Hi there,
I'm running a testnode (bare Jessie with a special partition layout) and it was running smooth and shiny with kernel 4.4.35-2, absolutely no issues and all the lvm commands were as quick as usual.
But after upgrading to kernel-4.4.49, suddenly all lvm related commands took way too long (about 20-30 seconds) to complete. After completion - all lvs are recognized, the command is just too slow.
Running lvscan / pvscan with debug parameters (-vvv) is showing my lots of /dev/loopX devices are scanned, but also scanning mdadm devices seems way slower to identify than with the previous kernel.
Booting back to 4.4.35: all issues are gone, lvm commands response as quick as they should.
Are there any changes related to lvm with the most recent kernel ? I couldn't find anything related in the release notes for kernel 4.4.49. Do I have a add/change a filter in lvm.conf to avoid this issue ?
Anybody else with such problems or is it just me ?
Thanks and Regards
Peter
I'm running a testnode (bare Jessie with a special partition layout) and it was running smooth and shiny with kernel 4.4.35-2, absolutely no issues and all the lvm commands were as quick as usual.
But after upgrading to kernel-4.4.49, suddenly all lvm related commands took way too long (about 20-30 seconds) to complete. After completion - all lvs are recognized, the command is just too slow.
Running lvscan / pvscan with debug parameters (-vvv) is showing my lots of /dev/loopX devices are scanned, but also scanning mdadm devices seems way slower to identify than with the previous kernel.
Booting back to 4.4.35: all issues are gone, lvm commands response as quick as they should.
Are there any changes related to lvm with the most recent kernel ? I couldn't find anything related in the release notes for kernel 4.4.49. Do I have a add/change a filter in lvm.conf to avoid this issue ?
Anybody else with such problems or is it just me ?
Thanks and Regards
Peter