Droidspaces - Discussion — Telegram group messages

13 November 2026

Deo K.Droidspaces v7.2.0 - Checking system requirements... [MUST HAVE] These features are required for Droidspaces to work: [✓] Root privileges [✓] Linux version [✓] PID namespace [✓] Mount namespace [✓] UTS namespace [✓] IPC namespace [✓] pivot_root syscall [✓] /proc filesystem [✓] /

did you try setting SELinux to Permissive (by setenforce 0 in Termux root shell) ?

04:30

D
Deo K.
yes, via adb
04:31
1. A
  avelith
  ADB root? If it really set it to permissive then now it should work. But if didn't, prioritise Directory-based method instead of Sparse
  04:32

Whole thread · 2 replies →

Deo K.

Deo K.same issue

yes, it is permissive

04:33

fatalcoder524

New mount API kernel 5.2 : fsopen() and friends systemd 259 and newer keep service credentials in a plain directory and ProtectProc= does nothing. On such a container Droidspaces answers EPERM for these calls so systemd takes those fallbacks instead of failing, and gives every unit and login shell LIBMOUNT_FORCE_MOUNT2=always so mount keeps using the classic call. A mount run behind an environment reset, sudo for one, needs that variable passed along. Older systemd is left alone. Loop device, ext4 CONFIG_BLK_DEV_LOOP, CONFIG_EXT4_FS: Sparse image containers cannot mount.

04:33

A
avelith
reference to my Systemd services getting failed when enabling Sandbox
04:35
D
Deo K.
Is it relavant to kernel 4.19
04:37
1. A
  avelith
  It only matters if you intended to use Systemd version 258 or above. For that you need to use droidspaces-backports without droidspaces-backports on nGKI kernel Systemd 258 and above won't boot at all meaning goodbye to Arch Linux, Fedora, etc.
  04:37

Whole thread · 3 replies →

avelith

on my kernel 4.14, the moment I enable 'Allow Sandboxing' or use noseccomb privilege mode, every core services of Systemd fails deepseek says it is a kernel bug (super grab) that Droidspaces can't workaround it

04:36

avelith

Photo

though I have those backports/patches 🗿

04:40

fatalcoder524

Deo K.same issue

val mountOptions = "loop,rw,nodelalloc,noatime,nodiratime,init_itable=0" val mountCmd = "${Constants.BUSYBOX_BINARY_PATH} mount -t ext4 -o $mountOptions ${quote(imgPath)} ${quote(mountPoint)} || " + "mount -t ext4 -o $mountOptions ${quote(imgPath)} ${quote(mountPoint)}" Failing command

04:46

fatalcoder524

Did you try daemon mode??

04:48

D
Deo K.
Photo
yes
04:56

Whole thread · 1 reply →

Deo K.

Increasing CONFIG_BLK_DEV_LOOP_MIN_COUNT fixed the issue

🤔1

04:57

S
Subhu.prop #m55xq
How much did u set it to?
05:19
1. D
  Deo K.
  64
  05:19

Whole thread · 2 replies →

Deo K.

But i am not sure if it is correct way to do so

04:57

Deo K.

losetup -f before installation works as well

05:02

D
Deo K.
According to claude: The command that fails is the installer's loop mount: busybox mount -t ext4 -o loop,rw,nodelalloc,noatime,nodiratime,init_itable=0 rootfs.img rootfs , with the system mount as its fallback. With -o loop, mount needs a free loop device, and on your phone all of them are in use after boot, so the kernel has to create a new one mid-mount. The new device's node in /dev/block shows up a moment too late, and the mount gives up before it exists. Running losetup -f first, or booting with 64 pre-created devices, means a free device with its node already exists, so the mount never has to create one.
05:18

Whole thread · 1 reply →

Deo K.

maybe, the app should handle such cases?

05:03