11

I just updated one of our debian jessie servers and the kernel was updated, nothing special, as we have done this many times. But the first time there were some warnings when the grub configuration file was being generated. I have never seen them before. As far as I can tell the system runs nicely after a reboot.

Setting up linux-image-3.16.0-4-amd64 (3.16.7-ckt25-2+deb8u3) ...
/etc/kernel/postinst.d/initramfs-tools:
update-initramfs: Generating /boot/initrd.img-3.16.0-4-amd64
/etc/kernel/postinst.d/zz-update-grub:
Generating grub configuration file ...
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
Found linux image: /boot/vmlinuz-3.16.0-4-amd64
Found initrd image: /boot/initrd.img-3.16.0-4-amd64
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
  WARNING: lvmetad is running but disabled. Restart lvmetad before enabling it!
done

I searched for the warning online, but I couldn't find a decent explanation that made sense to me (maybe not understood?) and also couldn't understand if this can be ignored. Anyone here has an idea? Thanks

Improve this question
10
  • As entry point: # systemctl list-unit-files | grep -i lvm
    – user178085
    Commented Jul 4, 2016 at 14:22
  • What is locking_type set to in /etc/lvm/lvm.conf?
    – Ferenc Wágner
    Commented Jul 4, 2016 at 16:33
  • locking_type = 1. Seems to be the standard.
    – Preexo
    Commented Jul 5, 2016 at 6:26
  • And use_lvmetad = 1? Is lvmetad really running? What does systemctl status lvm2-lvmetad say? Maybe the zz-update-grub hook temporarily overrides some of these settings? Don't you get similar warnings when you run LVM commands manually?
    – Ferenc Wágner
    Commented Jul 5, 2016 at 7:08
  • 3
    I've seen this too on some Jessie systems recently, but I've ignored it because I think it's just an unfortunately worded error message. In lvm.conf there's this comment: If lvmetad has been running while use_lvmetad was 0, it MUST be stopped before changing use_lvmetad to 1 and started again afterwards. So, I think this warning is just trying to remind you that you're in a weird situation of lvmetad running but disabled via configuration and only if you want to enable it in the config file, you will need to restart it. Oddly, the conf comment and warning disagree about order.
    – Brad Spencer
    Commented Jul 5, 2016 at 14:06

2 Answers 2

Reset to default
8
+50

according to info from Peter Rajnoha about an old 2014 fedora bug 1152185, "The warning is there because if lvmetad is already instantiated and running, then using use_lvmetad=0 will cause LVM commands run under this setting to not notify lvmetad about any changes - therefore lvmetad may miss some information - hence the warning.".

https://bugzilla.redhat.com/show_bug.cgi?id=1152185

However, in our case use_lvmetad = 0, so I tend to believe the warnings appear only during the update and the grub reconfiguration.

According to the explanations in the bug report, this is connected with lvm2-monitor, which is happily running on my system, I believe on yours too. Please check out the Process line:

# systemctl status lvm2-monitor
â lvm2-monitor.service - Monitoring of LVM2 mirrors, snapshots etc. using     dmeventd or progress polling
   Loaded: loaded (/lib/systemd/system/lvm2-monitor.service; enabled)
   Active: active (exited) since Sat 2016-07-09 04:04:49 EEST; 34min ago
     Docs: man:dmeventd(8)
           man:lvcreate(8)
           man:lvchange(8)
           man:vgchange(8)
  Process: 328 ExecStart=/sbin/lvm vgchange --monitor y --ignoreskippedcluster (code=exited, status=0/SUCCESS)
 Main PID: 328 (code=exited, status=0/SUCCESS)
   CGroup: /system.slice/lvm2-monitor.service

I do not see any traces of the warning after reboot and based on the other information I believe the warning is safe to ignore at this stage. If you get any more or other warnings, you should look into it further.

Also, I used to receive LVM warnings on each image update or grub reconfiguration about the names I believe, which turned out to be unimportant and most probably connected to the old hardware. So this is not uncommon.

Preexo, I hope that this has answered your two concerns. Rubo77, I hope I have been helpful for you too.

Kind regards!

Improve this answer
2

It looks like the Debian default is to have use_lvmetad=1 and the warning comes because you have it locally turned off, yet lvmetad is running.

So one solution is to set use_lvmetad=1 in /etc/lvm/lvm.conf. This worked for me.

Improve this answer
1

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .