CaptainSandwich

Did you ever find out the answer to this question? I'm looking to setup a script to set nvidia driver persistence mode when a VM shuts down, so the gpu can enter low power state. Thanks. EDIT - looks like /etc/libvirt/hooks/qemu is the one you want to edit. I modified the existing script to the below to set nvidia persistence mode when a VM releases all resources. The script update appears to persist between boots. #!/usr/bin/env php <?php if ($argv[2] == 'release' && $argv[3] == 'end'){ shell_exec('date +"%b %d %H:%M:%S libvirt hook: Setting nVidia Promiscuous mode to 1" >> /var/log/syslog'); shell_exec('nvidia-smi --persistence-mode=1'); } if (!isset($argv[2]) || $argv[2] != 'start') { exit(0); } $strXML = file_get_contents('php://stdin'); $doc = new DOMDocument(); $doc->loadXML($strXML); $xpath = new DOMXpath($doc); $args = $xpath->evaluate("//domain/*[name()='qemu:commandline']/*[name()='qemu:arg']/@value"); for ($i = 0; $i < $args->length; $i++){ $arg_list = explode(',', $args->item($i)->nodeValue); if ($arg_list[0] !== 'vfio-pci') { continue; } foreach ($arg_list as $arg) { $keypair = explode('=', $arg); if ($keypair[0] == 'host' && !empty($keypair[1])) { vfio_bind($keypair[1]); break; }

I've finally managed to get my setup to a stage where the disks in the array will spin down pretty consistently. But I think I'm now running into an issue with Fan Control. There are 2 SSD's not impacted by the array fan; the cache drive has been excluded from Auto Fan Control however an NVME drive is mounted via unassigned devices and does not show up in the list for exclusion. However, the log seems to indicate that when all the array drives have spun down, Auto Fan Control then refers to the NVME's temperature to determine fan speed, which means fan speed and noise increases with the array spun down. Is there any way to exclude drives mounted in unassigned devices from auto fan control?