mtp under [arch]linux: mtpfs vs gvfs-mtp [mount update]

Wanted to copy some file from a Moto G to my laptop. Installed mtpfs. Mounting took minutes and failed. As a last hope I tried gvfs-mtp. I don't know how to use the gvfs-* CLI though. But it's supposed to work out of the box in GTK apps. So Chromium + Ctrl-O. I can see the smartphone listed. Walk through the phone FS. It's instantaneous, snappy as I like it. But what about the CLI ?

Well gvfs-info says nothing. There's gvfs-mount, which says nothing. gvfs-mount --list outputs the mounted drives on this machine. And the last one is mtp://[usb:005,009]/ ... Hmm maybe gvfs-ls takes an URI ?

$ gvfs-ls mtp:/// # Error: The specified location is not mounted
$ gvfs-ls mtp://  # Error: The specified location is not mounted
$ gvfs-ls mtp:///sdcard # same as above

could the [usb:...] be mandatory ? odd syntax but alas

$ gvfs-ls mtp://[usb:005,009]/
Internal Storage # yay

$ gvfs-tree mtp://[usb:005,009]/ | head -n 32
mtp://[usb:005,009]/
`-- Internal Storage
    |-- .CM_Cloud
    |   |-- cm
    |   |   |-- cm_cloud_sdk_expand.db
    |   |   `-- cm_cloud_sdk_expand.db-journal
    |   `-- cmqp
    |       |-- cm_cloud_sdk_expand.db
    |       `-- cm_cloud_sdk_expand.db-journal
    |-- .DataStorage
    |   `-- ContextData.xml
    |-- .UTSystemConfig
    |   `-- Global
    |       `-- Alvin2.xml
    |-- .com.zwsoft.zwcad
    |   |-- local_Thumb
    |   |   |-- Fonts
    |   |   `-- Sample Drawings
    |   |       |-- Villa.dwg.png
    |   |       |-- ar_house_2.dwg.png
    |   |       |-- crane.dwg.png
    |   |       `-- l_1a.dwg.png
    |   |-- resources
    |   |   |-- adinit.dat
    |   |   |-- arialuni.ttf
    |   |   |-- plotstyle
    |   |   |   |-- Icad.ctb
    |   |   |   |-- Monochrome.ctb
    |   |   |   |-- zwcad-Color.stb
    |   |   |   |-- zwcad.ctb
    |   |   |   `-- zwcad.stb
    |   |   |-- zwcad.dwt

Didn't try to copy anything yet, I guess it won't be worse than doing a full http roundtrip over wifi as I did before.

ps: there was a side effect of trying gvfs through GTK file open dialog. It mounted the volume on click. Thus the ability to use gvfs-* afterwards.

So there's a few things to do in cli beforehand.

source: http://stackoverflow.com/questions/483460/how-to-mount-from-command-line-like-the-nautilus-does

$ gvfs-mount --list
Volume(0): XT1032
  Type: GProxyVolume (GProxyVolumeMonitorMTP)

This shows XT1032 (Moto G) is detected but not mounted. What we want is the following:

$ gvfs-mount --list
Volume(0): XT1032
  Type: GProxyVolume (GProxyVolumeMonitorMTP)
  Mount(0): XT1032 -> mtp://[usb:005,013]/
    Type: GProxyShadowMount (GProxyVolumeMonitorMTP)
Mount(1): mtp -> mtp://[usb:005,013]/
  Type: GDaemonMount

Usually to mount a 'storage' device, one uses /dev/sd*. But there's no sd* node for the XT1032.
gvfs-mount --list --info lists a bit more info (sic) (errata: --list -i)

$ Volume(0): XT1032
  Type: GProxyVolume (GProxyVolumeMonitorMTP)
  ids:
   unix-device: '/dev/bus/usb/005/013'
  activation_root=mtp://[usb:005,013]/
  themed icons:  [phone]
  symbolic themed icons:  [phone-symbolic]  [phone]
  can_mount=1
  can_eject=0
  should_automount=1

And trying with the /dev/bus/... path

$ gvfs-mount -d /dev/bus/usb/005/013
Mounted /dev/bus/usb/005/013 at /run/user/1000/gvfs/mtp:host=%5Busb%3A005%2C013%5D

$ gvfs-ls mtp://[usb:005,013]/                                                                                                                  
Internal Storage

# yay

usb key replicate

* usb key duplicate

* partition table

  manual cgdisk

* rsync partitions

  rsync -auv --progress <source></source> <target>

* rsync caveat : hard links, sparse file

  warning: some sparse files may have huge virtual size
    rsync, will attempt to expand them fully
    i.e: docker devicemapper (60MB on disk, virtually 100G)

  warning: some programs have one fat binary with hard links
    rsync, will attempt to copy them fully
    i.e: git, which has 114 hard links with different names

* BIOS boot (ef02) partition

  dd if=/dev/sdX1 of=/dev/sdY1 bs=1M # simply

* grub

  sudo grub-install --target=i386-pc --debug --boot-directory=/<root-mountpoint>/boot/ /dev/sdY

  warning: do not mess the device names and mountpoint names as it may modify your host system grub config


* grub UUID

  grub menuentry refers to source key UUID

* adjust root UUIDs
  /etc/fstab still refers to source key UUID

  blkid /dev/sdX? &gt;&gt; /etc/fstab
  vi /etc/fstab

  (some sed-fu would be nice)</root-mountpoint></target>

nixos 16.03 live iso boot "directly" from grub

I hate extracting iso images. I prefer to run them through grub loopback. Nixos 16.03 release is a few days old, I couldn't resist. Alas, booting it this way drops me in a rescue shell.

The initrd is set up to only have one layer of virtual fs. So it assumes the squashfs nix-store would be in /mnt-root/iso and complains since it is not.

So after trying to MITM myself and mount the desired FS where they should be, I replicate the final steps of the nixos initrd init script. Kernel panic.

I'm dumbfounded, I need to be more thorough. The init scripts reads a fsinfo file (fstab-like in a different format say) to prepare the new root for the real stage 2 kernel.

btw, my mistake was to simply mount the nix-store squashfs file in /mnt-root/nix/.ro-store and then symlink /mnt-root/nix/store to it. It works only partially as the stage2 kernel find its init and proceed to run systemd but hangs after a few services doing nothing. Only option: CTRL-ALT-DEL. (I even tried to instrument stage2 systemd into booting the rescue target, hoping for a rescue shell to investigate the hang .. no luck). Then I realize that the fsinfo mounts the squashfs in nix/.ro-store BUT, also prepares a unionfs in nix/store ! Maybe stage2 is waiting for a unionfs mount in nix/store and my symlink is confusing things up.

So I insert the missing layer in there. And just sh /init. It fails horribly saying 'cannot stop udev/queue' then 'Killed'. Not thrilled.

I extract the final steps of the init script (populate /mnt-root with proc,sys,run,dev; kill a few things; restore the path for `modprobe` in the new /proc/sys/kernel/modprobe value). And then exec switch_root ...

I can now enjoy nixos 16.03 live booting directly from grub. If you allow directly to allow user intervention .. heh.