You are not logged in.
Hi I need some help.
I've followed the guide to change from init to systemd:
https://wiki.archlinux.org/index.php/Sy … stallation
And now my computer takes more thant 2 minutes to boot to kdm, on init it only take 12 seconds.
On the logs I can only see this, why it takes 9 seconds and then without any error bumps to 95 seconds? I cannot understand what is going on with systemd.
Any clues?? logs below:
Aug 20 21:31:59 localhost kernel: [ 8.820304] Console: switching to colour frame buffer device 240x75
Aug 20 21:31:59 localhost kernel: [ 8.824568] fb0: radeondrmfb frame buffer device
Aug 20 21:31:59 localhost kernel: [ 8.824569] drm: registered panic notifier
Aug 20 21:31:59 localhost kernel: [ 8.824574] [drm] Initialized radeon 2.16.0 20080528 for 0000:01:00.0 on minor 0
Aug 20 21:31:59 localhost kernel: [ 8.877167] input: HDA ATI HDMI HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input15
Aug 20 21:31:59 localhost kernel: [ 9.073313] EXT4-fs (sdb2): mounting ext2 file system using the ext4 subsystem
Aug 20 21:31:59 localhost kernel: [ 9.074452] EXT4-fs (sdb2): mounted filesystem without journal. Opts: (null)
Aug 20 21:31:59 localhost [ 9.223791] systemd-journald[142]: Received SIGUSR1
Aug 20 21:31:59 localhost kernel: [ 95.619960] skge 0000:04:00.0: eth0: enabling interface
Aug 20 21:31:59 localhost kernel: [ 95.624363] ADDRCONF(NETDEV_UP): eth0: link is not ready
Aug 20 21:31:59 localhost kernel: [ 97.325063] skge 0000:04:00.0: eth0: Link is up at 100 Mbps, full duplex, flow control none
Aug 20 21:31:59 localhost kernel: [ 97.326393] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Aug 20 21:32:06 localhost kernel: [ 104.466952] EXT4-fs (sdd1): mounted filesystem with ordered data mode. Opts: (null)
Aug 20 22:37:36 localhost kernel: [ 0.000000] Initializing cgroup subsys cpuset
Aug 20 22:37:36 localhost kernel: [ 0.000000] Initializing cgroup subsys cpu
Aug 20 22:37:36 localhost kernel: [ 0.000000] Linux version 3.4.9-1-ARCH (tobias@T-POWA-LX) (gcc version 4.7.1 20120721 (prerelease) (GCC) ) #1 SMP PREEMPT Wed Aug 15 18:59:31 CEST 2012Last edited by playmobil (2012-08-20 22:38:36)
Há nos confins da Ibéria um povo que nem se governa nem se deixa governar ( Caio Júlio César, [ 100 a.C- 44 a.C])
Offline
It might be helpful to boot with systemd debug logging enabled. That may give you an extra clue.
Simply attach something like this to your kernel args:
systemd.log_level=debug systemd.log_target=kmsg log_buf_len=1MEdit: Also, don't forget to stick output in code tags!
Last edited by Zancarius (2012-08-20 22:31:21)
He who has no .plan has small finger.
~Confucius on UNIX.
Offline
Hi, thanks Zancarius, sorry for forgetting the code tags.
Hum i've added the line to the kernel args.
And now I can see what is causing the delay, but I'm more confused.
Systemd is trying to make a fsck on a Windows partition and after 95 seconds it gives a timeout:
9.470060] systemd-journald[142]: Received SIGUSR1
Aug 20 23:30:57 localhost [ 95.907209] systemd[1]: Job dev-sda3.device/start timed out.
Aug 20 23:30:57 localhost [ 95.907413] systemd[1]: Job dev-sda3.swap/start failed with result 'dependency'.
Aug 20 23:30:57 localhost [ 95.907419] systemd[1]: Job dev-sda3.device/start failed with result 'timeout'.fdisk -l
Device Boot Start End Blocks Id System
/dev/sda1 * 2048 206847 102400 7 HPFS/NTFS/exFAT
/dev/sda2 206848 234438655 117115904 7 HPFS/NTFS/exFATSo why is systemd trying to fdisk windows partition, and how to disable.
Also anyone knows why on every reboot systemd makes a fsck ??
Thank you
Há nos confins da Ibéria um povo que nem se governa nem se deixa governar ( Caio Júlio César, [ 100 a.C- 44 a.C])
Offline
That is kind of odd. If you're using ntfs-3g, you might be able to add norecover to your options to cause it to fail mounting if NTFS is still marked as dirty. That'll at least help narrow down what that problem might be. Otherwise, you could try commenting out those lines in your fstab and see if it'll boot without delay (mostly to confirm that it is, in fact, NTFS causing grief).
He who has no .plan has small finger.
~Confucius on UNIX.
Offline
Thanks Zancarius, you are right ![]()
I've found also this bug report:
https://bugzilla.redhat.com/show_bug.cg … &id=694647
So systemd do not like drives on fstab that are defined as /dev/sd*, because on boot the drive can change.
To solve the problem i've added a label to swapp and voila, fast boot again ![]()
Problem solved, thank you for helping out.
Há nos confins da Ibéria um povo que nem se governa nem se deixa governar ( Caio Júlio César, [ 100 a.C- 44 a.C])
Offline
That's good to know! I use LVM but don't mount using labels/UUIDs. Although, all my Windows partitions are mounted with the device. I wonder if it's because my BIOS is fairly well-behaved and doesn't change around drive order? Interesting nevertheless.
Don't forget to mark your thread solved if it looks like using labels fixed your problem!
He who has no .plan has small finger.
~Confucius on UNIX.
Offline
This is a support question. Moving from Arch Discussion to Applications & Desktop Environments.
aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies
Offline