dopeytree

Is it possible to add: fatrace - https://slackbuilds.org/repository/15.0/system/fatrace/ blktrace - https://slackbuilds.org/repository/15.0/system/blktrace/ Both help trace processes that are spinning up disks.

Did you find a way to install this?

Apparently the way on linux you can check which process is making a disk call is with 'fatrace' & or 'blktrace' but not sure how you'd install those packages to unraid. https://forum.openmediavault.org/index.php?thread/43919-fatrace-find-out-which-process-is-accessing-which-filepath-and-causing-your-hdd/#

Ah thanks yeah looks like I must have recently misconfigured it. All the excluded folders were set to include 🙈

To work out bandwidth you must look deep within at how you plan to run it. i.e - '1x minisas --> 4x sata cables' or '1x minisas to 1x minisas' into a backplane that has 12x drives? What's the most number of drives to be connected & are they spinners or sata ssds? i.e - Max read speed in MB of each drive X the number of drives = your maximum needed bandwidth. Then you see what is practical i.e if you want to keep a graphics card you wouldn't use that slot etc. Enjoy the new toy.

Ok noted but how do you find out what is making the spinup call? the file activity & open files plugins are for files accessed & read and as such not disks that are spun up and not read.

Anyone else getting random disk spin up to read smart status of drives?

Hi am noticing something keeps spinning up my disk to do smart health checks. Or they are being spun up and then doing a health check. No read happens on drives so I can't use the file activity plugin or the open files plugin to narrow down.. Is there a way I can turn the frequency of smart checks down? It doesn't need to start up a sleeping pool.. i.e merlin Could be a 6.12.8 bug? Have removed disk location, zfs master etc incase it's a plugin but doesnt seem to be. Its not plex as plex doesnt touch the merlin drive pool. Diagnostics attached. Thanks moulin-rouge-diagnostics-20240312-2336.zip

Well the solution is to write one's own code to do the task - as the existing solution doesn't function properly.

Friends nearly everyone commenting here already has unraid and thus free updates FOR LIFE™. So please just let new folks choose to either pay a yearly fee or they can choose to pay a higher fee to upgrade to lifetime. When we build 2nd servers we can also make the same decisions. If no one pays then unraid won't get any new features...

Have you seen the openfancontroller? It may be a better way to go? https://sasakaranovic.com/projects/openfan-controller/ Available on the unraid app store.

Hey is there a newer version of this code to turn off the screen? I've got a usb crystalfontz off ebay. not getting anything with the quoted code. My device is on: /dev/ttyUSB1

Sounds good at the moment I'm using a manual fan controller with a dial I turn but with your designed software & hardware I could automate this with some scripts. Awesome project for me to get ready for the warmer summer months.

Realtek network drivers has been fecked over the latest LINUX KERNAL because someone reported a bug. So it is upstream of any linux based OS. Once / IF resolved the fix should come back downstream to us all. Until then you have to manually enable ASPM.

This looks perfect!! It wont let me install without giving it access to a controller (which I don't have yet) so just quick question - Is this able to access all server sensors? so all HD temps & CPU / motherboard temps too? Rather than solely external sensors on the openFan controller?

I was playing with this plugin the other day before a new build. It seems to work but does the bug where it shows blank. Also it includes any unassigned drives and I couldn't quite workout the code to exclude them. It doesnt seem to activate itself after a reboot. So you seem to have to go in choose the PWM fan and click apply again. Then the fan should spin down from your bios default 100%. The plugin definitely needs some work. I've changed tract and am just using bios and a manual fan controller at 50%. If you do want to control from within unraid I'd suggest maybe using a usb controller and liquid CTL instead of this plugin

Hi just trying to understand SR-IOV... If you use this does it still need to be GPU passthrough'ed? I.E gpu is then solely available for VM's & not available for docker containers?

Hmm thanks yeah turbo seems to turn back on with this command as now boosting back up past 2.6Ghz which is the turbo off limit.

CPU Intel 13500H The issue is if I turn turbo off then turn back on it stays off until a reboot. It's a minor issue but have seen it on 2x Intel cpus. First one is a 11900T & this one is a 13500H. Does anyone else experience this? The new CPU is very good at stepping down so not such an issue.

Is this being tracked? The connect plugin just needs to occasionally clean up the .git folder on the usb stick.

Not sure if aware but ZFS in the main array is kind of pointless as no bit rot protection. It will let you know there is a problem but cannot fix it (where as zfs pools can). You can do ZFS receive tho to receive snapshots from a ZFS pool. I keep trying to add this info to the documentation but no one is merging it so..

powersave & performance

There's an AI category in the unraid 'app store' so this can now be added to the template.

New 13th Gen cpu on 6.12.8 - I can disable turbo but when toggle it back to ON doesn't yield a result without a reboot. Any ideas?

Why would you loose everything on a ZFS pool? Maybe if you only use a stripe setup with no redundancy but a normal z1 or z2 (parity) should provide the ability to heal.