones täglich über Unraid gesichert haben möchte, wie stelle ich das am besten an? Bisher habe ich am PC Acronis laufen und die App macht ber WLAN in der Nacht ein Backup.

Yes.

Hi Increasing the Shutdown timeout fixed this for me. Thanks very much!

Thank you ich777. I will await your advisement. Just glad to hear I didnt botch something here lol.

Dear Jorgeb It work correctly with reversed cable thank for you advice Best Regards

Ambient temperature is about 73F All are inside the 1U case, except as noted. Temperature : 28 C (82 F) Temperature : 27 C (80 F) Temperature : 28 C (82 F) Temperature : 26 C (78 F) Temperature : 28 C (82 F) Temperature : 28 C (82 F) Temperature : 28 C (82 F) Temperature : 28 C (82 F) Temperature : 43 C (109 F) SSD in a poor air flow spot in the case Temperature : 25 C (77 F) Temperature : 25 C (77 F) Temperature : 25 C (77 F) Temperature : 25 C (77 F) Temperature : 28 C (82 F) laptop drive outside connected, in free air Temperature : 41 C (105 F) outside USB connected drive I'd say, this NR12000 does a remarkable job keeping the disks cool.

Ich hab es doch erklärt wie sich die 65W TDP ergeben. Auch der 10900 kann über 200W bei Volllast ziehen: https://www.pcgameshardware.de/Core-i9-10900-CPU-276170/Tests/Kaufen-Benchmark-Review-vs-Ryzen-9-3900X-1356342/ Das ist richtig, hängt aber klar vom Spiel ab. PCGH hat zB beim 10900 bei Spielen 55W gemessen: Das ist aber über alle Spiele hinweg der Durchschnitt. Einzelne Ergebnisse und Spitzen ergeben sich dadurch nicht. Also noch mal zusammengefasst: Bei durchgehender Last mit zwei laufenden Cities Skylines VMs wird die CPU nicht mehr als 65W ziehen, sie kann aber nach einer Ruhephase kurz über 200W ziehen, weshalb das Netzteil auf keinen Fall zu klein geplant werden darf. Durchgehend 200W zieht sich aber niemals. Auch nicht der 10900K. Der wäre auf 125W limitiert. Runtertakten bringt vielleicht 1W und erhöht nur die Instabilität des Systems. Der 2400G mit altem Chipsatz ist im Leerlauf vermutlich ein größerer Stromschlucker als ein 9900K (natürlich ohne GPUs). Sparsam wäre nur ein 4xxxG mit B550 Chipsatz. Hier ein Beispiel: https://www.hardwareluxx.de/community/threads/die-sparsamsten-systeme-30w-idle.1007101/page-64#post-27807563 Die Idee wäre optimal. Ein sparsamer Server wäre zB ein C246N-WU2 ITX Board gepaart mit einem i3. Da kämst du inkl. Festplatten bei unter 8W raus. Messung von @Anym001: Da das Board recht teuer ist und du auf ECC RAM keinen Wert legst, würde ich dir eher zu einem günstigen H310N raten: https://geizhals.de/gigabyte-h310n-2-0-a2042493.html?hloc=at&hloc=de Da dann einen gebrauchten Pentium Gold oder i3 für 50 € drauf, die restliche Hardware von deinem Ryzen PC und fertig ist der sparsame 24/7 Server. Ist nur die Frage ob dir 1x NVMe und 4x SATA beim 24/7 reichen. Wobei man später, wenn es mehr Platten werden sollen auch eine SATA Karte nachrüsten könnte. Wenn dann der Gaming-Server nur bei Bedarf hochgefahren wird, kann einem der Mehrverbrauch der CPU auch egal sein und man kann einen 10900 oder 10900K nehmen.

Apparently it's the name they gave their server.

Sorry for being late to the party. I think the question about why it is a good idea to stop containers when running a backup has been addressed adequately. Check out the following post for how you can achieve it:

This would be a question better directed at Plex, rather than unraid. They would be able to confirm what those IPs are for, or if they are foreign. Most likely, however, they are the IPs for the Plex server as the Plex application allows both direct and indirect connections thru their own "proxy" service on their side.

Forgot to mention, unrelated to this issue but the server was running out of RAM for some time before this, you should also look into that.

There were errors in multiple disks: Jul 3 21:43:36 Tower kernel: md: disk4 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk1 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk2 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk6 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk8 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk0 read error, sector=8177890032 Jul 3 21:43:36 Tower kernel: md: disk29 read error, sector=8177890032 Also the cache device, could be a connection/power problem or HBA problem.

The OP's issue has nothing to do with the GUI, it's not the first time you post the same wrong information, please stop.

Please post the diagnostics: Tools -> Diagnostics

Enable syslog mirror to flash then post that log after a crash.

Ich benutze ein AES verschlüsseltes 7z Archiv 😄

You can always click on the flash drive on the Main tab and select the option to download a backup as a zip file.

Here's another version - looks the same but I got bored and decided to recreate it from scratch.

Open device manager You will see your unknown devices. Right click the unknown device and select "update driver". Select "Browse my computer for the driver software" Click browse Select the CDROM Drive virtio-win-x.x Then click next. Windows will scan the entire device for the location of the best-suited driver. It should find a RedHat network adapter driver, follow the prompts and you're in business. ** I never bothered to locate the actual subfolder of the driver on the virtio-win-1-1 image, I just let windows do it for me. ** Hope this helps.

Here's a sample that will only allow the 192.168.2.* machines write access to the NFS share. Everybody else gets readonly Additionaly, the options allow root user access from the 192.168.2.* machines. Everybody else gets mapped to uid=99 192.168.2.0/24(sec=sys,rw,no_root_squash,insecure) *(sec=sys,ro,insecure,anongid=100,anonuid=99,all_squash) So, if the client IP fails to match whatever you have listed, they don't get access.