iscsi multipath failure with libvirtError: Failed to open file '/dev/mapper/Mar': No such file or directory
hello All, I know the issue is related to libvirt,but i dont know where to ask. i have centos 6.6 running KVM as compute node in openstack icehouse when i try to attach volume to instance it shows 2596: error : virStorageFileGetMetadataRecurse:952 : Failed to open file '/dev/mapper/Mar': No such file or directory in libvirt log This does not always happen when it happens no one will be able to attach volume to instance using EMC VNX as storage backend. multipath.conf # Skip the files uner /dev that are definitely not FC/iSCSI devices # Different system may need different customization devnode ^(ram|raw|loop|fd|md|dm-|sr|scd|st)[0-9]* devnode ^hd[a-z][0-9]* devnode ^cciss!c[0-9]d[0-9]*[p[0-9]*] # Skip LUNZ device from VNX device { vendor DGC product LUNZ } } defaults { user_friendly_names no flush_on_last_del yes } devices { # Device attributed for EMC CLARiiON and VNX series ALUA device { vendor DGC product .* product_blacklist LUNZ path_grouping_policy group_by_prio path_selector round-robin 0 path_checker emc_clariion features 1 queue_if_no_path hardware_handler 1 alua prio alua failback immediate } } Can any one help me with this issue -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Number of threads for virtual machine process
Thanks for the response. Can a vm process create any number of extra threads based on i/o requirement ,or is there any relation between number of VCPU allowed and these extra threads. I hope these threads are for disk i/o and for network it uses vhost-'pid' process. Thanks for your help On Mon, Feb 16, 2015 at 9:21 PM, Paolo Bonzini pbonz...@redhat.com wrote: On 16/02/2015 11:54, mad Engineer wrote: Hello all, On a RHEL 6.4 server i created a vm with 2 VCPU and expecting to see single process with 2 threads on host. but top -p pid-of-qemu -H shows many threads randomly being created and destroyed,pid-of-qemu remains the same but other threads' pid keep on changing. The extra threads are doing I/O. Paolo -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Number of threads for virtual machine process
Hello all, On a RHEL 6.4 server i created a vm with 2 VCPU and expecting to see single process with 2 threads on host. but top -p pid-of-qemu -H shows many threads randomly being created and destroyed,pid-of-qemu remains the same but other threads' pid keep on changing. Can some one help me understand this better.I was in a belief that virtual machine with 2 VCPU will have 2 threads,but sometimes total thread count has reached upto 8. top output from host when running multiple dd 6 qemu 20 0 2480m 381m 4664 R 100.2 3.9 1:57.97 /usr/libexec/qemu-kvm -name HAP -S -M rhel6.1.0 -enable-kvm -m 2048 -smp 2,sockets=2,cores=1,threads=1 -uuid 2b8c1cd4-2de7-e59 7 qemu 20 0 2480m 381m 4664 R 99.6 3.9 0:34.97 /usr/libexec/qemu-kvm -name HAP -S -M rhel6.1.0 -enable-kvm -m 2048 -smp 2,sockets=2,cores=1,threads=1 -uuid 2b8c1cd4-2de7-e594 11101 qemu 20 0 2480m 381m 4664 S 0.0 3.9 0:13.33 /usr/libexec/qemu-kvm -name HAP -S -M rhel6.1.0 -enable-kvm -m 2048 -smp 2,sockets=2,cores=1,threads=1 -uuid 2b8c1cd4-2de7-e594 31278 qemu 20 0 2480m 381m 4664 S 0.0 3.9 0:00.00 /usr/libexec/qemu-kvm -name HAP -S -M rhel6.1.0 -enable-kvm -m 2048 -smp 2,sockets=2,cores=1,threads=1 -uuid 2b8c1cd4-2de7-e594 HT is ON on host. Even Though total CPU consumed by first two threads are 200%,i don't understand why there are these many extra threads Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Centos 6.6 on Centos 6.6,internal error process exited
I have a server with centos 6.5 installed and later upgraded to 6.6 after restart, installed kvm and created centos 6.5 as guest and later upgraded guest to 6.6. Now after host reboot i am not able to start guest as it shows Internal error process exited while reading console log output Supported machines are: pc RHEL 6.5 pc RHEL 6.5.0 RHEL 6.5 pc RHEL 6.4.0 RHEL 6.4 pc Even though by changing Machine in xml config from RHEL 6.6.0 to RHEL 6.5.0 vm has started and its working. But why is 6.6 guest not supported in 6.6 host,i suspect its because of some upgradation issue from 6.5 to 6.6.Can some one please help me understand this. Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Network performance drop when compared to other hypervisor with vhost_net on for UDP
I am running RHEL6.5 as Host and Guest on HP server. Server has 128G and 48 Core[with HT enabled.] 3 VMs are running 2 pinned to first 24 PCPU with proper NUMA pinning, Guests: VM1: 6 VCPU pinned to 6 PCPU NUMA node 1,with 16G RAM VM2: 6 VCPU pinned to 6 PCPU on NUMA node 0,with 16G RAM VM3: 2 VCPU ,no pinning,4G RAM HOST host has 10 free CPU+24 HT threads which is not allocated and is available. Host also runs a small application that is single threaded,that uses ~4G RAM. Total resource to host is 10 CPU+24 HT=34 and 92G unallocated RAM[VMS dont even use 70% of allocated RAM] also ksm is not running. Networking: Uses linux bridge connected to 1Gbps eth0,with ip assigned on eth0 [This IP is called for accessing application running on host] All vms use virtio and VHOST is on . Traffic on virtual machines are ~3MBps and combined traffic on host is ~14MBps VHOST-pid-of-qemu-process sometimes uses ~35% CPU. There is no packet loss,drop or latency,but the issue is with the same setup on Vmware with same sizing of virtual machines,with the only difference as application running on host has moved to fourth VM.So in Vmware there are 4 VMs. Application gives better number ie on KVM that number is 310 and on vmware it is 570.Application uses UDP to communicate. I tried removing VHOST,still value is same.(I hope VHOST-NET UDP issue is solved) Thanks for any help -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ovirt-users] Network performance drop when compared to other hypervisor with vhost_net on for UDP
Thanks Martin, How can we see the changes made by tuned? for virtual guest i see it changes scheduler to deadline.Is there any way to see what parameters each profile is going to change Thanks On Wed, Jan 14, 2015 at 6:14 PM, Martin PavlĂk mpav...@redhat.com wrote: Hi, from the top of my head you could try to play with tuned both with guest and host ###Install### yum install tuned /etc/init.d/tuned start chkconfig tuned on ###usage### list the profile: tuned-adm list change your profile: tuned-adm profile throughput-performance maybe try to experiment with other profiles. HTH Martin Pavlik RHEV QE On 14 Jan 2015, at 12:06, mad Engineer themadengin...@gmail.com wrote: I am running RHEL6.5 as Host and Guest on HP server. Server has 128G and 48 Core[with HT enabled.] 3 VMs are running 2 pinned to first 24 PCPU with proper NUMA pinning, Guests: VM1: 6 VCPU pinned to 6 PCPU NUMA node 1,with 16G RAM VM2: 6 VCPU pinned to 6 PCPU on NUMA node 0,with 16G RAM VM3: 2 VCPU ,no pinning,4G RAM HOST host has 10 free CPU+24 HT threads which is not allocated and is available. Host also runs a small application that is single threaded,that uses ~4G RAM. Total resource to host is 10 CPU+24 HT=34 and 92G unallocated RAM[VMS dont even use 70% of allocated RAM] also ksm is not running. Networking: Uses linux bridge connected to 1Gbps eth0,with ip assigned on eth0 [This IP is called for accessing application running on host] All vms use virtio and VHOST is on . Traffic on virtual machines are ~3MBps and combined traffic on host is ~14MBps VHOST-pid-of-qemu-process sometimes uses ~35% CPU. There is no packet loss,drop or latency,but the issue is with the same setup on Vmware with same sizing of virtual machines,with the only difference as application running on host has moved to fourth VM.So in Vmware there are 4 VMs. Application gives better number ie on KVM that number is 310 and on vmware it is 570.Application uses UDP to communicate. I tried removing VHOST,still value is same.(I hope VHOST-NET UDP issue is solved) Thanks for any help ___ Users mailing list us...@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Again with same Cgroup issue :)
Hello All, From the last couple of days i have been spamming this mailing list with request for configuring Cgroup with libvirtd on Centos systems. I still can not find a permanent solution to limit host RAM to particular value,tried creating a separate hierarchy mykvm and changed in sysconfig/libvirtd after that vm's memory cgroup reflects this.But it is not obeying memory.limit_in_bytes set in mykvm group,i als specified it in cgrules.conf and restarted it.If i change that in /cgconfig/memory/mykvm/libvirt/qemu/memory.limit_in_bytes its working.But that is dynamic as i am not able to find a way to mention that in cgconfig.conf. How can i make sub hierarchies follow what is set in parent cgroup? eg: if change variables in /cgconfig/memory/mykvm then all instances coming under /cgconfig/memory/mykvm/libvirt/qemu/ should follow that value..How is it possible.Can you please help -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
C group hierarchy and libvirtd
On Centos 6.4 x64,with libvirt-0.10.2-18.el6.x86_64 i am trying to set memory.limit_in_bytes for all qemu process. changed cgconfig.conf group mygroup{ perm { admin { uid = root; gid = root; } task { uid = qemu; gid = kvm; } } memory { memory.limit_in_bytes = 61G; } } and also added CGROUP_DAEMON=memory:/mygroup in sysconfig/libvirtd and in cgrules.conf and then restarted services. Now i can see created virtual machines use cgroup hierarchy /cgroup/memory/mygroup/libvirt/qemu/virtualmachine1/ instead of /cgroup/memory/libvirt/qemu/virtualmachine1/. The issue is memory.limit_in_bytes set to mygroup is getting applied only to libvirtd process.VM are not following memory.limit_in_bytes set to mygroup. 1. how can i set this globally so that all virtual machines follow that, i dont want to create a new group for that if libvirt supports it. 2.Is there any way i can avoid extra hierarchy and create virtual machines memory cgroup under /cgroup/memory/kvm/ instead of /cgroup/memory/mygroup/libvirt/qemu/ ? please help me fix this issue.I have this working on ubuntu servers where i specified libvirt-qemu user and VMs follow that cgroup for memory. Regards, -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
using cgroups with KVM
hello all, i am trying to limit RAM usage of guests using memory cgroup. testing on a virtual machine with Nested virtualization enabled. What happens when memory.usage_in_bytes = memory.limit_in_bytes ? Is it going to swap extra memory? Thanks for your help -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
odd number of VCPU and performance impact?
Hi all, Is it a good practice to always create vm with even number of VCPU? What could be the impact if we create vms with odd number of CPU on NUMA or SMP systems.Is there any recommendation Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
ksmd high cpu usage from almost a week with just one vm running
Hello All, I am using centos6.5 x64 on a server with 48 G RAM and 8 Cores.Managed by Ovirt There is only one running VM with RAM 34 G and with 6 VCPU (pinned to proper numa nodes) from top top - 06:42:48 up 67 days, 20:05, 1 user, load average: 0.26, 0.20, 0.17 Tasks: 285 total, 2 running, 282 sleeping, 0 stopped, 1 zombie Cpu(s): 1.0%us, 1.4%sy, 0.0%ni, 97.5%id, 0.1%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 49356468k total, 33977684k used, 15378784k free, 142812k buffers Swap: 12337144k total,0k used, 12337144k free, 343052k cached PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND 101 root 25 5 000 R 27.4 0.0 5650:04 [ksmd] 26004 vdsm 0 -20 3371m 64m 9400 S 9.8 0.1 1653:27 /usr/bin/python /usr/share/vdsm/vdsm --pidfile /var/run/vdsm/vdsmd.pid 20963 qemu 20 0 38.5g 33g 6792 S 3.9 71.6 5225:43 /usr/libexec/qemu-kvm -name Cinder -S -M rhel6.5.0 -cpu Nehalem -enable-kvm -m 34096 -realtime mlock=off -smp 6,maxcpus=160,sockets=80,c from /sys/kernel/mm/ksm pages_unshared 7602322 pages_shared 207023 pages_to_scan 64 pages_volatile31678 Any idea why ksmd is not coming to normal CPU usage ,on a different server ksmd was disabled and for testing when i enabled it initially CPU usage was high but later settled down to 3% ,in that host i have 4 VMs running. Before turning off ksmd can any one help me find out why ksmd is behaving like this.Initially it had 2 virtual machines,because of high CPU utilization of this guest other is migrated to another host. Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Allocating dedicated RAM to host that guest can not use
Hi, Is there any way to set some RAM dedicated to host that guest can not access? Similar to setting RAM to Dom0 in Xen. I am over committing RAM for the instances but don't want host to swap. i understand that virtual machines are process,but can we achieve this Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Allocating dedicated RAM to host that guest can not use
never tried that. can we do that transparently ie with out setting cgroups for each virtul machines?. A global group such that all combined virtual machines RAM utilization to be with in a specific value ? On Thu, Nov 27, 2014 at 4:59 PM, Wanpeng Li wanpeng...@linux.intel.com wrote: On Thu, Nov 27, 2014 at 05:12:52PM +0530, mad Engineer wrote: Hi, Is there any way to set some RAM dedicated to host that guest can not access? Similar to setting RAM to Dom0 in Xen. I am over committing RAM for the instances but don't want host to swap. i understand that virtual machines are process,but can we achieve this How about limit the memory of which guest can access through memory cgroup? Regards, Wanpeng Li Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Allocating dedicated RAM to host that guest can not use
a random thought can we set qemu user/group rss to a particular hard limit in limits.conf Can this work? On Thu, Nov 27, 2014 at 5:39 PM, mad Engineer themadengin...@gmail.com wrote: never tried that. can we do that transparently ie with out setting cgroups for each virtul machines?. A global group such that all combined virtual machines RAM utilization to be with in a specific value ? On Thu, Nov 27, 2014 at 4:59 PM, Wanpeng Li wanpeng...@linux.intel.com wrote: On Thu, Nov 27, 2014 at 05:12:52PM +0530, mad Engineer wrote: Hi, Is there any way to set some RAM dedicated to host that guest can not access? Similar to setting RAM to Dom0 in Xen. I am over committing RAM for the instances but don't want host to swap. i understand that virtual machines are process,but can we achieve this How about limit the memory of which guest can access through memory cgroup? Regards, Wanpeng Li Thanks -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line unsubscribe kvm in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html