Bug 2056725
| Summary: | qemu-kvm leaks on qemu_rbd_snap_list | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 7 | Reporter: | Germano Veit Michel <gveitmic> |
| Component: | qemu-kvm | Assignee: | Stefano Garzarella <sgarzare> |
| Status: | CLOSED ERRATA | QA Contact: | Tingting Mao <timao> |
| Severity: | urgent | Docs Contact: | |
| Priority: | urgent | ||
| Version: | 7.9 | CC: | coli, jferlan, jinzhao, jmaloy, juzhang, kkiwi, kpfleming, lob+redhat, sgarzare, system-engineering, virt-maint, ymankad |
| Target Milestone: | rc | Keywords: | Triaged, ZStream |
| Target Release: | --- | Flags: | pm-rhel:
mirror+
|
| Hardware: | x86_64 | ||
| OS: | All | ||
| Whiteboard: | |||
| Fixed In Version: | qemu-kvm-1.5.3-175.el7_9.6 | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2022-05-18 16:15:18 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: | |||
Stefano and Bermano: Thanks for your info, when tried with 10 RBD disks and removed all the prints/sleeps from the stress test, I reproduced the issue. And also tried with the latest qemu, there is no the issue anymore.
So I will set this bug as verified accordingly. Thanks again.
Reproduced with:
qemu-kvm-1.5.3-175.el7_9.5.x86_64
kernel-3.10.0-1160.el7.x86_64
Results:
The used memory of qemu-kvm pid is increased continuely, and after about 1 hour 10 minutes, the used memory increased to ~20G as below. And the size is still increased continuely.
# cat /proc/3274843/status | grep VmRSS
VmRSS: 21124024 kB
Verified with:
qemu-kvm-1.5.3-175.el7_9.6.x86_64
kernel-3.10.0-1160.el7.x86_64
Results:
The used memory of qemu-kvm pid is not increased, and after about 1 hour 20 minutes, the memory size is still ~1G.
# cat /proc/3412360/status | grep VmRSS
VmRSS: 970676 kB
Steps:
1. Define the VM
# virsh define rhel7.9.xml
Note:
# cat rhel7.9.xml
<domain type='kvm' id='2'>
<name>rhel7.9</name>
<uuid>33e879e7-6591-4f07-8ac0-2ddc33807f1b</uuid>
<memory unit='KiB'>8388608</memory>
<currentMemory unit='KiB'>8388608</currentMemory>
<vcpu placement='static'>2</vcpu>
<resource>
<partition>/machine</partition>
</resource>
<os>
<type arch='x86_64' machine='pc-i440fx-rhel7.0.0'>hvm</type>
<boot dev='hd'/>
</os>
<features>
<acpi/>
<apic/>
</features>
<cpu mode='custom' match='exact' check='full'>
<model fallback='forbid'>Skylake-Server-IBRS</model>
<vendor>Intel</vendor>
<topology sockets='1' cores='1' threads='2'/>
<feature policy='disable' name='ds'/>
<feature policy='disable' name='acpi'/>
<feature policy='require' name='ss'/>
<feature policy='disable' name='ht'/>
<feature policy='disable' name='tm'/>
<feature policy='disable' name='pbe'/>
<feature policy='disable' name='dtes64'/>
<feature policy='disable' name='monitor'/>
<feature policy='disable' name='ds_cpl'/>
<feature policy='disable' name='vmx'/>
<feature policy='disable' name='smx'/>
<feature policy='disable' name='est'/>
<feature policy='disable' name='tm2'/>
<feature policy='disable' name='xtpr'/>
<feature policy='disable' name='pdcm'/>
<feature policy='disable' name='dca'/>
<feature policy='disable' name='osxsave'/>
<feature policy='disable' name='tsc_adjust'/>
<feature policy='require' name='clflushopt'/>
<feature policy='disable' name='intel-pt'/>
<feature policy='require' name='pku'/>
<feature policy='disable' name='ospke'/>
<feature policy='require' name='md-clear'/>
<feature policy='require' name='stibp'/>
<feature policy='require' name='ssbd'/>
<feature policy='require' name='hypervisor'/>
<feature policy='disable' name='arat'/>
</cpu>
<clock offset='utc'>
<timer name='rtc' tickpolicy='catchup'/>
<timer name='pit' tickpolicy='delay'/>
<timer name='hpet' present='no'/>
</clock>
<on_poweroff>destroy</on_poweroff>
<on_reboot>restart</on_reboot>
<on_crash>destroy</on_crash>
<pm>
<suspend-to-mem enabled='no'/>
<suspend-to-disk enabled='no'/>
</pm>
<devices>
<emulator>/usr/libexec/qemu-kvm</emulator>
<disk type='file' device='disk'>
<driver name='qemu' type='qcow2' cache='none'/>
<source file='/home/timao/RHEL-7.9-x86_64-latest.qcow2'/>
<backingStore/>
<target dev='sda' bus='scsi'/>
<alias name='scsi0-0-0-0'/>
<address type='drive' controller='0' bus='0' target='0' unit='0'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdb' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk1'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdc' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk2'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdd' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk3'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sde' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk4'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdf' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk5'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdj' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk6'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdh' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk7'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdi' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk8'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdg' bus='scsi'/>
</disk>
<disk type='network' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<auth username='admin'>
<secret type='ceph' uuid='7c1912af-3a87-40c1-a1e4-9924ec83cb14'/>
</auth>
<source protocol='rbd' name='rbd/disk9'>
<host name='10.73.114.12' port='6789'/>
</source>
<target dev='sdk' bus='scsi'/>
</disk>
<controller type='usb' index='0' model='ich9-ehci1'>
<alias name='usb'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x7'/>
</controller>
<controller type='usb' index='0' model='ich9-uhci1'>
<alias name='usb'/>
<master startport='0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0' multifunction='on'/>
</controller>
<controller type='usb' index='0' model='ich9-uhci2'>
<alias name='usb'/>
<master startport='2'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x1'/>
</controller>
<controller type='usb' index='0' model='ich9-uhci3'>
<alias name='usb'/>
<master startport='4'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x2'/>
</controller>
<controller type='pci' index='0' model='pci-root'>
<alias name='pci.0'/>
</controller>
<controller type='scsi' index='0' model='virtio-scsi'>
<alias name='scsi0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/>
</controller>
<interface type='bridge'>
<mac address='52:54:00:78:3c:d9'/>
<source bridge='switch'/>
<target dev='vnet0'/>
<model type='rtl8139'/>
<alias name='net0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
</interface>
<serial type='unix'>
<source mode='bind' path='/tmp/rhel8.6'/>
<target type='isa-serial' port='0'>
<model name='isa-serial'/>
</target>
<alias name='serial0'/>
</serial>
<console type='unix'>
<source mode='bind' path='/tmp/rhel8.6'/>
<target type='serial' port='0'/>
<alias name='serial0'/>
</console>
<input type='mouse' bus='ps2'>
<alias name='input0'/>
</input>
<input type='keyboard' bus='ps2'>
<alias name='input1'/>
</input>
<graphics type='vnc' port='5900' autoport='yes' listen='0.0.0.0'>
<listen type='address' address='0.0.0.0'/>
</graphics>
<video>
<model type='cirrus' vram='16384' heads='1' primary='yes'/>
<alias name='video0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
</video>
<memballoon model='virtio'>
<alias name='balloon0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/>
</memballoon>
</devices>
<seclabel type='dynamic' model='selinux' relabel='yes'>
<label>system_u:system_r:svirt_t:s0:c296,c922</label>
<imagelabel>system_u:object_r:svirt_image_t:s0:c296,c922</imagelabel>
</seclabel>
<seclabel type='dynamic' model='dac' relabel='yes'>
<label>+107:+107</label>
<imagelabel>+107:+107</imagelabel>
</seclabel>
</domain>
2. Start the VM
# virsh start rhel7.9
3. Stress to the VM
# while true; do virsh domblkinfo rhel7.9 --all;done
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 (qemu-kvm 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-2022:4639 |
Description of problem: Customer qemu-kvm processes memory usage over time far exceeds the memory allocated for the Guest. For example, a Guest configured with 8G of RAM is seen using 32G and so on. OOM killer is frequently seen unless the customer keeps migrating VMs around. Using valgrind massif, we found the root cause: n5: 1623668055 (heap allocation functions) malloc/new/new[], --alloc-fns, etc. n3: 1358331883 0x67D775D: g_malloc (gmem.c:99) n1: 1290516000 0x1B7BB4: qemu_rbd_snap_list (rbd.c:953) n1: 1290516000 0x1A3D1B: bdrv_query_snapshot_info_list (qapi.c:46) n1: 1290516000 0x1A3FD4: bdrv_query_image_info (qapi.c:162) n1: 1290516000 0x1A42B8: bdrv_query_info (qapi.c:249) n1: 1290516000 0x1A451A: qmp_query_block (qapi.c:314) n1: 1290516000 0x299B17: qmp_marshal_input_query_block (qmp-marshal.c:787) n1: 1290516000 0x3298A6: qmp_call_cmd (monitor.c:4511) n1: 1290516000 0x3298A6: handle_qmp_command (monitor.c:4577) n1: 1290516000 0x3D7DA7: json_message_process_token (json-streamer.c:104) n1: 1290516000 0x3E708A: json_lexer_feed_char (json-lexer.c:310) n1: 1290516000 0x3E714D: json_lexer_feed (json-lexer.c:360) n1: 1290516000 0x3285F2: monitor_control_read (monitor.c:4598) n1: 1290516000 0x291030: qemu_chr_be_write (qemu-char.c:167) n1: 1290516000 0x291030: tcp_chr_read (qemu-char.c:2496) n1: 1290516000 0x67D2118: g_main_dispatch (gmain.c:3175) n1: 1290516000 0x67D2118: g_main_context_dispatch (gmain.c:3828) n1: 1290516000 0x263269: UnknownInlinedFun (main-loop.c:187) n1: 1290516000 0x263269: UnknownInlinedFun (main-loop.c:235) n1: 1290516000 0x263269: main_loop_wait (main-loop.c:475) n1: 1290516000 0x1818BF: main_loop (vl.c:1997) n0: 1290516000 0x1818BF: main (vl.c:4367) Tracked this down to this change which was not backported to RHEL 7.9.z. commit 9e6337d0818650362149b734d53edf9489f3acaa Author: Stefan Hajnoczi <stefanha> Date: Wed Sep 25 16:00:48 2013 +0200 rbd: avoid qemu_rbd_snap_list() memory leaks When there are no snapshots qemu_rbd_snap_list() returns 0 and the snapshot table pointer is NULL. Don't forget to free the snaps buffer we allocated for librbd rbd_snap_list(). When the function succeeds don't forget to free the snaps buffer after calling rbd_snap_list_end(). Cc: qemu-stable Signed-off-by: Stefan Hajnoczi <stefanha> Signed-off-by: Kevin Wolf <kwolf> diff --git a/block/rbd.c b/block/rbd.c index f6d3237179..4a1ea5b5ce 100644 --- a/block/rbd.c +++ b/block/rbd.c @@ -943,7 +943,7 @@ static int qemu_rbd_snap_list(BlockDriverState *bs, do { snaps = g_malloc(sizeof(*snaps) * max_snaps); snap_count = rbd_snap_list(s->image, snaps, &max_snaps); - if (snap_count < 0) { + if (snap_count <= 0) { g_free(snaps); } } while (snap_count == -ERANGE); @@ -967,6 +967,7 @@ static int qemu_rbd_snap_list(BlockDriverState *bs, sn_info->vm_clock_nsec = 0; } rbd_snap_list_end(snaps); + g_free(snaps); done: *psn_tab = sn_tab; Then built a test package for the customer based on latest 7.9.z with the patch above. http://brew-task-repos.usersys.redhat.com/repos/scratch/gveitmic/qemu-kvm/1.5.3/175.el7.5.sfdc02961627/x86_64/ After this the problem was gone. Version-Release number of selected component (if applicable): qemu-kvm-1.5.3-175.el7_9.5.x86_64 How reproducible: Always Steps to Reproduce: 1. Configure qemu-kvm guest with RBD disk, and no snapshots 2. Run this in a loop while true ; do date ; virsh domblkinfo <vm name> --all ; sleep 0.1 ; done Actual results: * qemu-kvm memory usage goes through the roof Expected results: * no leaks