Description of problem: Configure the host with the libvirt split modular daemons, instead of libvirtd https://libvirt.org/daemons.html#switching-to-modular-daemons In this scenario no daemon is running initially. They are launched on-demand when a client connects to the socket. The initialization of devices happens in the background though, so if an API call is made very quickly, many devices won't be known yet. This can be seen by runing $ virsh -c nodedev:///system nodedev-list computer It only shows one or two devices. Wait 1 second and run it again, and you'll see all devices. Version-Release number of selected component (if applicable): libvirt-6.0.0-10.el8 How reproducible: Semi-reliable Steps to Reproduce: 1. Ensure virtnodedevd.service is *NOT* currently active 2. Ensure virtnodedevd.socket IS currently active 3. Run virsh -c nodedev:///system nodedev-list Actual results: Only 1 or 2 devices are shown Expected results: All devices are shown. Additional info:
Fix posted at https://www.redhat.com/archives/libvir-list/2020-March/msg00491.html
Fixed in 6.2.0 with commit 008abeb03c262149b756ad5a226ff6cbc5e37e2c Author: Daniel P. Berrangé <berrange> Date: Fri Mar 13 11:53:25 2020 +0000 nodedev: fix race in API usage vs initial device enumeration
I can reproduce it with the steps in comment 0 with libvirt-6.0.0-18 And test on libvirt-libs-6.3.0-1.module+el8.3.0+6478+69f490bb.x86_64, no such issue. The bug is fixed. # systemctl stop libvirtd # systemctl start virtnodedevd.socket # virsh -c nodedev:///system nodedev-list block_sda_36d09466077a944002306f0d40902561f computer drm_card0 net_eno1_20_04_0f_ea_f1_7c net_eno2_20_04_0f_ea_f1_7d net_eno3_20_04_0f_ea_f1_7e net_eno4_20_04_0f_ea_f1_7f net_enp6s0f0_f8_f2_1e_30_df_50 ....
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (virt:8.3 bug fix and enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2020:5137