From mboxrd@z Thu Jan 1 00:00:00 1970 References: <5AC35747020000F9000B344F@prv-mh.provo.novell.com> From: Zdenek Kabelac Message-ID: Date: Tue, 3 Apr 2018 10:10:16 +0200 MIME-Version: 1.0 In-Reply-To: <5AC35747020000F9000B344F@prv-mh.provo.novell.com> Content-Language: en-US Content-Transfer-Encoding: 7bit Subject: Re: [linux-lvm] PVmove can not work on lvm2-2.02.177 Reply-To: LVM general discussion and development List-Id: LVM general discussion and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , List-Id: Content-Type: text/plain; charset="us-ascii"; format="flowed" To: LVM general discussion and development , Gang He Dne 3.4.2018 v 04:28 Gang He napsal(a): > Hello list, > > As you know, pvmove can run on old version (e.g. lvm2-2.02.120 on SLE12SP2), > But with new version lvm2-2.02.177, I can not run pvmove successfully in the cluster. > Here, I paste some information from my test, > if you can know the cause, please help to figure out. > > tb0307-nd1:/ # cat /etc/issue > Welcome to openSUSE Tumbleweed 20180307 - Kernel \r (\l). > tb0307-nd1:/ # uname -r > 4.15.7-1-default > tb0307-nd1:/ # rpm -qa | grep lvm > lvm2-cmirrord-2.02.177-4.1.x86_64 > liblvm2app2_2-2.02.177-4.1.x86_64 > liblvm2cmd2_02-2.02.177-4.1.x86_64 > lvm2-clvm-2.02.177-4.1.x86_64 > lvm2-lockd-2.02.177-4.1.x86_64 > lvm2-2.02.177-4.1.x86_64 > lvm2-testsuite-2.02.177-4.1.x86_64 > > tb0307-nd1:/ # pvs > PV VG Fmt Attr PSize PFree > /dev/vdb cluster-vg2 lvm2 a-- 40.00g 20.00g > /dev/vdc cluster-vg2 lvm2 a-- 40.00g 40.00g > tb0307-nd1:/ # vgs > VG #PV #LV #SN Attr VSize VFree > cluster-vg2 2 1 0 wz--nc 79.99g 59.99g > tb0307-nd1:/ # lvs > LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert > test-lv cluster-vg2 -wi-ao---- 20.00g > tb0307-nd1:/ # pvmove -i 5 /dev/vdb /dev/vdc > Increasing mirror region size from 0 to 16.00 KiB > Error locking on node a4311a8: Device or resource busy > Failed to activate cluster-vg2/test-lv <<== Failed, but in fact, I can mount cluster-vg2/test-lv with ocfs2 file system in the cluster, and read/write files from each node. Hi Yep - it's a work in progress - if you take upstream from git - there are already fixes committed. Hopefully next release (2.02.178) will address most regressions cause by newer (hopefully better) code. Good to see someone is really using it... Regards Zdenek