[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: How can I force a full fsck on a remote system at next reboot?



On 03/13/2015 02:38 PM, Michael Biebl wrote:
Am 2015-03-13 19:35, schrieb Michael Biebl:

Maybe [1] has the missing clue. If your mentry is under the Advanced
menu, you need to use a special syntax.
I just tried the following
1. Set GRUB_DEFAULT=saved, as you did and ran update-grub
2. Checked which menu entries I had:
    grep menuentry /boot/grub/grub.cfg
3. Constructed the default entry:
    grub-set-default "Erweiterte Optionen für Debian GNU/Linux>Debian
GNU/Linux, with Linux 3.16.0-4-amd64 (recovery mode)"

Note how I explicitly specified the kernel as a subcomponent of the
advanced menu.
(sorry for the Geman locale)

I just noticed, that Elimar responded about the same time: He correctly
mentions, that you can also use numeric identifiers here.
The important part is, that you need to make sure, if the menu entry is
in submenu, you specify it like that.


Hi, Michael.

The problem was that I had read while searching that grub started with '0' as its first menu number and that grub2 started with '1'. I was using '7' when I should have used '6' (the last entry in my menu, which is the one that runs fsck).

I don't know why grub-set-default wasn't using the actual quoted menu entry. I was probably not paying enough attention to how some characters might need to be escaped in that environment.

I'll probably just be using grub-reboot instead of grub-set-default when I'm doing this manually on a remote system. Though I'd have still been typing '7' under the mistaken notion that the menu numbering started with '1'.

Your advice has been a real help since even reverting to sysvinit wouldn't have been of any help to me, since I wasn't using "shutdown -F" in the first place, and now wouldn't choose to use it after learning from you how it actually accomplishes the running of fsck. (I imagine, at least, that the initramfs-tools update may have had the same effect on my tune2fs strategy under sysvinit as it has had under systemd-sysv. (?) Anyway, I'm not interested enough to revert a system to test the hypothesis unless that's called for in following the bug report.

Many thanks for your help, Michael.

Best,
JP


Reply to: