Package: lvm2 Version: 2.03.02-1 Severity: important At boottime I get: WARNING: Device /dev/md0 not initialized in udev database even after waiting 10000000 microseconds. two times. System boots to maintenance stage. There I have to login as root and run `vgchange -ay' and <ctrl-d> to get a running system. /dev/md0 was assembled fine in the first boot stage. I am running a custom 4.19.13 which ran fine on stable. Adding a initramfs-tools/script/local-top/lvm which instructs to run `vgchange -ay' let the system boot without intervention but pulls out the 'WARNING' 4 times while booting Thanks in advance
* Elimar Riesebieter <riesebie@lxtec.de> [2019-01-07 18:21 +0100]: The pv is running on top of /dev/md0 (RAID 1).
Hi, Elimar Riesebieter wrote: I get the same warning, too, at boot time and so far also at any time I call a LVM command. In anycase the command (and hence the boot itself, too) takes about 7 minutes! (Which makes every reboot very annoying, and I had plenty of them recently due to #918764.) So far I've seen it in the following circumstances: * At boot, probably due to calls to vgchange or so. * Calling vgs * Calling lvs * Calling lvresize I get this type of warning 44 times per boot or LVM command: # time lvs WARNING: Device /dev/md0 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-0 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/md1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/md2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-3 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-4 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-5 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-6 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-7 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-8 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-9 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-10 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-11 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-12 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-13 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-14 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-15 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-16 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-17 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-18 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-19 not initialized in udev database even after waiting 10000000 microseconds. /dev/sde: open failed: No medium found /dev/sdf: open failed: No medium found /dev/sdg: open failed: No medium found /dev/sdh: open failed: No medium found /dev/sdi: open failed: No medium found WARNING: Device /dev/md0 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-0 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/md1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/md2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-3 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-4 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-5 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-6 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-7 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-8 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-9 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-12 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-13 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-14 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-15 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-16 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-17 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-18 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/dm-19 not initialized in udev database even after waiting 10000000 microseconds. LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert home vgc6 -wi-ao---- <38.61g log vgc6 -wi-ao---- <1.86g root vgc6 -wi-ao---- 53.00g var vgc6 -wi-ao---- 4.25g anx vgwd -wi-ao---- 200.00g apt vgwd -wi-ao---- 1.00g bkup vgwd -wi-ao---- 150.00g c3pio-sd vgwd -wi-a----- 65.00g crsh vgwd -wi-ao---- 20.00g dph vgwd -wi-ao---- 90.00g games vgwd -wi-ao---- 2.00g img vgwd -wi-ao---- 70.00g med vgwd -wi-ao---- 3.00g pbr vgwd -wi-ao---- 8.00g scr vgwd -wi-ao---- 60.00g swap vgwd -wi-ao---- 16.00g tmp vgwd -wi-a----- 30.00g wdm-test vgwd -wi-a----- 10.00g lvs 0.44s user 0.59s system 0% cpu 7:01.58 total # ^^^^^^^^^^^^^ Please also note the 7 minutes this command took to finish. /dev/sde to /dev/sdi are the five different slots of an internal USB 3.0 multi-format card reader. That's a clear difference to my case, probably because of using a stock kernel: System boots more or less fine, but takes about 8.5 minutes minutes to boot of which likely 7 minutes are due to LVM. vgc6 is a VG on two SSDs coupled to a RAID1 with md. vgwd is a VG on two 3TB spinning disks coupled to a RAID1 with md. There's also a third md-based RAID1 without LVM for /boot/. I am running stock Debian unstable kernels. So far it happened with 4.18.20-2 (last 4.18 which was in sid/buster) and 4.19.12-1 (current is 4.19.13-1). Oh, and JFTR: I'm using sysvinit-core as PID1 and it so far happened with at least udev 240-2 and 240-3. Regards, Axel
Hi, about 20 minutes and two reboots later... Axel Beckert wrote: Same with 4.19.13-1 and 4.20-1~exp1. Regards, Axel
Perhaps related to this, I've noticed horrific slowdowns with v2.03.02-1 when running stuff inside chroot unless I bind-mount /run from the host environment. Yesterday I noticed that it can take hours to run update-grub, and I get tons of this same error message for each device/partition WARNING: Device /dev/loop0 not initialized in udev database even after Problem goes away immediately when I mount /run. Something related to missing /run/udev. # apt policy lvm2
* Axel Beckert <abe@debian.org> [2019-01-10 05:57 +0100]: For my system this seems to be fixed. Can't reproduce which update pulled the bug out: ii udev 240-4 ii mdadm 4.1-1 ii lvm2 2.03.02-1 Custom 4.19.15. No custom lvm script. Elimar
* Elimar Riesebieter <riesebie@lxtec.de> [2019-01-16 22:08 +0100]: My /boot resides on a separated lv on a vg on top of a mdadm. I've installed grml-rescueboot. grub can't find vg0 and therefor doesn't boot a grml-image (2018-12). It did boot last in October with the very same layout.... Elimar
Hi, * Ville Korhonen [Thu Jan 10, 2019 at 06:45:40PM +0200]: I can confirm this problem, it seems to have appeared with the upload of lvm 2.03.02-1. STR (with `/dev/mapper/$VG-$LV` corresponding to a LV): mkfs.ext4 /dev/mapper/$VG-$LV mount /dev/mapper/$VG-$LV /mnt debootstrap buster /mnt chroot /mnt apt -y install lvm2 mount --bind /proc /mnt/proc mount --bind /sys /mnt/sys mount --bind /dev /mnt/dev mount --bind /dev/pts /mnt/dev/pts chroot /mnt vgs Then the vgs process is horribly slow and causing many messages like: WARNING: Device /dev/... not initialized in udev database even after waiting 10000000 microseconds As soon as /run/udev is available inside the chroot, then lvm behaves as expected: mount --bind /run/udev /mnt/run/udev regards, -mika-
metoo Sample (run on my laptop): # pvdsplay WARNING: Device /dev/sda not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/root not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sda1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/swap not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/export not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb3 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/root not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sda1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/swap not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/vg00/export not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb1 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb2 not initialized in udev database even after waiting 10000000 microseconds. WARNING: Device /dev/sdb3 not initialized in udev database even after waiting 10000000 microseconds. --- Physical volume --- PV Name /dev/sda1 VG Name vg00 PV Size 107.13 GiB / not usable 3.16 MiB Allocatable yes (but full) PE Size 4.00 MiB Total PE 27425 Free PE 0 Allocated PE 27425 PV UUID GKAL1e-CUkg-y3GD-y1WR-GcKl-xXkI-fI3Gik This breaks "dpkg-reconfigure linux-image-something" in a chroot. It takes hours to complete, while you are eager to get your server up and running again as fast as possible. Obviously pvdisplay worked without udev database entries, so I don't see a reason why it took so long. Moving back to lvm2 2.03.01-2 fixes the delay. Regards Harri
Hi, we’re also hit by this (for a while already, but now on more systems). I discovered the following things: During boot, when the “even after waiting 10000000 microseconds” message comes, hitting ^C allows the boot to continue, although several things (such as the framebugger console) are missing/not initialised. Logging in and doing sudo /etc/init.d/udev stop sudo /etc/init.d/udev start fixes it. Or, letting it boot, and getting the “sudo lvs” to hang. The same commands (udev stop/start) fix it, until the next reboot. All with sysvinit. On one system, I did not get it any more after today’s dist- upgrade (in sid), although that was the one on which I got it only late. On my own X61 work laptop, I don’t get it. Axel, do you still get it on your laptop after dist-upgrading to latest sid? bye, //mirabilos
(answered in chat) the bug is apparently gone in latest sid. @submitter can you confirm? XTaran? bye, //mirabilos
(answered in chat) the bug is apparently gone in latest sid. @submitter can you confirm? XTaran? bye, //mirabilos
Thorsten Glaser wrote: I can't remember having seen it recently, but for checking I need to reboot and I don't do that every few days. Another thing I need to check before being able to answer is if I still have that "sleep 5" workaround in the udev init script which Michael Biebl suggested back then and which at least helped with some recent udev/LVM issues on some, but not all machines. So: Will tell as soon as will do the next reboot with at least that machine where I remember having experienced this heavily. (There were others, too, where it was only about a minute of introduced lag and where I don't remember anymore which machines were affected. So I should probably remove that workaround from any such machine and check them all.) Regards, Axel
Thorsten Glaser wrote: I can't remember having seen it recently, but for checking I need to reboot and I don't do that every few days. Another thing I need to check before being able to answer is if I still have that "sleep 5" workaround in the udev init script which Michael Biebl suggested back then and which at least helped with some recent udev/LVM issues on some, but not all machines. So: Will tell as soon as will do the next reboot with at least that machine where I remember having experienced this heavily. (There were others, too, where it was only about a minute of introduced lag and where I don't remember anymore which machines were affected. So I should probably remove that workaround from any such machine and check them all.) Regards, Axel
Ah. I normally put rootdelay=5 onto the kernel commandline for that. OK. Take care, //mirabilos
Ah. I normally put rootdelay=5 onto the kernel commandline for that. OK. Take care, //mirabilos
As Ville Korhonen already reported, this bug affects all chroot based stuff (debootstick, multistrap, schroot) making them effectively unusable. There's some more info here: https://bbs.archlinux.org/viewtopic.php?id=242594
The Arch post mentioned above helpd me a lot but was not enough for me since all lvm tools where still stuck in the chroot env. My solution was to also bind mount /run/udev from the host to the chrooted system. My complete story and solution below: I faced this bug today while playing with LVM volumes on a new EFI machine. I was using a live Debian Buster and chrooted Debian Buster. In the chroot, I tried =strace /sbin/lvdisplay= and discovered attempts to open files in /run/udev/. I also warn you that for Debian Buster to be chrooted, your live CD must be equipped with same LVM2 version or so. This is the first time I face this issue (I have debug lots of situations in the past using, for example, quite old rescuecd w.r.t. chrooted system). # On the host: apt-get install lvm2 pvscan vgchange -a y mount /dev/vg/debian-amd64 /mnt/debian-amd64 # Prepare chroot: root=/mnt/debian-amd64 mkdir -p /mnt/debian-amd64/run/lvm mkdir -p /mnt/debian-amd64/run/udev mount --bind /dev $root/dev mount --bind /proc $root/proc mount --bind /sys $root/sys mount --bind /dev/pts $root/dev/pts mount --bind /run/lvm $root/run/lvm mount --bind /run/udev $root/run/udev # Do not forget EFI partition: mount /dev/sda1 $root/boot/efi # Run chroot: chroot $root /bin/bash # In the chroot: update-grub grub-install --efi-directory=/boot/efi Hope this helps. Nicolas
Greetings, from The illuminati world elite empire. Bringing the poor, the needy and the talented to limelight of fame, riches, powers and security, get recognized in your business, political race, rise to the top in whatever you do, be protected spiritually and physically! All these you will achieve in a twinkle of an eye when you get initiated to the great Illuminati empire. Once you are initiated to the illuminati empire you will get numerous benefits and reward. Note: that this email message was created solely for the purpose of our recruitment scheme which will end next month and this offer is for unique ones only, if you are not serious on joining the illuminati empire, then you are advise not to contact us at all. This is because disloyalty is highly not tolerated here in our organization. Do you agree to be a member of the illuminati new world order? If YES!. Then kindly reply us back on our direct recruitment email only at: joinilluminatin@hotmail.com Please note, Kindly make sure all your response are send directly to the email stated above only at:> joinilluminatin@hotmail.com For more instructions on our membership process. Note: Some email providers incorrectly place official Illuminati messages in their spam / junk folder or promotion folder. This can divert and exclude our responses to your emails. The Illuminati.
Greetings, from The illuminati world elite empire. Bringing the poor, the needy and the talented to limelight of fame, riches, powers and security, get recognized in your business, political race, rise to the top in whatever you do, be protected spiritually and physically! All these you will achieve in a twinkle of an eye when you get initiated to the great Illuminati empire. Once you are initiated to the illuminati empire you will get numerous benefits and reward. Note: that this email message was created solely for the purpose of our recruitment scheme which will end next month and this offer is for unique ones only, if you are not serious on joining the illuminati empire, then you are advise not to contact us at all. This is because disloyalty is highly not tolerated here in our organization. Do you agree to be a member of the illuminati new world order? If YES!. Then kindly reply us back on our direct recruitment email only at: joinilluminatin@hotmail.com Please note, Kindly make sure all your response are send directly to the email stated above only at:> joinilluminatin@hotmail.com For more instructions on our membership process. Note: Some email providers incorrectly place official Illuminati messages in their spam / junk folder or promotion folder. This can divert and exclude our responses to your emails. The Illuminati.
Greetings, from The illuminati world elite empire. Bringing the poor, the needy and the talented to limelight of fame, riches, powers and security, get recognized in your business, political race, rise to the top in whatever you do, be protected spiritually and physically! All these you will achieve in a twinkle of an eye when you get initiated to the great Illuminati empire. Once you are initiated to the illuminati empire you will get numerous benefits and reward. Note: that this email message was created solely for the purpose of our recruitment scheme which will end next month and this offer is for unique ones only, if you are not serious on joining the illuminati empire, then you are advise not to contact us at all. This is because disloyalty is highly not tolerated here in our organization. Do you agree to be a member of the illuminati new world order? If YES!. Then kindly reply us back on our direct recruitment email only at: joinilluminatin@hotmail.com Please note, Kindly make sure all your response are send directly to the email stated above only at:> joinilluminatin@hotmail.com For more instructions on our membership process. Note: Some email providers incorrectly place official Illuminati messages in their spam / junk folder or promotion folder. This can divert and exclude our responses to your emails. The Illuminati.