dmacias

I'll look into it. It could just be a nut problem with this ups.

What, no X10s? I'll see if I can implement the ole hi, med, low I had started before for the X9s

That sucks. Where'd you get that info or did you test on an X9SCM-F or similar?

First make sure your BIOS and BMC are on the latest firmware. Are your fans 4pin pwm fans? For missing FANA, check the Settings page under Display settings and Global available sensors and make sure FANA is checked. You also need to use the Sensor Config editor to change your fan thresholds. Check back through the thread around 2018/05/20 is when I implemented fan control for X9 boards.

I assume you have it shutdown based on time on battery. What time did you set it to? If you have a monitor hooked up, watch the console when that time expires and see if it's starting shutdown.

I updated a few c compiler packages and added a few. Also updated glibc to add a couple missing files. The files with o extensions (e.g. crtl.o) in /usr/lib64. You'll need to deselect glibc, apply then reselect and apply

Sorry for the late reply. I would set FAN1234 to CPU temp and FANA to hard drive temps.

I was refering to adding an override to the plugin settings. Latest update provides an override for virtual machines.

I think I got it fixed. Updated plugin and added delay after full fans. I think the first fan control commands were happening too soon after full fan commands and not taking. So fan speeds weren't changing on script start and eventually changing on the next poll when the temp changed.

Thanks for the info. I looked back through the releases and lbzip2 was included in 6.6.0. I have removed it from the 6.6 repo

I was wondering if your BRNAME was actually capitalized as BR0? It may have an affect on the python script and command line. While unRAID is case sensitive, maybe the python script doesn't handle it well. You had mentioned earlier running the command line and using br0 and it worked.

Interesting. My X10 and now X11 would only take multiples of 100. Have you checked if your bios is up to date and bmc? Just curious if that's the norm for the X9 boards.

I'll update in a little bit. I fixed the error reporting. But here's the command. You'll have to edit the file manually too usr/sbin/ipmi-sensors-config --filename=/boot/config/plugins/ipmi/ipmi-sensors.config --commit

I cannot find any emacs with no X available for slackware

Sorry this requires too many packages for this plugin

For whatever reason I could never get the writing of a bad config to come back as an error. It's not saving because it doesn't like the value you are using for the lower threshold. If you were to run the command from the command line you'd see the complaint. What values have you tried? Try only numbers like 300, 400, 500

You can add this to the plg install chmod +0755 /usr/local/emhttp/plugins/dynamix.cache.dirs/scripts/* Or create an "install" directory within the dynamix.cache.dirs.txz package and place a doinst.sh file with that command in the install directory. That's the way all slackware packages handle permissions and other install functions. E.g. This is from NerdPack doinst.sh #!/bin/shln -sf /usr/local/emhttp/plugins/NerdPack/scripts/packagemanager /usr/sbin/packagemanagerchmod +0755 /usr/sbin/packagemanager

virbr0 would not work. The script monitors all packets looking for wol packets on the bridged network for your VMs. It gets the bridge name from BRNAME in /boot/config/domain.cfg. Do you have that variable in the domain.cfg? Maybe that variable isn't set anymore? Do you have bridging enabled in network settings?

Is the VM stopped or paused? It has to be shutdown or hibernated. It can't be sleeping or in any other state. I would stop the service. Then run "ps aux | grep libvirtwol". To make sure another instance isn't running then start the service and run that again. You can also run manually /etc/rc.d/rc.libvirtwol start/stop. You should see /usr/bin/python /usr/local/emhttp/plugins/libvirtwol/scripts/libvirtwol.py br0 running. Edit: also is your bridge named BR0 or br0? Could make a difference. The service gets this from /boot/config/domain.cfg

I don't remember if any other y cables came with the noctuas but the one I'm using only senses rpm on one of the fans. I think it's this cable: Noctua NA-YC1 Fan 4-Pin Y-Cable. But that's plugged into a 4-pin pwm fan header. Definately would get another fan for the lower intake.

What version of unRAID?

Glad you figured out your noise problem. Besides being in the closet, my unifi router and switch drown out the server. That was the exact reason I started working on my IPMI plugin. Besides getting accurate sensor readings, I wanted to control the hdd fans based on hdd temp. You might try the dynamix auto fan control. Your board may be supported by lm-sensors

I'll look at updating python but I'm usually hesitant since it usually breaks things. I don't compile those myself and sometimes they are compiled using different packages than are available in unRAID. The path for python 2 is /usr/lib64/python2.7. I believe 3 is similar and all packages from pip I believe are in the sub directory site-packages. It's all non persistent. So if you reboot its gone till it's all installed again.

Same noise level? Maybe with some sort of fan control it would be tolerable for you under normal everyday use. Just loud during parity checks. Also did you have the filters on or off? When I had the DS380, I did a lot of testing with and without.

Correct need IPMI for the plugin. I added newer pics in the OP to show how I have it now. I used Corsair type 3 (same as type 4) for the peripheral power and an extender kit for the main atx cable and for the motherboard/cpu power cable. I still think you might get even better cooling results from flipping like I did. That way power suppy, system and hdds are all drawing from their own external air. Edit. I can remember for sure but I believe I have all drives sequentially from top to bottom in the case. With parity on the top