Red Hat Bugzilla – Bug 1025596
Read/Randread/Randrw performance regression
Last modified: 2013-12-05 05:42:19 EST
This bug has been copied from bug #1022548 and has been proposed to be backported to 6.5 z-stream (EUS).
Test the following scenarios according to Kevin and Fam's suggestions on qemu-kvm-0.12.1.2-2.415.el6_5.1. But there's some problem in scenario (2). (1) Internal snapshot with qcow2 => Pass. a. Boot a guest with qcow2 format b. Create internal snapshot 1 during the guest boot. (qemu) savevm 1 c. Login guest after guest boot up and ping external host. Then create another internal snapshot (qemu) savevm 2 d. (qemu) loadvm 1 c. (qemu) loadvm 2 Result: Internal snapshot could be created and loaded successfully. (2) qemu-img create -f vmdk -o subformat=monolithicFlat /tmp/a.vmdk 1G Formatting '/tmp/a.vmdk', fmt=vmdk size=1073741824 compat6=off subformat='monolithicFlat' zeroed_grain=off [root@localhost home]# [root@localhost home]# [root@localhost home]# qemu-img info /tmp/a.vmdk image: /tmp/a.vmdk file format: raw virtual size: 0 (0 bytes) disk size: 4.0K [root@localhost home]# ==> As a result, the file format is raw instead of "vmdk", and the virtual size is 0 instead of 1G. Downgrade the qemu-kvm to qemu-kvm-0.12.1.2-2.415.el6, no this issue: # rpm -qa | grep qemu gpxe-roms-qemu-0.9.7-6.10.el6.noarch qemu-img-0.12.1.2-2.415.el6.x86_64 qemu-kvm-0.12.1.2-2.415.el6.x86_64 qemu-kvm-debuginfo-0.12.1.2-2.415.el6.x86_64 qemu-kvm-tools-0.12.1.2-2.415.el6.x86_64 [root@localhost qemu-kvm-415]# qemu-img create -f vmdk -o subformat=monolithicFlat /tmp/b.vmdk 1G Formatting '/tmp/b.vmdk', fmt=vmdk size=1073741824 compat6=off subformat='monolithicFlat' zeroed_grain=off [root@localhost qemu-kvm-415]# [root@localhost qemu-kvm-415]# [root@localhost qemu-kvm-415]# qemu-img info /tmp/b.vmdk image: /tmp/b.vmdk file format: vmdk virtual size: 1.0G (1073741824 bytes) disk size: 4.0K Hi, Kevin and Fam Could you help have a look at scenario (2)? Is this a regression? Thanks, Qunfang
Change the status back to ASSIGNED according to comment 9, please help have a check and help fix it asap. Thanks a lot.
Test the following scenarios on qemu-kvm-0.12.1.2-2.415.el6_5.2.x86_64, passed. (1) Internal snapshot with qcow2 => Pass. a. Boot a guest with qcow2 format b. Create internal snapshot 1 during the guest boot. (qemu) savevm 1 c. Login guest after guest boot up and ping external host. Then create another internal snapshot (qemu) savevm 2 d. (qemu) loadvm 1 c. (qemu) loadvm 2 Result: Internal snapshot could be created and loaded successfully. (2) # qemu-img create -f vmdk -o subformat=monolithicFlat /tmp/a.vmdk 1G Formatting '/tmp/a.vmdk', fmt=vmdk size=1073741824 compat6=off subformat='monolithicFlat' zeroed_grain=off [root@localhost home]# [root@localhost home]# qemu-img info /tmp/a.vmdk image: /tmp/a.vmdk file format: vmdk ==> This is correct. virtual size: 1.0G (1073741824 bytes) disk size: 4.0K [root@localhost home]# qemu-img create -f vmdk -o subformat=monolithicFlat /tmp/b.vmdk 100G Formatting '/tmp/b.vmdk', fmt=vmdk size=107374182400 compat6=off subformat='monolithicFlat' zeroed_grain=off [root@localhost home]# [root@localhost home]# [root@localhost home]# qemu-img info /tmp/b.vmdk image: /tmp/b.vmdk file format: vmdk ==> This is correct. virtual size: 100G (107374182400 bytes) disk size: 4.0K Besides, we will run performance test, function test (virtual block, glusterfs, storage vm migration, qcow2, vmdk) to verify this bug.
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, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1750.html