Description of problem: The built-in lvm filter-mpath detects and ignores multipath component devices, but sometimes component devs can get through the filter and be seen by lvm commands [1]. When this happens, lvm sees the same PV on different devices and warns about "duplicate PVs". More critically, lvm activation commands will not activate LVs when these duplicate PVs are seen. This often leads to a boot failure when the root LV fails to be activated (or some other LV required for boot.) LVM now has access to device wwids, and this can be used to detect and eliminate multipath components if they happen to get through the filter. This means that cases of failed multipath component detection should not longer prevent lvm from operating normally. upstream commit: https://sourceware.org/git/?p=lvm2.git;a=commit;h=212b1fc529a7d550c49f32cb7cbecb017ceb35b0 [1] filter-mpath has traditionally relied on the kernel sysfs info from dm multipath devs to detect components, but sometimes the multipath device is not set up when needed, causing the components to get through the filter. More recently the filter began using /etc/multipath/wwids as a secondary method of detecting a component device. If both of these methods fail, then the new wwid comparison of duplicate PVs described above will resolve the problem (although it occurs later in the command than the filter, so is somewhat less efficient because lvm will read the component devs prior to eliminating them.) Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
This bug was created as a reference for the many cases that include problems with duplicate PVs related to multipath. bug 1971263 bug 2012751 bug 1933038
Marking Verified:Tested (SanityOnly) with the latest rpms. All devicesfile regression scenarios passed. kernel-4.18.0-398.g366e.el8.kpq1 BUILT: Tue Jun 7 04:56:38 CDT 2022 lvm2-2.03.14-4.el8 BUILT: Wed Jun 15 17:14:34 CDT 2022 lvm2-libs-2.03.14-4.el8 BUILT: Wed Jun 15 17:14:34 CDT 2022
All devicesfile regression scenarios passed on latest kernel/lvm2 as well. kernel-4.18.0-411.el8 BUILT: Wed Jul 20 18:42:42 CDT 2022 lvm2-2.03.14-5.el8 BUILT: Thu Jul 14 09:23:13 CDT 2022 lvm2-libs-2.03.14-5.el8 BUILT: Thu Jul 14 09:23:13 CDT 2022
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 (lvm2 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-2022:7792