NAS-newbie

I have spent a lot of time researching what may prevent my Unraid server from spinning down the array but so far have not found the cause and need some more sugestions. I have installed the Disk Activity and Active Streams plugins that, when I dont exepct activity dont show any even though I, when disks spin up in dashboards see a short spike of both reads and write bandwidth use (300-400 kb/s). Most recently I even tried disabling VMs and Dockers totally and spinning down the disks but within minutes something is waking them up. Configuration attached - happy for any help! nas-diagnostics-20250713-0953.zip

I have a multimedia share that right now is set to only use array but still a coresponding dataset (I use separate zfs filesystems for each disk in the array) is for some reason created "automatically" in the cache. It seems like only empty directories are created in this cache dataset and not the files I am actually copying to the share making it even more strange... The share is only updated by myself manually copying files from other shares mostly using midnight commander container and some odd time using unraid shell. What am I doing wrong / how can I avoid this from happening?

I tried this and the behavior was really strange - when connecting with the UnRAID built in console all keys including "~" worked but when pressing this key the VM was really laggy responding and after this "~" auto-repeated several times and from this point on the keyboard mapping acted totally wrong - no keys worked as they where expected returning various symbold instead of the expected characters. Do this key have some special function in remote sessions or how can this be? When I connected with TeamViewer, seeing thew same desktop, the keys worked correctly but of course ~ still did not work as I had not changed anything in this environment. I can switch back and forth between the two and see the same behaviour, i.e. wrong mapping in UnRAID console and mostly working in TeamViewer... Is something seriously wrong with my VM or wtf is going on here?!?!

Thanks will try that!

I have created a Linux Mint Cinnamon VM that I have tried accessing both with the UnRAID VM level console as well as with TeamViewer (from a Linux Mint Cinnamon laptop) and in both cases my keybinding (Swedish/Swedish) works perfectly witgh ONE VERY ANNOYNING EXCEOṔTION - the tilde "~" key that instead of the expected character seem to be mapped to some "combination key" that when I press another key prints a blank (or non displayable character?). When I connect with SSH to the VM all characters including tilde work as expected. I have psecified the keybinding for the console to Swedish and also set Swedish/Swedish as keyboard binding both in the guest OS and in the laptop I use to connect with (that of ocurse handles tilde correctly when used by itself). Any thoughts on additional things I may need to do in the VM or UnRAID configursation or how to further investigate the problem are very appreciated!

My USB drive have given up and I am trying to replace it with a new SAMSUNG BAR (got a few of them after recomendation that they where stable for UnRAID). Unfortunatly the new USB Creator, for some reason, flag it as blacklisted even though it is brand new so I am trying the manual method as proposed in a forum post. My problem is that I am still running UnRAID 6.12.6 and this release is not available on the download page (I supose I should have saved the downloaded version somewhere but was assuming geting hold of it if needed would not be an issue) - can I use a later 6.12.y release like 6.12.9 for my flash or do I have to use the exact same version? How to best proceed?

I can see that there periodically (even with Docker & VMs disabled) is small amounts of both read and write activity that wakes up all the disks. Nothing shows up under neither disk or shares activity in the "file activity" plugin. Are there other plugins or step one can take to track this activity down?

I have only shares for appdata, domains, isos and system only on cache and syslog and some other folders for media etc as cache->array. Nothing explicitly for /mnt/user (this is not even a share by default in UNRAID). I was kind of assuming the share "system" was mounted as /mnt/user/system - is this not the case?

Thanks for the suggestion - had that enabled some time ago when I turned some directories into datasets but have turned it off again (found it in disk settings). Besides I do not know what would write even to ONE disk (even when I have both Dockers and VMs disabled)? I do run the directory listing plugin (periodically listing directories on the disk to keep this cached in memory) installed but that is supposed to help AVOIDING disk access and has worked will in the past (I have not changed any of its settings) but I guess I could try disabling it...

I once had my UNRAID system working as I intended with most disk spun down and activity mostly going to cache but lately this is no longer the case - all disk are now running all the time 😪 I have done some first steps to diagnose the problem but so far not found the reason: I tried disabling first docker and then also VMs in settings than spin down the array but after just like 1 or two minutes all disks lit up green again (even my latest disk, the third disk in array, that hardly have any data on it). I have tried the disk activity plugin but it shows no share activity and when I enable cache/disks I only see cache activity. Not sure what to try next? Included my diagnostics if that can provide any clues... nas-diagnostics-20241117-0737.zip

I get a bit confused about some advice in some forum posts about specifying the docker appdata setting to a path under /cache. As I understand it, if you simply have the appdata share configured to cache only, this is enough to ensure the container settings are stored in cache - i.e. what you specify in container appdata is where the appdata share will be mounted (in addition to showing up under /mnt/cache/appdata)?! In my case I have appdata configuration in docker set to /mnt/user/appdata/. Similarly if I have specified that the system share should be stored on cache only, is it enough to specify docker file location as /mnt/user/system/docker/docker.img to have it residing in cache (i.e. is the system share mounted as /mnt/user/system?)? Also for this I have seen some mention of specifying a location under /mnt/cache but this seem pointless to me?! I have tried finding the recommended settings for Docker vdisk and appdata (as mentioned above my understanding is /mnt/user/system/docker/docker.img and /mnt/user/appdata/) but has failed... Can somebody confirm or correct my understanding of this?

It was indeed a network problem (bad router port it seems) - thanks for the suggestion!

All of a sudden in the middle of the day I lost all contact with my UNRAID 6.12.6 server (and everything running on it like Plex etc). I switched to the text console that was frozen. Not seeing any other option I reset the server. When it came up again I noticed a few odd things in the boot printout: * Cant stat /usr/local/bin/mover * Logger send message failed: Bad File Descriptor * Most relevant I guess the IPv4 address set was an 169.254.xx.xx address. With no proper IP set I still cant connect to the web console. I also tried logging in with "root" and no password in the text console (that I thought was the default for the text console) but this was not accepted. Also tried "safe boot" but still not getting a proper IP. My network seem to work perfectly for other machines as usual. I obviously do not want to keep re-booting without a plan how to fix this so postponing further experiments hoping for some advice what to try next here... Can create another bootable USB with Linux or whatever to validate there are no problems with the machine getting an IP when not running UNRAID but I guess this will not find any problem...

I run on UnRaid and have had zero problems for more than a year but after the most recent update of the container the client do not work as it should 😞 After seemingly succeeding to connect I see no list of torrents in progress and I cant add new torrents. I can however I see and even change the preferences. In Deluged container log I see no errors but in the file config/deluged.log (that I can list when connecting to the container console) I find this: Deluge client disconnected: Connection to the other side was lost in a non-clean fashion: Connection lost. Any tips?

I have 63W (at 7% with a T-series i9 CPU) but then I run two enterprise SSD for cache (U.2 NVMe), 3xPCIe sticks (idle at this point) and 4x20TB WD Enterprise HDDs (powered down right now). No GPU as I rely on the one in the CPU for my plex needs.

Yes I also run with two 32GB modules at 4800 (and do does several people who have reported this in the thread) I but would like to know if anybody managed to run with 4x32GB and in that case with what modules and what speed...

Still no experiences of running this board with the full 128GB RAM? Curious about what ECC sticks that works for this and what speed that was possible?

I am using Core i9-13900T - probably total overkill unless you will run a lot of containers, have many users etc. but I will also use it as a build server and some other things and since the build anyhow became quite expensive I wanted some future margin on capacity rather than save some bucks on the CPU. It runs cool and does not consume a lot of power so has worked out nicely. If you really intend to push the CPU at times and want to avoid temp throttling go for a good cooler (when I pushed it really hard it went up to 200W+ according to the display on the UPS)...

Sorry missed that "root cause" and yes it was indeed the problem - must say it is really unexpected and odd that this makes a difference - I would even go so far as saying the other (non-working) order is the more intuitive than the one that work.... To avoid more users having this problem I would really try fixing this and please also change the default - at least for a cache pool and probably even in general on UnRAID I would guess an alternative with redundancy is by far the most likely the desired (and the one with the least "catastrophes" risk if selected by misstake). An odd thing I noticed that even with my RAID0 cache pool the shares placed on it (cache only) was shown as "green" (that I interpreted as "protected") even though they was in reality highly insecure to a single disk failure. Due to this I did not notice the problem until I realized that the size of the cache array was to large for using mirror...

Mine reads "1 group of two devices" and this is ok right? I can actually mention that about a week or so ago I managed to create a ZFS cache pool (I tried several things then also before it worked like first formating the disks for ZFS before adding them, leaving them empty etc.) so not sure how I did it to finally succeed but then I used "default settings" and was burned by same thing as customer in the referenced thread i.e. I got RAID0 (not a good default) and that was the reason I had to recreate the pool again (after all the usual steps to empty it etc) but now it does not work again and with these correct settings I can again figure out how to make it work....

I have looked at space invaders tutorial and believe I followed that quite well - I did the following steps - create a new cache pool (name cache, zfs, mirror, two devices) and assigning two empty (no filesystem) disks. When I start the array the disks are expectedly shown as unmountable. I specify that I want to format unmountable array disks but this does NOT result in them getting partitioned/formatted instead it finished more or less immediately with no error message and the disks still "unmountable". I can add that I previously have had a BTRFS cache pool working perfectly on the same hardware and that I have a non-cache ZFS pool that was created without any issues. The hardware have worked perfectly up to now so I think the risk of hardware problems is very low (recently did extensive RAM test when motherboard was new, server have ECC memory etc). What am I missing?

I need to copy files from an unassigned HD (was previously part of an array but I have done a new configuration with larger disks so uses it as "backup") to my UnRAID array but for a lot of directories and files I get the error message from rsync "failed to stat "/mnt/disk1/:somefilename Invalid or incomplete multibyte or wide character (84)" and when it is a directory that is the problem also "*** Skipping any contents from this failed directory ***" I assume these faulty files and directories (there are many hundred of them) was created by SMB from data on a Windows machine (I live in Sweden where we have åäöÅÄÖ and I can see from the failing names that it is the ones including these international characters). From the error messages it SEEMS from a few example that Ä in directory names result in \#303\#204 and ä results in #303\#244 while in ordinary files Ä is #216 and ä #204. This seems strange to me - i.e. do the files use more than one encoding?! As there are so manty affected files and directories spread out over the whole disk it is not very doable to manually rename them one by one and I instead need to find a way to tackle the encoding issue with a script and some utility working on character encodings. I have googled for these types of problems and it seems one can tell rsync to convert file names between encodings but then I need to know for sure what encodings I have and want and sadly I don't and have no idea about how to find out 😞 There is also a Linux utility "convmv" but once again I need to somehow figure out what encoding to encode from and to and also this utility in not available in UnRAID (and I cant find it in NerdTools either)... Sadly I know about nothing about character encoding in general as well as in Linux/UnRAID in particular so not sure what I need to do in order to fix this? Anybody with expertise on character encoding that can give me some tips on how I can figure out what my current and desired encodings are and how to best try solving the problem? I have not changed anything related to character encoding in UnRAID neither in SMB, when mounting disk or in general so all is "default". I run UnRAID 6.12.2.

Yes could be the SATA cable I used... just like you say I rarely keep track of new or old rather just have a box with them and other disk accessories.... The slim SAS cable looked definitely more premium and more importantly was for sure new and not bent and folded many times as old cables may have endured over time....

I would try ASUS support for the password problem.

Happy to hear you got it working - I actually did some performance measurements and compared the 4 fixed SATA with the four over the slim SAS connector and as expected there where no difference in performance level (due to same protocol and nominal speed) but for some reason the performance variation over time in both bandwidth and latency was a little bit lower (i.e. less jitter) - perhaps there is another controller in the chipset or some other difference between them?! Anyhow I am using these slightly "better" slim SAS/SATA for my parity and largest/most used drives - this is partly a hobby project for me so making every little improvement I can think of is part of the fun 🙂