Bug 1977798
| Summary: | RHEL8.5 guest network interface name changed after upgrade to qemu-6.0 | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux Advanced Virtualization | Reporter: | Igor Mammedov <imammedo> |
| Component: | qemu-kvm | Assignee: | Igor Mammedov <imammedo> |
| qemu-kvm sub component: | PCI | QA Contact: | Lei Yang <leiyang> |
| Status: | CLOSED ERRATA | Docs Contact: | |
| Severity: | medium | ||
| Priority: | unspecified | CC: | jinzhao, juzhang, mtessun, virt-maint |
| Version: | 8.5 | ||
| Target Milestone: | rc | ||
| Target Release: | 8.5 | ||
| Hardware: | Unspecified | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | qemu-kvm-6.0.0-26.module+el8.5.0+12044+525f0ebc | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-11-16 07:54:28 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | 6.1 |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 1957194 | ||
Fix posted upstream: "[PATCH 2/2] acpi: pc: revert back to v5.2 PCI slot enumeration" https://lore.kernel.org/qemu-devel/20210624204229.998824-3-imammedo@redhat.com/ Hi Igor I tried to reproduce the problem, but I failed. The reason is that neither qemu-kvm-6.0.0-21.module+el8.5.0+11555+e0ab0d09.x86_64 nor qemu-kvm-5.2.0-16.module+el8.4.0+11536+725e25d9.2.x86_64 supports the "pc-i440fx-5.2" machine type. So I would like to confirm a few questions with you. Could you please help me check it? 1. Since neither qemu-kvm-6.0.0-21 nor qemu-kvm-5.2.0-16 supports "pc-i440fx-5.2", Could you help confirm whether the command line is correct? 2. Does this changes affect other PCI devices? ==>Test CLI /usr/libexec/qemu-kvm \ -name 'rhel' \ -M pc-i440fx-5.2 \ -nodefaults \ -device VGA,bus=pci.0,addr=0x2 \ -m 7168 \ -smp 6,maxcpus=6,cores=3,threads=1,dies=1,sockets=2 \ -cpu 'Haswell-noTSX',+kvm_pv_unhalt \ -device usb-tablet,id=usb-tablet1,bus=usb1.0,port=1 \ -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \ -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/kvm_autotest_root/images/rhel850-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \ -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \ -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \ -device pci-bridge,chassis_nr=1,id=pci.1,bus=pci.0,addr=0x3 \ -device virtio-net-pci,id=nic1,bus=pci.1,addr=0x1 \ -device virtio-net-pci,id=nic2,bus=pci.1,addr=0x2 \ -device virtio-net-pci,id=nic3,bus=pci.1,addr=0x3 \ -vnc :0 \ -rtc base=utc,clock=host,driftfix=slew \ -boot menu=off,order=cdn,once=c,strict=off \ -enable-kvm \ ==>Error info: qemu-kvm: -M pc-i440fx-5.2: unsupported machine type Use -machine help to list supported machines Additional info: I try to use the latest qemu packages: qemu-kvm-6.0.0-22.module+el8.5.0+11677+542c8d77.x86_64. Use -machine help to list supported machines: # /usr/libexec/qemu-kiwi -machine help Supported machines are: pc-i440fx-4.2 Standard PC (i440FX + PIIX, 1996) (deprecated) pc-i440fx-2.11 Standard PC (i440FX + PIIX, 1996) (deprecated) pc RHEL 7.6.0 PC (i440FX + PIIX, 1996) (alias of pc-i440fx-rhel7.6.0) pc-i440fx-rhel7.6.0 RHEL 7.6.0 PC (i440FX + PIIX, 1996) (default) pc-i440fx-rhel7.5.0 RHEL 7.5.0 PC (i440FX + PIIX, 1996) pc-i440fx-rhel7.4.0 RHEL 7.4.0 PC (i440FX + PIIX, 1996) pc-i440fx-rhel7.3.0 RHEL 7.3.0 PC (i440FX + PIIX, 1996) pc-i440fx-rhel7.2.0 RHEL 7.2.0 PC (i440FX + PIIX, 1996) pc-i440fx-rhel7.1.0 RHEL 7.1.0 PC (i440FX + PIIX, 1996) pc-i440fx-rhel7.0.0 RHEL 7.0.0 PC (i440FX + PIIX, 1996) Machine type 'pc-i440fx-4.2' is deprecated: Not supported by RHEL Best Regards Lei The acpi-index feature is available only on PC machine so far, so for testing on RHEL you can use the latest available PC machine type (i.e. pc-i440fx-rhel7.6.0). Hi Igor I tried to test on RHEL the latest available PC machine(pc-i440fx-rhel7.6.0), I can reproduced this bug. But the premise is that the pci-bridge binding method is used for the nic in the command line. ==Test Steps 1. Install qemu-kvm-5.2.0-11.module+el8.4.0+10268+62bcbbed.x86_64 and boot a guset /usr/libexec/qemu-kvm \ -name 'avocado-vt-vm1' \ -sandbox on \ -machine pc-i440fx-rhel7.6.0 \ -nodefaults \ -device VGA,bus=pci.0,addr=0x2 \ -m 6144 \ -smp 6,maxcpus=6,cores=3,threads=1,dies=1,sockets=2 \ -cpu 'Haswell-noTSX',+kvm_pv_unhalt \ -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \ -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/rhel850-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \ -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \ -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \ -vnc :1 \ -rtc base=utc,clock=host,driftfix=slew \ -boot menu=off,order=cdn,once=c,strict=off \ -monitor stdio \ -device pci-bridge,chassis_nr=1,id=pci.1,bus=pci.0,addr=0x3 \ -device virtio-net-pci,id=nic1,netdev=hostnic1,bus=pci.1,addr=0x1 \ -netdev tap,id=hostnic1,vhost=on \ -device virtio-net-pci,id=nic2,netdev=hostnic2,bus=pci.1,addr=0x2 \ -netdev tap,id=hostnic2,vhost=on \ -device virtio-net-pci,id=nic3,netdev=hostnic3,bus=pci.1,addr=0x3 \ -netdev tap,id=hostnic3,vhost=on \ 2.Check guest log # journalctl |grep virtio virtio_net virtio1 ens1: renamed from eth0 virtio_net virtio3 ens3: renamed from eth2 virtio_net virtio2 ens2: renamed from eth1 3. shutdown guest 4. Install qemu-kvm-6.0.0-21.module+el8.5.0+11555+e0ab0d09.x86_64 and boot guest again 5. Check the guest log # journalctl |grep virtio virtio_net virtio1 ens1: renamed from eth0 virtio_net virtio3 ens3: renamed from eth2 # journalctl |grep exists systemd-udevd[496]: Error changing net interface name 'eth1' to 'ens3': File exists systemd-udevd[496]: could not rename interface '3' from 'eth1' to 'ens3': File exists systemd-udevd[778]: Error changing net interface name 'eth1' to 'ens3': File exists systemd-udevd[778]: could not rename interface '3' from 'eth1' to 'ens3': File exists Additional info: If QE does not use the pci-bridge method to add the nics, but uses the following command line to execute the above steps again, the bug will not reproduce. qemu CLI: /usr/libexec/qemu-kvm \ -name 'avocado-vt-vm1' \ -sandbox on \ -machine pc-i440fx-rhel7.6.0 \ -nodefaults \ -device VGA,bus=pci.0,addr=0x2 \ -m 6144 \ -smp 6,maxcpus=6,cores=3,threads=1,dies=1,sockets=2 \ -cpu 'Haswell-noTSX',+kvm_pv_unhalt \ -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \ -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/rhel850-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \ -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \ -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \ -vnc :1 \ -rtc base=utc,clock=host,driftfix=slew \ -boot menu=off,order=cdn,once=c,strict=off \ -monitor stdio \ -device virtio-net-pci,mac=9a:61:1f:b8:cd:a2,id=id5r9s2h,netdev=idJ7MN1Z,bus=pci.0,addr=0x5 \ -netdev tap,id=idJ7MN1Z,vhost=on \ -device virtio-net-pci,mac=9a:61:1f:b8:cd:a3,id=idYzDe0T,netdev=idkzcAZj,bus=pci.0,addr=0x6 \ -netdev tap,id=idkzcAZj,vhost=on \ -device virtio-net-pci,mac=9a:61:1f:b8:cd:a4,id=id7UK0tA,netdev=idh0Vjk8,bus=pci.0,addr=0x7 \ -netdev tap,id=idh0Vjk8,vhost=on \ Best Regards Lei upstream commit 7193d7cdd9 acpi: pc: revert back to v5.2 PCI slot enumeration Martin, can you give release + dev-ack flags to this bz, pls? QE bot(pre verify): Set 'Verified:Tested,SanityOnly' as gating/tier1 test pass. ==Test Steps 1. Install qemu-kvm-5.2.0-16.module+el8.4.0+11923+e8b883e4.4.x86_64 and boot a guset /usr/libexec/qemu-kvm \ -name 'avocado-vt-vm1' \ -sandbox on \ -machine pc-i440fx-rhel7.6.0 \ -nodefaults \ -device VGA,bus=pci.0,addr=0x2 \ -m 6144 \ -smp 6,maxcpus=6,cores=3,threads=1,dies=1,sockets=2 \ -cpu 'Haswell-noTSX',+kvm_pv_unhalt \ -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \ -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/rhel850-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \ -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \ -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \ -vnc :1 \ -rtc base=utc,clock=host,driftfix=slew \ -boot menu=off,order=cdn,once=c,strict=off \ -monitor stdio \ -device pci-bridge,chassis_nr=1,id=pci.1,bus=pci.0,addr=0x3 \ -device virtio-net-pci,id=nic1,netdev=hostnic1,bus=pci.1,addr=0x1 \ -netdev tap,id=hostnic1,vhost=on \ -device virtio-net-pci,id=nic2,netdev=hostnic2,bus=pci.1,addr=0x2 \ -netdev tap,id=hostnic2,vhost=on \ -device virtio-net-pci,id=nic3,netdev=hostnic3,bus=pci.1,addr=0x3 \ -netdev tap,id=hostnic3,vhost=on \ 2.Check guest log # journalctl |grep virtio Aug 02 15:47:06 localhost.localdomain kernel: virtio_net virtio1 ens1: renamed from eth0 Aug 02 15:47:06 localhost.localdomain kernel: virtio_net virtio3 ens3: renamed from eth2 Aug 02 15:47:06 localhost.localdomain kernel: virtio_net virtio2 ens2: renamed from eth1 3. shutdown guest 4. Install qemu-kvm-6.0.0-21.module+el8.5.0+11555+e0ab0d09.x86_64 and boot guest again 5. Check the guest log # journalctl |grep virtio Aug 02 16:22:23 localhost.localdomain kernel: virtio_net virtio1 ens1: renamed from eth0 Aug 02 16:22:23 localhost.localdomain kernel: virtio_net virtio2 ens3: renamed from eth1 # journalctl |grep exists Aug 02 16:22:23 localhost.localdomain systemd-udevd[486]: Error changing net interface name 'eth2' to 'ens3': File exists Aug 02 16:22:23 localhost.localdomain systemd-udevd[486]: could not rename interface '4' from 'eth2' to 'ens3': File exists Aug 02 16:22:27 localhost.localdomain systemd-udevd[780]: Error changing net interface name 'eth2' to 'ens3': File exists Aug 02 16:22:27 localhost.localdomain systemd-udevd[780]: could not rename interface '4' from 'eth2' to 'ens3': File exists ==Reproduced with qemu-kvm-6.0.0-21.module+el8.5.0+11555+e0ab0d09.x86_64 ==Verified with qemu-kvm-6.0.0-26.module+el8.5.0+12044+525f0ebc.x86_64 1. Install qemu-kvm-6.0.0-26.module+el8.5.0+12044+525f0ebc.x86_64 and boot guest /usr/libexec/qemu-kvm \ -name 'avocado-vt-vm1' \ -sandbox on \ -machine pc-i440fx-rhel7.6.0 \ -nodefaults \ -device VGA,bus=pci.0,addr=0x2 \ -m 6144 \ -smp 6,maxcpus=6,cores=3,threads=1,dies=1,sockets=2 \ -cpu 'Haswell-noTSX',+kvm_pv_unhalt \ -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \ -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/rhel850-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \ -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \ -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \ -vnc :1 \ -rtc base=utc,clock=host,driftfix=slew \ -boot menu=off,order=cdn,once=c,strict=off \ -monitor stdio \ -device pci-bridge,chassis_nr=1,id=pci.1,bus=pci.0,addr=0x3 \ -device virtio-net-pci,id=nic1,netdev=hostnic1,bus=pci.1,addr=0x1 \ -netdev tap,id=hostnic1,vhost=on \ -device virtio-net-pci,id=nic2,netdev=hostnic2,bus=pci.1,addr=0x2 \ -netdev tap,id=hostnic2,vhost=on \ -device virtio-net-pci,id=nic3,netdev=hostnic3,bus=pci.1,addr=0x3 \ -netdev tap,id=hostnic3,vhost=on \ 2.Check guest log # journalctl |grep virtio Aug 02 16:28:03 localhost.localdomain kernel: virtio_net virtio3 ens3: renamed from eth2 Aug 02 16:28:03 localhost.localdomain kernel: virtio_net virtio2 ens2: renamed from eth1 Aug 02 16:28:03 localhost.localdomain kernel: virtio_net virtio1 ens1: renamed from eth0 3.Based on above, move to "VERIFIED". Best Regards Lei 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:av 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-2021:4684 |
Description of problem: Commit [1] moved _SUN variable from only hot-pluggable to all devices. This made linux kernel enumerate extra slots that weren't present before. If extra slot happens to be be enumerated first and there is a device in th same slot but on other bridge, linux kernel will add -N suffix to slot name of the later, thus changing NIC name compared to QEMU 5.2. This in some case confuses systemd, if it is using SLOT NIC naming scheme and interface name becomes not the same as it was under QEMU-5.2. 1) b7f23f62e40 (pci: acpi: add _DSM method to PCI devices) Version-Release number of selected component (if applicable): qemu-kvm-6.0.0-21 How reproducible: 100% Steps to Reproduce: use rhel8 guest with following CLI -M pc-i440fx-5.2 -nodefaults \ -device pci-bridge,chassis_nr=1,id=pci.1,bus=pci.0,addr=0x3 \ -device virtio-net-pci,id=nic1,bus=pci.1,addr=0x1 \ -device virtio-net-pci,id=nic2,bus=pci.1,addr=0x2 \ -device virtio-net-pci,id=nic3,bus=pci.1,addr=0x3 Actual results: v6.0: kernel: virtio_net virtio0 ens1: renamed from eth0 kernel: virtio_net virtio2 ens3: renamed from eth2 systemd-udevd[299]: Error changing net interface name 'eth1' to 'ens3': File exists systemd-udevd[299]: could not rename interface '3' from 'eth1' to 'ens3': File exists Expected results: v5.2: kernel: virtio_net virtio0 ens1: renamed from eth0 kernel: virtio_net virtio2 ens3: renamed from eth2 kernel: virtio_net virtio1 enp1s2: renamed from eth1 Additional info: 5.2 (slot 2 is assigned to empty bus 0 slot and virtio1 is assigned to 2-2 slot, and renaming falls back, for some reason, to path based naming scheme) 6.0 (with commit [1] kernel assigns virtio2 to 3-2 slot since bridge advertises _SUN=0x3 and kernel assigns slot 3 to bridge. Still it manages to rename virtio2 correctly to ens3, however systemd gets confused with virtio1 where slot allocation exactly the same (2-2) as in 5.2 case and tries to rename it to ens3 which is rightfully taken by virtio2)