Increasing Disk Detection Timeout at Boot with Linux/Systemd

bootbtrfslinuxsystemdtimeout

I have a machine with a lot of disks, and an extra SAS controller in HBA mode.
This seems to cause Linux to think for at least 8-10 seconds once in initramfs before the disks actually appear. The timeout for disk detection is 10 seconds. This causes BTRFS/MDADM/etc fail to mount a RAID1 that I have in my system, dropping me to an emergency shell from where I can actually mount the disks and continue just fine.

My question is, how do I increase this timeout at boot from 10 seconds? Is it in systemd? is it in udev? somewhere else? I'm not sure where to start looking, and googling about this problem mostly seems to yield people looking to raise the I/O timeout or some other (scsi/lun/etc) timeouts, but I'm not looking for that.

Best Answer

I've finally found it! It's of course but a simple kernel parameter, found here https://www.kernel.org/doc/html/latest/admin-guide/kernel-parameters.html

The parameter I was specifically looking for is rootdelay, I had already tried rootwait but apparently that wasn't enough, as it still aborted the wait after 10 seconds. Now it actually does not wait the full 30 seconds specified, but only about 10-15 seconds depending on how long it takes for my disks to show up, so setting a really high value doesn't seem to hurt, although I've only set 30 for my use case, which so far seems to have completely resolved the issue!

You can add it to your kernel boot parameters in Grub or systemd-boot.

Grub: /etc/defaults/grub -> GRUB_CMDLINE_LINUX_DEFAULT="rootdelay=30 quiet"

systemd-boot: /boot/loader/entries/yourentry.cfg -> options rootdelay=30 [other options]

Related Solutions

Linux Systemd – Managing Dependencies and Boot Order

Use systemctl edit smb.service to update the dependencies.

After=dirsrv.target - Will ensure the smb.service is started after dirsrv.target.

For robustness, (which will be worth while if you're tinkering with this stuff) you may also wish to include some of the following:

Requires=dirsrv.target - Activate dirsrv.target when smb.service is activated. Will cause smb.service to fail if dirsrv.target fails.

Wants=dirsrv.target - Activate dirsrv.target when smb.service is activated. Won't cause smb.service to fail if dirsrv.target fails.

BindsTo=dirsrv.target - If dirsrv.target is deactivated, deactivate smb.service.

Source: http://www.freedesktop.org/software/systemd/man/systemd.unit.html

systemd-ui provides a GUI for systemd. Gives a good view of the state of systemd but you'll still have to use a text editor to modify the unit files.

Linux – mdadm software RAID isn’t assembled at boot during initramfs stage

You could mirror the drives with btrfs itself, instead of creating that fs on top of the software raid: mkfs.btrfs -d raid1 /dev/sdc /dev/sdd

Otherwise try:

    umount /dev/md0 if mounted
    mdadm --stop /dev/md0
    mdadm --assemble --scan
    mv /etc/mdadm/mdadm.conf /etc/mdadm/mdadm.conf.bak
    /usr/share/mdadm/mkconf > /etc/mdadm/mdadm.conf

If cat /proc/mdstat shows the correct output now then create your filesystem and mount it, use blkid to get the UUID for /dev/md0 and edit /etc/fstab accordingly.

If you are still having issues you could try this before proceeding with the above mentioned instructions:

    mdadm --zero-superblock /dev/sdc /dev/sdd
    mdadm --create /dev/md0 --level=1 --raid-devices=2 /dev/sdc1 /dev/sdd1

I tested this on a system running Debian Jessie with 3.16.0-4-amd64 kernel, and I wrote gpt partition tables to the two block devices that I mirrored together. The array is properly assembled at boot and mounted as specified.

Best Answer

Related Solutions

Linux Systemd – Managing Dependencies and Boot Order

Linux – mdadm software RAID isn’t assembled at boot during initramfs stage

Related Topic