boomam

You don't need to add it back, its built in now. On a different note, anyone having any container network DNS issues after the upgrade?

Pretty seamless upgrade overall. Only thing noted is that the CPU governor reset to defaults for me.

Has anyone had any luck getting phpldapadmin working lately? Looks like the repo for its dead (last commit was 3 years ago).

Hi, I get multiple of these in my log per day - sshd[22149]: Failed publickey for root from 1.2.3.4 port 52346 ssh2: RSA SHA256:XXXXXXXXXXX/ or sshd[28177]: error: PAM: Authentication failure for root from 1.2.3.4 or Failed keyboard-interactive/pam for root from 1.2.3.4 port 52366 ssh2 alert or sshd[22149]: Read error from remote host 1.2.3.4 port 52346: Connection timed out Ordinarily not an issue, but i have certain alerts go to Telegram, so needless to say, they have become annoying. 😛 The IP referenced, is always my main computer, but I get these errors whether its on or off, so im guessing its connections being killed perhaps. Any way to either tune these out, or straight up stop them from occurring in the first place? Ordinarily I'd just adjust the sshd config/etc, but on Unraid, not sure the effect of doing so. Thanks!

...Ok, some testing with the system at Idle (running 29 containers, 2x VMs), nothing heavy. Note for formatting - "before/after" # Test 1 Governor: (on-demand/power save) Dropped 20*c in temperature, and the in0 voltage is now 736mv (368mv x2), much more in line with what I'd expect. # Test 2 Governor: on-demand Boost Performance: (yes/no) Dropped 15-20*c in temperature, in0 voltage is now 1.088v (544mv x2) at max, occasionally dropping down to 750mv. So I'm guessing on that limited testing that 'boost performance' is preventing the CPU from ramping down its voltage perhaps? Which now that I read the wiki linked in the block of text, https://wiki.unraid.net/index.php/Tips_and_Tweaks#Add_CPU_scaling_for_AMD_systems, seems to be about right....but the powernow_k8 module doesn't exist anymore according to the CLI output, so perhaps a bug elseware?

Are there any general tips for what settings to use for a Ryzen system? Seeing a lot of tips for Intel, but most AMD stuff seems to be anecdotal Also, how are we monitoring voltages for the CPU? A few people have mentioned it, but 'sensors' on the CLI doesn't seem to show (or at least explicitly named) CPU voltage? Few reports I'm reading around online seems to show "in0" is the CPU voltage, but its often 50% of actual value. Which if true, means my CPU is running at a constant 1.4v...surely not?

Thanks! 🙂

I must be looking in the wrong place, as cat'ing the files associated with the plugin in /boot shows zero commands. Where do you see the raw commands being ran?

Is there a github page anywhere with this plugin on? I'd like to have a look at the exact commands its using to see if i can diagnose whats going on.

Unfortunately not the case for me. I never use the /cache paths, always the /mnt/XYZ path, as I dont want to have to reconfigure things should I change the primary location.

Another random question - is there rhyme or reason why backups from a specific tar file are not necessarily up-to-date? For example, I added some content to BitWarden a 4 days ago. Yesterday, i had to do a restore of appdata to fix an issue from 1 day ago , so I picked a backup from 2 days ago. The data from 4 days ago was NOT in the restored BitWarden data. Is there some sort of 'modified time' variable it uses to backup perhaps?

Urgh, i dont think i can stomach that at this time of the weekend. 😛 Did a tar unzip for now, will reboot later in the week if that's indeed the fix/workaround.

Anyone else having issues doing restores? I can select a backup, press restore, confirm the warning, then nothing happens?

The environment variable "WG_DEFAULT_ADDRESS" does not work to adjust the address range in use. Is this a known issue? ##EDIT## Further testing, neither do the environment variables for DNS.

You dont need the software to use it. Just plug it into the relevant ports on your motherboard and if applicable, USB headers. Whilst the CAM software doesn't work with Unraid to my knowledge, the only thing you are missing out on is RGB control and fine tuned control of the pump - the defaults are fine though.

Do we know if the latest AGESA code/Kernel versions still have the issue with the lock up/Kernel panics?

FYI - Not sure how to expose it on Linux, but hwinfo on Windows (for my 3900x) reports 8x different temp sensors on the CPU (CPU Tctl/Tdie, CPU Die Average, CCD1, CCD2, L3 Average, L3 Cache CCD1:1, L3 Cache CCD1:2,L3 Cache CCD2:1, L3 Cache CCD2:2) - and obviously the motherboard reports its own (from the socker sensor). Shame Linux support for the HW sensors is a little behind Windows. 😞

ok, then that'll be why. The NCT6755 is a sensor on the socket, not the CPU. Technically correct for what its measuring, but not the same as the cpu die temp. I'm guessing the others, like me, are using the K10 temp (the on-die sensor), as your NCT temp is the same ball park as my NCT temp, so your temps are basically the same. 🙂 Thanks for the feedback though, helped understand the difference as i was like 'eh!? how!?' 😛

How is yours idling below everyone else's? Do you have the right sensor selected? The k10 sensor? As the other CPU sensor shows mid-30's for me too.

Thanks, difficult to tell with the 5000 series. First thoughts are always a paste problem!

Hi, I just replaced the CPU in my X570M Pro4, changing from a 3200g to a 5700g. I can no longer find the setting in the UEFI called 'Power Supply Idle Control' - anyone know if its been renamed, removed, or fixed? (Or even needed anymore with newer agesa's?) I was using it as a workaround/fix for the Ryzen kernel panic bug in unRAID/Linux Kernel. Thanks!

Hi, I just upgraded the CPU in my home server from a 3200g, which idled between 38 & 42c. This is in a Thermaltake Core V21 mATX case, stock front fan (200mm), 140mm phantec exhaust, and the Wraith Prism from a 3900x. Same setup, new thermal paste (thermal grisly), and I'm now seeing idle temps in the 52-59*c range. Turned off PBO, but not a lot of difference has been seen, if any. Pretty sure my thermal paste application is fine - same as I've done for years, thin spread across the entire IHS up-to about 1mm from the edge, then a small "half a grain of rice" blob in the middle to help fill any gaps when it gets squashed by the HSF. Normal temps for a 5700G? Or something amiss? Thanks!

Great script! I'll be modifying it at the weekend to alert me to BTRFS scrub errors/corrupt files, and was exactly what I was looking for!

Hi, Is there any magic sauce or workaround to get NFS shares to export properly on Unraid? I've been trying for the last hour to get a test share to mount from a few different systems, but each time, dependent on the export rules in place, I either get one of 'not exported', or 'unmatched host'. Currently testing with - 192.168.0.0/24(rw,sync) However I would expect the following to work too, as I've used it on a different system as a test, so I know it works. 192.168.0.0/24(rw,sync,root_squash,anonuid=1008,anongid=100) Where the anonid/anongid are the IDs for a specific local user on the host server. Thanks!

Following up - no further issues (touch wood!) have been experianced. Drive partitions are writing (and reading) without error, and the monthly (1st of each) parity check ran without error too. Next up, sending alerts from CLI...