Not enough disk space (<141.4TB)

[jonathan-realman]

Note: This application is provided "as-is" without support or warranty. If you need help with issues, consider making a donation to support the development.

Describe the bug
When I try to take a Snapshot or set the application up I get a message that the space need would be above 140TB dispite that being more than 50 time the space on my machine
sda 8:0 0 59.6G 0 disk
├─sda1 8:1 0 477M 0 part /boot
└─sda2 8:2 0 59.2G 0 part /
sdb 8:16 0 931.5G 0 disk
└─sdb1 8:17 0 931.5G 0 part /home/jonathan/.local (excluded in the snapshot)
sdc 8:32 0 931.5G 0 disk
└─sdc1 8:33 0 931.5G 0 part /run/timeshift/backup
sdd 8:48 0 116.5G 0 disk
├─sdd1 8:49 0 100G 0 part /home
└─sdd2 8:50 0 16.5G 0 part [SWAP]

Screenshots

[Warning] Deleted invalid lock
App config loaded: /etc/timeshift.json

(timeshift-gtk:25411): GLib-GIO-CRITICAL **: 09:18:38.838: g_file_get_path: assertion 'G_IS_FILE (file)' failed

** (timeshift-gtk:25411): CRITICAL **: 09:18:38.838: tee_jee_file_system_path_combine: assertion 'path1 != NULL' failed

** (timeshift-gtk:25411): CRITICAL **: 09:18:38.838: tee_jee_file_system_dir_exists: assertion 'dir_path != NULL' failed

/dev/sdc1 is mounted at: /run/timeshift/backup, options: rw,relatime

System:

• Linux Distribution Name and Version: Arch Linux Rolling
• Desktop MATE+i3
• Application Version v20.03

[Lampwick]

I have the same issue. Also on Arch with timeshift 20.03

[jonathan-realman]

I have the same issue. Also on Arch with timeshift 20.03

with exactly the same amount of storage?

[Lampwick]

sda1 - 1.0TB - Timeshift destination
sda2 - 3.0TB - (Not to be backed up)
sdb1 - 21GB - /var
sdb2 - 3.0TB - /home (Not to be backed up)
sdc1 - 536MB - /boot
sdc2 - 512GB - /

I get the same, "Not enough disk space (<141.4TB)", message.

[nnleaf]

I have the same issue. I tried deleting a bunch of files to make room, reduced my drive to just 30gb with a 200gb drive for backup, but it still gives me the 'Not enough disk space (<140.8 TB)' error.

I'm on Arch Linux and there was a recent update with Timeshift, so I'm guessing it has something to do with that.

Edit : Tried downgrading to 19.08.1-2 with no success. Odd as Timeshift was working great a few days ago.

[Flygrounder]

Same here with timeshift 20.03 on Arch. Tried even excluding /** , but nothing has changed.

[mannp]

Getting the same on arch. Swapped to a larger disk but have the same issue. Also getting this on the new 1tb drive. All was working great until the latest update.

Synching files with rsync...
193.57% complete (00:00:00 remaining)

[Flygrounder]

I have noticed, that I also have #588 issue as well. I have a hypothesis, that it is related to this issue. If timeshift has problems with filters maybe it does not see default filters like /run/* /dev/* etc. It may lead to an infinite recursion, which can cause such a huge estimated backup size.
@jonathan-realman @Lampwick @nnleaf @mannp do you have this issue as well?

[nnleaf]

Hey everyone, I think I found the problem.

It seems most people with problems are running Arch, and I saw a post here about a dhcpcd's >9.0 update. It seems to mount some system directories in /var/lib/dhcpcd/ so excluding that folder might be the solution!

I've excluded it from Timeshift and the error dissappears, it also allows me to create a snapshot and it seems to be running without any odd errors like #591 or #590 . I will update in the morning, as I'm creating a fresh backup and it will take a few hours.

As it does exclude the dhcpcd folder, I'm not sure if there is anything important there that I would need if I needed to do a system restore, would the better solution be to downgrade dhcpcd to 8.x until a more official solution is found?

[Lampwick]

I can confirm excluding /var/lib/dhcpd/ fixed it for me too.
Not sure if the folder is necessary for system restore though.

[johncparr]

I have found that by disabling and stopping dhcpcd I could get timeshift working again. I then installed networkmanager and switched over to it for networking and I seem to have a permanent workaround/fix

[nnleaf]

Wanted to give an update. Excluding the folder worked, but I simply removed dhcpcd and used networkmanager itself. I guess with dhcpcd's changes made to 9.0, Timeshift will have to exclude that folder by default or give out a warning.

Thanks for everyone's help, glad to be using Timeshift again - I even made a pacman hook to run a snapshot before updates if anyone wants to check it out because timeshift is such a great tool.

[UltraBlackLinux]

Same issue here - 140,8 TB....
But fixed! Thank you nnleaf!

[JimDeadlock]

It's not a fix, it's a workaround. Just saying.

[gardotd426]

Yeah I also have this issue on vanilla Arch, what's weird is that it doesn't occur on Manjaro. And I run Unstable Branch of Manjaro, and the dhcpcd package versions are the same. So it's not something that dhcpcd did. Though, Manjaro actually provides a repo prebuilt package in the official repos, instead of only an AUR package, so maybe they patched it, like they've patched Pacman to allow database files larger than 25MB.

[NavySaint]

Can anyone confirm that a full system restore works properly even after excluding the aforementioned folder?

[gardotd426]

Yes.

[teejee2008]

"/var/lib/dhcpd/" has been added to list of default filters

[biohazard13]

bug still present, fortunately i found this thread

[Alzg3852]

unfortunately, problem is still not solved even after filtering out '/var/lib/dhcpd'.

[shmu26]

I started having this issue on Ubuntu 21.10
I excluded

/var/lib/dhcpcd/**
/mnt/**

but it didn't help

[shmu26]

Ubuntu 21.10 got a system update that included also a Timeshift update and this seems to have fixed it. :)
Timeshift is now on version 20.11.1