Bug 1436534
| Summary: | It's very slow to boot guest with multiple virtio-serial-pci in specify host. | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Peixiu Hou <phou> |
| Component: | qemu-kvm-rhev | Assignee: | pagupta |
| Status: | CLOSED WORKSFORME | QA Contact: | Qianqian Zhu <qizhu> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 7.4 | CC: | chayang, juzhang, knoel, lijin, michen, phou, virt-maint, wyu, xfu, xiagao |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2018-12-13 07:57:00 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: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 1473046 | ||
|
Description
Peixiu Hou
2017-03-28 06:39:25 UTC
Hi, "qemu-kvm: virtio_bus_set_host_notifier: unable to init event notifier: Too many open files (-24)" qemu-kvm: virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower). This error is coming because enough open file descriptors are not available. Can you please check if increasing the open file descriptors solves the issue for you? Thanks, Pankaj Sorry for late to request, I tried to increase the open file descriptors through add "ulimit -n 16384" to /etc/profile, the error ""qemu-kvm: virtio_bus_set_host_notifier: unable to init event notifier: Too many open files (-24)" qemu-kvm: virtio_bus_start_ioeventfd: failed. Fallback to userspace (slower)"" disappeared. But the guest boot so slowly also can be reproduced on my localhos, Boot guest with 29 virtio-serial-pci, boot command as "-device virtio-serial-pci,id=virtio-serial$i,max_ports=511. Also keep with 29 virtio-serial-pci, change the boot device command to "-device virtio-serial-pci,id=virtio-serial$i" it can be booted quilckly. the host cpu info as follows: 1. [root@localhost qemu-14]# lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 8 On-line CPU(s) list: 0-7 Thread(s) per core: 2 Core(s) per socket: 4 Socket(s): 1 NUMA node(s): 1 Vendor ID: GenuineIntel CPU family: 6 Model: 42 Model name: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz Stepping: 7 CPU MHz: 3700.023 BogoMIPS: 6784.80 Virtualization: VT-x L1d cache: 32K L1i cache: 32K L2 cache: 256K L3 cache: 8192K NUMA node0 CPU(s): 0-7 2. [root@localhost qemu-14]# cat /proc/cpuinfo processor : 7 vendor_id : GenuineIntel cpu family : 6 model : 42 model name : Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz stepping : 7 microcode : 0x29 cpu MHz : 3686.476 cache size : 8192 KB physical id : 0 siblings : 8 core id : 3 cpu cores : 4 apicid : 7 initial apicid : 7 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts bogomips : 6784.80 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: used version: kernel-3.10.0-680.el7.x86_64 qemu-img-rhev-2.9.0-14.el7.x86_64 virtio-win-prewhql-137 (In reply to Peixiu Hou from comment #3) > Sorry for late to request, I tried to increase the open file descriptors > through add "ulimit -n 16384" to /etc/profile, the error ""qemu-kvm: > virtio_bus_set_host_notifier: unable to init event notifier: Too many open > files (-24)" > qemu-kvm: virtio_bus_start_ioeventfd: failed. Fallback to userspace > (slower)"" disappeared. > > But the guest boot so slowly also can be reproduced on my localhos, Boot > guest with 29 virtio-serial-pci, boot command as "-device > virtio-serial-pci,id=virtio-serial$i,max_ports=511. > > Also keep with 29 virtio-serial-pci, change the boot device command to > "-device virtio-serial-pci,id=virtio-serial$i" it can be booted quilckly. Can you please check if 'max_ports' value is <=31 works fine? Thanks, Pankaj > > the host cpu info as follows: > 1. [root@localhost qemu-14]# lscpu > Architecture: x86_64 > CPU op-mode(s): 32-bit, 64-bit > Byte Order: Little Endian > CPU(s): 8 > On-line CPU(s) list: 0-7 > Thread(s) per core: 2 > Core(s) per socket: 4 > Socket(s): 1 > NUMA node(s): 1 > Vendor ID: GenuineIntel > CPU family: 6 > Model: 42 > Model name: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz > Stepping: 7 > CPU MHz: 3700.023 > BogoMIPS: 6784.80 > Virtualization: VT-x > L1d cache: 32K > L1i cache: 32K > L2 cache: 256K > L3 cache: 8192K > NUMA node0 CPU(s): 0-7 > > 2. [root@localhost qemu-14]# cat /proc/cpuinfo > processor : 7 > vendor_id : GenuineIntel > cpu family : 6 > model : 42 > model name : Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz > stepping : 7 > microcode : 0x29 > cpu MHz : 3686.476 > cache size : 8192 KB > physical id : 0 > siblings : 8 > core id : 3 > cpu cores : 4 > apicid : 7 > initial apicid : 7 > fpu : yes > fpu_exception : yes > cpuid level : 13 > wp : yes > flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat > pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm > constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc > aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 > ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes > xsave avx lahf_lm epb tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm > ida arat pln pts > bogomips : 6784.80 > clflush size : 64 > cache_alignment : 64 > address sizes : 36 bits physical, 48 bits virtual > power management: > > used version: > kernel-3.10.0-680.el7.x86_64 > qemu-img-rhev-2.9.0-14.el7.x86_64 > virtio-win-prewhql-137 (In reply to pagupta from comment #4) > (In reply to Peixiu Hou from comment #3) > > Sorry for late to request, I tried to increase the open file descriptors > > through add "ulimit -n 16384" to /etc/profile, the error ""qemu-kvm: > > virtio_bus_set_host_notifier: unable to init event notifier: Too many open > > files (-24)" > > qemu-kvm: virtio_bus_start_ioeventfd: failed. Fallback to userspace > > (slower)"" disappeared. > > > > But the guest boot so slowly also can be reproduced on my localhos, Boot > > guest with 29 virtio-serial-pci, boot command as "-device > > virtio-serial-pci,id=virtio-serial$i,max_ports=511. > > > > Also keep with 29 virtio-serial-pci, change the boot device command to > > "-device virtio-serial-pci,id=virtio-serial$i" it can be booted quilckly. > > Can you please check if 'max_ports' value is <=31 works fine? > > Thanks, > Pankaj > Tried with 'max_ports' value is <=31, booted in guest quickly. Also tried with 'max_ports=511', reproduced this issue, booted in guest so slowly. Best Regards~ Peixiu Hou For virtio-serial we support 'max_ports' <=31, see comment 5. So, we should use use maximum limit for ports. libvirt should take care of this. I am closing this BZ. Thanks, Pankaj (In reply to pagupta from comment #10) > For virtio-serial we support 'max_ports' <=31, see comment 5. So, we should > use use maximum limit for ports. libvirt should take care of this. > I am closing this BZ. > > Thanks, > Pankaj Well noted, then kvm qe will disable max_ports>31 test cases accordingly. Best regards, Junyi |