http://bugzilla.opensuse.org/show_bug.cgi?id=982329
http://bugzilla.opensuse.org/show_bug.cgi?id=982329#c2
--- Comment #2 from Manfred Hollstein ---
The trouble I see is (appears to be) kernel version independent. I have
standard TW kernel 4.5.4 and also the latest from Kernel:stable. The problem
only manifests with TW's lvm2 version, not with an earlier one, such as Leap's
version.
My /boot and /root are a plain (unencrypted) partition on /dev/sda (SSD), while
4 spinning disks are forming an MD-raid10 /dev/md0, which is used as the PV for
one VG providing space for several LVs, such as /var, swap, /home etc.
I'm not sure if both setups share the same problem, but, how long does it take
to run a plain "lvs" to list the logical volumes? With "use_lvmetad = 1" it
responds instantaneously, but with "use_lvmetad = 0" it takes over 4 seconds.
That's why dracut/mkinitrd runs take a minute per kernel in the meantime...
Nevertheless, the logic in lvm2's 69-dm-metad.rules file does not appear to
work, since the VG and LVs don't get activated on boot and therefore cannot be
used - which is a PITA with /var on an LV ;)
--
You are receiving this mail because:
You are on the CC list for the bug.