Astaroth

Tried that already. Setting smt doesn't seem to do anything, it's ignored entirely. And trying to set MQ and other stuff fails, as it looks like you cannot set parameters for something that's created upon VM start. Would you maybe consider adding a slider button so one could switch between the interactive version and pure xml? Sure, there would be compromises, but assuming this would be aimed at people that really know what they're doing, this could work, and implementation seems to be a bit easier - you're just dumping all the responsibility on the user and simply ingesting the given config.

Hello, I have no idea how I should describe the problem because it consists of a few things. Recently, I started doing some optimizations on VMs and I noticed while I can set things like mq, rss, vCPU topology, etc. While Unraid allows to set up some things (like physical cores that a VM should use), I saw that if I select 4 cores, it will force the config to present it to the guest as 2 cores with 2 threads, no matter whether physically it's an HT pair or not. All in all, I could accept the fact that I need to change some things in the config directly, but the issue is, whatever I do, if I enter/edit the VM in the GUI, even when I don't make any changes, it will overwrite the config. Is there a way to get around that? Except for never going into VM's edit menu that is. Or maybe there's a way to edit the xml via GUI so it'll save all the custom entries, it's just that I don't see it?

Updated both card's bios & firmware. Can't find a trace of "fault" in the log. There was one reset, but it looks like it was part of standard init. Temps are visible & disk status also works fine (actually it was working before the upgrade, after restart, but I suppose it'd break again soon after). I'll be monitoring the situation. For now, thank you for your help!

Here's a fresh one. astaroth-diagnostics-20260326-1715.zip

From what I've checked now, it looks like the BIOS as well as the FW on the card may be outdated (v13 & v9 respectively, where latest are both v16), but it also says it's in IT mode.

astaroth-diagnostics-20260326-1523.zip Here you go.

Hiya, In short, I've moved to a SAS controller (LSI 9305-24i) recently as a prerequisite to adding more disks and I noticed that Unraid doesn't show whether the drive is working or spun down - the dot is always grey and it looks odd that there's hundreds of MBs of transfers to a drive that's seemingly powered down. Also the temperature reading is missing, there's just an asterisk in its place. What's weirder is that smartctl shows the temps under the same ID as the drives connected to a SATA controller (194). It's the same with reporting the current state (although when querying for a drive connected to SAS, the response is a bit sluggish). Any idea what's happening here and is it something I can fix on my own?

Recently I've switched from three SATA controllers to a single HBA card (SAS9305-24i). Saw this topic (https://forums.unraid.net/topic/138035-temperature-sensors-on-sataraid-controller-lis-sas9211-8i/) and confirmed my suspicions that there's really no support for temperature reading from SAS cards. So here's a "ghetto-plugin" that uses storcli64 to pull the data about the controller temp and displays it on the footer, as seen on the right here: Feel free to test it out. It should support multiple controllers, but I don't have even a second one to verify that. I'm not really planning to improve it (alerting and whatnot), so if someone wants to add that kind of functionality, feel free to do so. Installation is pretty simple - here I'm assuming you unzip contents to /mnt/user/share: plugin install /mnt/user/share/storcli.footer.temp.plg storcli.footer.temp-2026.03.23-x86_64-1.zip