Vexz

The 1GB RAM limitations have been removed and of course after the reboot eveything is back to normal again. But will it stay that way? I think only time will tell. nas-diagnostics-20260611-1600.zip

Greetings. I woke up today and saw that my unRAID server suddenly has about 86% RAM usage, which is very unusual. I've been using unRAID for almost 1,5 years now and never had this happen, so I was surprised to see this. Usually the RAM usage is about 42%. So I asked the AI and it told me to execute a "free -h", which displays 16 Gi in the "buff/cache" column. It said everything is fine and it just means that my RAM is used efficiently. But we all know you can't blindly trust AI and I've never had this happen before, so I'm very skeptical. Would love to know what's actually happening here. Diagnostics attached. Thanks in advance. nas-diagnostics-20260611-0808.zip

Ich weiß nicht, ob es dafür (bereits) eine offizielle Lösung gibt, ich konnte zumindest keine finden, aber es ist zumindest eine, die man einfach, schnell und sicher durchführen könnte. Vielleicht wird es noch irgendwann eine extra Option dafür geben, unRAID von einer SSD auf eine Andere zu migrieren.

Freut mich, dass du es hingekriegt hast. Für das nächste Mal, sollte etwas schief gehen, mach am besten immer ein Backup von deinem unRAID System, damit du nicht nochmal alles neu installieren und konfigurieren musst. Im Grunde reicht es auch schon den Pfad /boot/config zu sichern. Da ist alles drin, um mit dem Unraid USB Flash Creator Tool dein unRAID wieder genau so wiederherstellen zu können, wie es vorher war.

Ist möglicherweise nicht die schönste Lösung, aber du könntest einen USB Stick in deinen Server einstecken, das unRAID System auf diesen migrieren und dann von da aus auf die Optane SSD.

You probably missed my edit of the OP, right at the top. I fixed it and everything is working fine now.

Update I fixed it with a restore of a backup on a new USB device and deleted all partitions via console on the SSD first. I just upgraded to 7.3.1 and installed an SSD before doing so to migrate the system to the SSD after the update. While going through the process I encountered this error: ➜ Starting configuration... ➜ Starting internal boot setup. This might take a while... ➜ Configuring internal boot pool... ✗ Internal boot setup returned an error: assignableDisks snapshot: [{"id":"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T","serialNum":"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T","device":"/dev/nvme0n1"}] assignableDisks resolved serialNum->device: [["Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T",{"bootId":"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T","devicePath":"/dev/nvme0n1"}]] Running: emcmd debug=cmdCreatePool,cmdAssignDisk,cmdMakeBootable Running: emcmd cmdCreatePool=apply&poolName=boot&poolSlots=1 mkbootpool: command failed or timed out emcmd request failed with status 500 Technical details { "request": { "operation": "CreateInternalBootPool", "variables": { "poolName": "boot", "devices": [ "Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T" ], "bootSizeMiB": 0, "updateBios": true, "reboot": false } }, "response": { "ok": false, "code": 1, "output": "assignableDisks snapshot: [{\"id\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"serialNum\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"device\":\"/dev/nvme0n1\"}]\nassignableDisks resolved serialNum->device: [[\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",{\"bootId\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"devicePath\":\"/dev/nvme0n1\"}]]\nRunning: emcmd debug=cmdCreatePool,cmdAssignDisk,cmdMakeBootable\nRunning: emcmd cmdCreatePool=apply&poolName=boot&poolSlots=1\nmkbootpool: command ...[truncated]" }, "status": 1, "error": { "name": "Error", "code": 1, "message": "Internal boot setup returned ok=false", "category": "unknown", "networkError": { "status": 1, "result": { "ok": false, "code": 1, "output": "assignableDisks snapshot: [{\"id\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"serialNum\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"device\":\"/dev/nvme0n1\"}]\nassignableDisks resolved serialNum->device: [[\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",{\"bootId\":\"Samsung_SSD_970_EVO_Plus_500GB_S4EVNJ0N234529T\",\"devicePath\":\"/dev/nvme0n1\"}]]\nRunning: emcmd debug=cmdCreatePool,cmdAssignDisk,cmdMakeBootable\nRunning: emcmd cmdCreatePool=apply&poolName=boot&poolSlots=1\nmkbootpool: command ...[truncated]" } } } } ➜ Updating Server Identity to NAS... ✓ Server Identity updated. ➜ Finalizing setup...I didn't reboot yet, I think the system will not be able to start if I do it. Help is appreciated.

Just upgraded from 7.1.3 to 7.2.0. Everything is working. Thank you!

So safe to ignore, if you don't run VMs inside unRAID?

@tobasium @SvensenDE @nicx Update! I just tested this (using unRAID 7.0.0) and found out that you do NOT need that part with the "/etc/rc.conf". I never created that file and it didn't exist for me, ergo I never wrote something inside that file. The part with the "/etc/rc.conf" was completely skipped. So after you install the microcode with un-get, check the contents of your flash drive in "/boot". Depending on what file you see there (the "intel-ucode.cpio" OR the "intel-ucode.img") you just need to change the part in the Syslinux Configuration in the web UI. In my case the Syslinux Configuration looks like this, because it downloaded the "intel-ucode.img": kernel /bzimage append initrd=/intel-ucode.img,/bzroot After that, just reboot your unRAID server and the microcode update will be applied. Result after the reboot: root@NAS:~# dmesg | grep microcode [ 0.000000] microcode: updated early: 0xe -> 0x1c, date = 2024-09-06 [ 1.975603] microcode: Microcode Update Driver: v2.2. Before the reboot "dmesg | grep microcode" wouldn't give me any output, so there definitely is a change and the patch was applied.

I know now how to find out and it was much easier than I thought after reading the original post again. In the third line of the "/etc/rc.conf" you point to the specific file. So before you reboot, you just check the contents of your unRAID USB drive. I can see the "intel-ucode.img" there, so I know I'd need to use this, not the CPIO-file.

@SvensenDE Then how did you know you needed to enter "intel-ucode.img"? How did you get to know about this file? I mean you must have found out about this somehow and the system must store this file somewhere.

Is there a way to find out which of the two it installed BEFORE you reboot? I'd rather not run into that issue. Would be a big hassle to solve this, because my unRAID NAS is not connected to a monitor (since there is no monitor in that room) and the USB drive is plugged in a USB port directly on the motherboard and cannot be removed, unless I remove the motherboard first.

@etsi @anhvo Works for me on unRAID 7. I downloaded the ITE IT87 Driver, then installed the Dynamix Auto Fan Control. Dynamix Auto Fan Control is a little confusing, but I managed to get it to work somehow. Set "PWM controller" to "it8613 - pwm3". Set "Fan control function" to "Enabled". Now click on the "DETECT" button next to "PWM fan". It might take a while and you might need to click on it twice. It sets the following value: "/sys/devices/platform/it87.2624/hwmon/hwmon2/fan3_input". "Minimum PWM value" is a little confusing. I set it to "100", which results in 1245 rpm. The value "120" gives me about 1457 rpm. You might wanna play around with the values here. Set low and high temperatures to your liking. I chose 35°C for low and 40°C for high. I think you should set the "Refresh interval (minutes)" value to "1", but that's just what I like. Save. Don't be confused if it says "Disabled" for "Fan control function" after saving. It does that, because it shows you the settings for the "it8613 - pwm2" fan controller now. You want that one to be disabled. In fact, you only want "it8613 - pwm3" to be enabled. Choose the "it8613 - pwm3" from the list and you will see that "Fan control function" will say "Enabled". The setting "Fan control function" doesn't enable or disable the fan control function in general - it just sets it for the PWM controller you selected from the list in "PWM controller". I don't know why the devs did this, because it's just confusing. Thanks for this thread. Two of my drives were constantly at 45°C. Now they are at 37°C.

I don't know where to check that, but it must have been my misinterpretation. It was my guess of what happens there, since it took so long. Just ignore it. Thank you very much for your help.