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

Re: systemd messages relating to non-existent devices



Richard Hector <richard@walnut.gen.nz> writes:

> On 7/08/19 3:16 AM, Kushal Kumaran wrote:
>> Richard Hector <richard@walnut.gen.nz> writes:
>> 
>>> Hi all,
>>>
>>> I'm getting messages like this in my logs:
>>>
>>> Aug  6 13:16:18 akl-host3 systemd[1]: dev-xvda9.device: Job
>>> dev-xvda9.device/start timed out.
>>> Aug  6 13:16:18 akl-host3 systemd[1]: Timed out waiting for device
>>> dev-xvda9.device.
>>> Aug  6 13:16:18 akl-host3 systemd[1]: Dependency failed for /dev/xvda9.
>>> Aug  6 13:16:18 akl-host3 systemd[1]: dev-xvda9.swap: Job
>>> dev-xvda9.swap/start failed with result 'dependency'.
>>> Aug  6 13:16:18 akl-host3 systemd[1]: dev-xvda9.device: Job
>>> dev-xvda9.device/start failed with result 'timeout'.
>>>
>>> /dev/xvda9 used to be my swap device, but no longer exists due to VPS
>>> weirdness (it's now on xvdi). I've changed my /etc/fstab to suit.
>>>
>>> Why does systemd keep trying to do stuff with it?
>>>
>>> I have the same issue with former LVM volumes on other systems as well.
>>>
>>> I suspect a reboot might fix it, but where is systemd keeping this info
>>> around, and why, and how can I stop it?
>>>
>> 
>> Run systemctl daemon-reload to regenerate the systemd mount units from
>> changed fstab file.
>> 
>
> Thanks - I thought I'd done that, but it must have been on one of the
> others. Hopefully the messages will stop now.
>
> I'm unclear what it's actually trying to do - (re)enable the swap space
> (or (re)mount the filesystem) that it thinks is supposed to be there?
>
> I'm not sure I like that; sometimes I deliberately umount a filesystem,
> and I don't want it remounted automatically while I might be doing some
> kind of maintenance.
>

I haven't had systemd attempt to automatically mount filesystems I've
explicitly unmounted.  The corresponding systemd mount unit just goes to
`inactive` state, as if I'd run `systemctl stop` on it.  swap might
behave differently.

-- 
regards,
kushal


Reply to: