<div dir="ltr"><div><div><div><div><div>I have a reoccurring problem with VM's failing to start. In the task log I'l see this:<br><br>kvm: -netdev type=tap,id=net0,ifname=tap721i0,script=/var/lib/qemu-server/pve-bridge,downscript=/var/lib/qemu-server/pve-bridgedown,vhost=on: tap: open vhost char device failed: Cannot allocate memory<br>kvm: -netdev type=tap,id=net0,ifname=tap721i0,script=/var/lib/qemu-server/pve-bridge,downscript=/var/lib/qemu-server/pve-bridgedown,vhost=on: Device 'tap' could not be initialized<br>TASK ERROR: start failed: command '/usr/bin/kvm -id 721 -chardev 'socket,id=qmp,path=/var/run/qemu-server/721.qmp,server,nowait' -mon 'chardev=qmp,mode=control' -vnc unix:/var/run/qemu-server/721.vnc,x509,password -pidfile /var/run/qemu-server/721.pid -daemonize -smbios 'type=1,uuid=5cd965b8-ee6c-4a23-a705-589329833829' -name VM721 -smp '4,sockets=1,cores=4,maxcpus=4' -nodefaults -boot 'menu=on,strict=on,reboot-timeout=1000' -vga qxl -no-hpet -cpu 'kvm64,hv_spinlocks=0xffff,hv_relaxed,+lahf_lm,+x2apic,+sep' -m 2048 -k en-us -device 'qxl,id=vga1,ram_size=67108864,vram_size=33554432,bus=pci.0,addr=0x18' -device 'AC97,addr=0x18' -device 'piix3-usb-uhci,id=uhci,bus=pci.0,addr=0x1.0x2' -readconfig /usr/share/qemu-server/pve-usb.cfg -chardev 'spicevmc,id=usbredirchardev0,name=usbredir' -device 'usb-redir,chardev=usbredirchardev0,id=usbredirdev0,bus=ehci.0' -chardev 'spicevmc,id=usbredirchardev1,name=usbredir' -device 'usb-redir,chardev=usbredirchardev1,id=usbredirdev1,bus=ehci.0' -chardev 'spicevmc,id=usbredirchardev2,name=usbredir' -device 'usb-redir,chardev=usbredirchardev2,id=usbredirdev2,bus=ehci.0' -chardev 'spicevmc,id=usbredirchardev3,name=usbredir' -device 'usb-redir,chardev=usbredirchardev3,id=usbredirdev3,bus=ehci.0' -spice 'tls-port=61004,addr=127.0.0.1,tls-ciphers=DES-CBC3-SHA,seamless-migration=on' -device 'virtio-serial,id=spice,bus=pci.0,addr=0x9' -chardev 'spicevmc,id=vdagent,name=vdagent' -device 'virtserialport,chardev=vdagent,name=com.redhat.spice.0' -device 'virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3' -iscsi 'initiator-name=iqn.1993-08.org.debian:01:46185e44b9c' -drive 'file=/mnt/cephfs/images/721/vm-721-disk-1.qcow2,if=none,id=drive-virtio1,format=qcow2,cache=writeback,aio=native,detect-zeroes=on' -device 'virtio-blk-pci,drive=drive-virtio1,id=virtio1,bus=pci.0,addr=0xb,bootindex=100' -drive 'file=/mnt/pve/ISO/template/iso/virtio-win-0.1-100.iso,if=none,id=drive-ide0,media=cdrom,aio=native' -device 'ide-cd,bus=ide.0,unit=0,drive=drive-ide0,id=ide0,bootindex=200' -netdev 'type=tap,id=net0,ifname=tap721i0,script=/var/lib/qemu-server/pve-bridge,downscript=/var/lib/qemu-server/pve-bridgedown,vhost=on' -device 'virtio-net-pci,mac=5A:8D:37:32:57:C6,netdev=net0,bus=pci.0,addr=0x12,id=net0,bootindex=300' -rtc 'driftfix=slew,base=localtime' -global 'kvm-pit.lost_tick_policy=discard'' failed: exit code 1<br><br><br><br></div>According the the proxmox status, the node has 32GB of ram  with  19 GB is use.<br><br></div>dmesg shows a stack dump occuring as well (attached).<br><br><br></div>proxmox 3.4<br></div>kernel 3.10-7<br><br></div>I've found that if I change the vm network card from PV to E1000 and back to PV that usually resolves the problem.<br clear="all"><div><div><div><div><div><div><div><br>-- <br><div class="gmail_signature">Lindsay</div>
</div></div></div></div></div></div></div></div>