Too long boot and shut down after kernel update

English Install/Boot/Login
1

Hello I am using OpenSUSE Tumbleweed. After upgrading to kernel 4.4.0.1 from 4.3.3.6. boot and shut down are taking a lot of time. Seems it is because of a cron job about LVM2, but I do not use it. It is a similar problem to this: https://forums.opensuse.org/showthread.php/494300-lvm2-is-stopping-the-boot-process

Doing:

systemctl mask lvm2-activation-early.service
systemctl mask lvm2-activation.service

(According to this recomendation: https://forums.opensuse.org/showthread.php/503293-Boot-takes-3-minutes-after-installing-Sundtek-USB-DVB-S2-driver )

Not solves the problem. The boot log shows:

http://pastebin.com/4TSi5Bkr

In the power off it says that could not unmount the linux partition because /var/run is busy.

This coomand

systemctl status lvm2-activation-net.service

shows:

lvm2-activation-net.service - Activation of LVM2 logical volumes
   Loaded: loaded (/etc/lvm/lvm.conf)
   Active: failed (Result: exit-code) since mar 2016-01-19 16:57:06 CET; 9min ago
     Docs: man:lvm2-activation-generator(8)
  Process: 1127 ExecStartPre=/usr/bin/udevadm settle (code=exited, status=1/FAILURE)

But if instead of starting with new kernel (4.4.0.1) I select the previous (4.3.3.6) then it is working:

http://pastebin.com/TSagp0GL

And

systemctl status lvm2-activation-net.service

output is:

systemd-udev-settle.service
   Loaded: masked (/dev/null)
   Active: inactive (dead)

Boot and shutdown are fast again. And returning to new installed kernel (4.4.0.1) drives to previous problem. (And every time I change kernel I reinstall nvidia driver).

My fdisk -l its:


Disk /dev/sda: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: dos
Disk identifier: 0x76692ca8

Device     Boot      Start        End    Sectors   Size Id Type
/dev/sda1             2048   30716279   30714232  14.7G 1c Hidden W95 FAT32 (L
/dev/sda2         30716280  239400629  208684350  99.5G  7 HPFS/NTFS/exFAT
/dev/sda3  *     239416688 1953520064 1714103377 817.4G  f W95 Ext'd (LBA)
/dev/sda5        239416751 1634742324 1395325574 665.4G  7 HPFS/NTFS/exFAT
/dev/sda6       1634744320 1638934527    4190208     2G 82 Linux swap / Solari
/dev/sda7       1638935298 1848647744  209712447   100G 83 Linux
/dev/sda8       1932539238 1953520064   20980827    10G 83 Linux

And blkid:

/dev/sda1: LABEL="RECOVERY" UUID="3C98-AC5D" TYPE="vfat" PARTUUID="76692ca8-01"
/dev/sda2: LABEL="OS" UUID="6E40872F4086FCD9" TYPE="ntfs" PARTUUID="76692ca8-02"
/dev/sda5: LABEL="DATA" UUID="B26C01CAB53315F1" TYPE="ntfs" PTTYPE="dos" PARTUUID="76692ca8-05"
/dev/sda6: UUID="de3f60c8-44dc-4aff-aadb-0cac1303db71" TYPE="swap" PARTUUID="76692ca8-06"
/dev/sda7: UUID="509f86da-eb1b-464a-840f-9d9c17575846" TYPE="ext4" PARTUUID="76692ca8-07"
/dev/sda8: LABEL="Android-x86" UUID="d37f905a-40e3-4451-8b2e-32f52ef08726" TYPE="ext3" PARTUUID="76692ca8-08"

(sda1 is recovery, sda2 is windows, sda3 is extended partition, sda5 is data partition, sda6 swap, sda7 opensuse and sda8 android-x86)

2

Hello I can not find problem. In YAST->services: *lvm2-activation-net , lvm2-lvmetad and lvm2-monitor *were disabled and inactives. Commands:

systemctl disable lvm2-lvmetad.service lvm2-lvmetad.socket
systemctl stop lvm2-lvmetad.service lvm2-lvmetad.socket

Not change anything. Editing parameters from lvm file in */etc/lvm/lvm.conf *also no changes. Now I just reinstalled new Tumbleweed ISO and problem appears from first boot.

systemd-analyze blame

shows:

2min 480ms systemd-udev-settle.service
         54.789s lvm2-activation-net.service
         13.897s purge-kernels.service
         11.115s apparmor.service
          5.890s dev-sda8.device
          4.012s ModemManager.service
          3.665s postfix.service
          3.627s SuSEfirewall2_init.service
          3.206s nscd.service
          3.187s alsa-restore.service
          3.183s avahi-daemon.service
          3.182s rc-local.service
          2.734s bluetooth.service

http://pastebin.com/WXkkRFpd

3

What happens if you run the same command (see /run/systemd/generator/lvm2-activation-net.service) manually after boot?

4

Two days ago I made »zypper up -r repo-oss« and the next time the computer took about 10 min to boot. I just waited (impatiently). The next – until today the last – update made KDE5 stable again…

Reboot and wait. Maybe it works.

5

I have the same problem. Did anybody find a solution?

Here is my:

systemd-analyze blame


   2min 603ms systemd-udev-settle.service
         59.868s lvm2-activation-net.service
         55.734s purge-kernels.service
          1.315s mysql.service
          1.111s apparmor.service
           898ms dkms.service
           776ms display-manager.service
           651ms dev-sda6.device
           637ms SuSEfirewall2.service
           493ms usr-local.mount
           492ms var-log.mount
           491ms \x2esnapshots.mount
           489ms var-lib-mailman.mount
           488ms srv.mount
           479ms opt.mount
           479ms var-tmp.mount
           478ms var-crash.mount
           468ms var-lib-named.mount
           464ms boot-grub2-x86_64\x2defi.mount
           457ms var-lib-libvirt-images.mount
           455ms var-lib-pgsql.mount
           453ms boot-grub2-i386\x2dpc.mount
           453ms apache2.service
           452ms tmp.mount

Thanks for your help

6

After searching for a long time I found the solution.

After running the commands:

systemctl mask systemd-udev-settle.service

systemctl mask lvm2-activation-net.service

the boot time of my system is back to normal. I am not sure whether this is the cleanest solution, but for me it works fine.