http://bugzilla.opensuse.org/show_bug.cgi?id=923150 Bernhard Wiedemann changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |lwang@suse.com, | |sfalken@opensuse.org, | |trenn@suse.com Assignee|bnc-team-screening@forge.pr |nfbrown@suse.com |ovo.novell.com | -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=923150 Diego Ercolani changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(diego.ercolani@gm | |ail.com) | --- Comment #2 from Diego Ercolani --- If I don't pass the "shell" option, after pivoting of root filesystem, systemd complaint a while of the lack of /dev/raid/mysql and /dev/raid/ROOTHOME, then after a while boot procedure drop to rescue shell where: - /dev/md/lvm is started lvscan -v say: inactive /dev/raid/mysql inactive /dev/raid/ROOTHOME -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=923150 --- Comment #4 from Diego Ercolani --- When the problem have began, I noticed that the boot process dropped to rescue shell, after the generation of a new initrd, now the boot process simply freezes (almost for 2 hours) trying to launch processes that depend on the disk it don't find. So I did this workaround: - disable mysql & disable bacula-dir service - added "nofail" attribute to fstab to - /dev/md/mysql - /dev/md/ROOTHOME - /dev/nonraid/tmp I rebooted the machine and, as you can see in the movie system boot up and: - /dev/md/lvm is correctly started - /dev/raid/mysql, /dev/raid/ROOTHOME, /dev/nonraid/tmp are inactive So to address: lvchange -ay /dev/nonraid/tmp lvchange -ay /dev/raid/mysql lvchange -ay /dev/raid/ROOTHOME (systemd mount "new" filesystems authomatically) rcmysql start rcbacula-dir start This should provide another bit of information: http://youtu.be/FzR_V5ULSK0 -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=923150 Diego Ercolani changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(diego.ercolani@gm | |ail.com) | --- Comment #6 from Diego Ercolani --- use_lvmetad was already "1" but it isn't enabled bacula:~ # systemctl status lvm2-lvmetad.service lvm2-lvmetad.service - LVM2 metadata daemon Loaded: loaded (/usr/lib/systemd/system/lvm2-lvmetad.service; disabled) Active: inactive (dead) Docs: man:lvmetad(8) systemctl status lvm2-lvmetad.socket systemctl status lvm2-monitor.service bacula:~ # systemctl status lvm2-lvmetad.socket lvm2-lvmetad.socket - LVM2 metadata daemon socket Loaded: loaded (/usr/lib/systemd/system/lvm2-lvmetad.socket; disabled) Active: inactive (dead) Docs: man:lvmetad(8) Listen: /run/lvm/lvmetad.socket (Stream) bacula:~ # systemctl status lvm2-monitor.service lvm2-monitor.service - Monitoring of LVM2 mirrors, snapshots etc. using dmeventd or progress polling Loaded: loaded (/usr/lib/systemd/system/lvm2-monitor.service; disabled) Active: inactive (dead) Docs: man:dmeventd(8) man:lvcreate(8) man:lvchange(8) man:vgchange(8) I enabled lvm2-lvmetad.socket as stated in /etc/lvm.conf [snip /etc/lvm/lvm.conf] # If lvmetad has been running while use_lvmetad was 0, it MUST be stopped # before changing use_lvmetad to 1 and started again afterwards. # If use_lvmetad set to 1, please make sure lvm2-lvmetad.socket is started use_lvmetad = 1 [/snip] and remade the initrd and rebooted.... Yes, I confirm that all the logical volumes are enabled after the boot I include also the bootlog (journalctl -b) -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=923150 Neil Brown changed: What |Removed |Added ---------------------------------------------------------------------------- Flags| |needinfo?(diego.ercolani@gm | |ail.com) --- Comment #8 from Neil Brown --- Excellent. So if you enable lvm2-lvmetad.socket the problem goes away. So the next question is: did you create the LVM volumes manually or did you use YaST to create them. If the former, then maybe this is a user-error or documentation error and I'd probably reassign this to the maintainer of 'lvm2' to see if they think any documents or defaults could be improved. If the latter, then I would reassign this to the YaST team as yast should probably enable lbm2-lvmetad.socket when any lvm volumes are created. -- You are receiving this mail because: You are on the CC list for the bug.