Installer crash when clicking on "Choose file" button
-
Hello,
UBports Installer v. 0.9.1-beta installed as 'snap' package on openSUSE Leap 15.1 (same behaviour on 15.2) with KDE5.
The installer starts up normally. I can select the phone model (Xiaomi Mi A2), the target OS (UT) and the channel to use (16.04/arm64/android9/stable).
It then tells me to download a 'vendor.img' file, which I did, and to select it for installation by clicking on the "Choose file" button, which is presumably meant to pop-up an "Open file" dialog box.
As soon as I click on that "Choose file" button, the UBPI application crashes with the following error message:Gtk-Message: 15:50:12.799: Failed to load module "canberra-gtk-module" info: Welcome to the UBports Installer version 0.9.1-beta! info: device selected: jasmine_sprout info: Installing Ubuntu Touch on your Xiaomi Mi A2 (jasmine_sprout) info: configuring... info: settings: {"bootstrap":true,"channel":"16.04/arm64/android9/stable"} (ubports-installer:10838): Gtk-WARNING **: 15:50:47.936: Could not load a pixbuf from /org/gtk/libgtk/icons/16x16/status/image-missing.png. This may indicate that pixbuf loaders or the mime database could not be found. ** Gtk:ERROR:../../../../gtk/gtkiconhelper.c:494:ensure_surface_for_gicon: assertion failed (error == NULL): Failed to load /org/gtk/libgtk/icons/16x16/status/image-missing.png: Unrecognized image file format (gdk-pixbuf-error-quark, 3) Bail out! Gtk:ERROR:../../../../gtk/gtkiconhelper.c:494:ensure_surface_for_gicon: assertion failed (error == NULL): Failed to load /org/gtk/libgtk/icons/16x16/status/image-missing.png: Unrecognized image file format (gdk-pixbuf-error-quark, 3) Aborted (core dumped)I've already tried the following, based on recommendations found on different forums (unrelated to UBPI):
# update-mime-database /usr/share/mime # gdk-pixbuf-query-loaders --update-cache # gtk-update-icon-cacheI've also tried to 'unset XDG_DATA_DIRS' before launching 'ubports-installer'. None of this made any difference.
I've also checked using 'ldd' that the 'ubports-installer' was able to correctly resolve all the shared libraries it was linked against, and I didn't notice any problem there.
And to conclude, I've run 'ubports-installer' through 'strace' to try and catch the underlying cause of the error, but that was inconclusive (because the application is multi-threaded, perhaps?).Can anyone here offer advice?
TIA
Ph. A. -
@phandersson
Try previous version of installer, 0.9.1 is known to have issues. -
@keneda said in Installer crash when clicking on "Choose file" button:
@phandersson
Try previous version of installer, 0.9.1 is known to have issues.Many thanks for your suggestion, @keneda, I'll gladly do that. But there doesn't seem to be any snap channel offering it. Here is the relevant part of 'snap info ubports-installer':
commands: - ubports-installer snap-id: mq0sTMV7a8744UTRBwQHimStAqsGzbjj tracking: latest/stable refresh-date: yesterday at 15:18 CEST channels: latest/stable: 0.9.1-beta 2021-10-04 (371) 152MB - latest/candidate: ↑ latest/beta: 0.9.1-beta 2021-10-04 (371) 152MB - latest/edge: 0.9.1-beta 2021-10-04 (371) 152MB - installed: 0.9.1-beta (371) 152MB -All available channels end up pointing to the same release, apparently.
Or am I missing something?Xiaomi Mi A2 (16.04 OTA-25/stable) with 2 SIMs
Daily driver: Google Pixel 3a (20.04 OTA-8/stable) [was: Nokia N900 (Maemo) from 2009]. -
@phandersson Sorry this is a bit rushed, but here is the link for https://github.com/ubports/ubports-installer/releases/tag/0.8.8-beta 0.8.8 You should be able to find all the previous versions of the installer or most anyway on github. Just don't have link handy. Any issue will be back later.
-
@phandersson
You can find all ubports installer versions here :
https://github.com/ubports/ubports-installer/releases -
@keneda and @Lakotaubp
I've downloaded 'ubports-installer_0.8.9-beta_linux_x86_64.AppImage', saved it locally to '~/platform-tools', made it executable and symlinked it as 'ubports-installer' (I've also removed the 0.9.1 snap that was causing trouble).This time, I can select the downloaded 'vendor.img' without issue, and the 1st step of the flashing proceeded without issue.
I was then instructed to reach recovery mode and to click continue, which I did (the phone shows a screen with "RECOVERY" and a 3-circles logo -- at the bottom, it also says "Halium 9" and "Slot b"), but the installer is now waiting forever to reconnect to the phone:
Waiting for device... adb is scanning for devicesThere was a note in the installer to create custom udev rules, but I skipped that as I already have similar rules in place (thanks to the openSUSE 'android-udev-rules' package) although the associated group is not "plugdev" but "adbusers" (a group which my account is a member of, obviously).
But I now notice that these old rules don't seem to apply:$ lsusb [...] Bus 004 Device 008: ID 18d1:0000 Google Inc. [...] $ ll ./bus/usb/004/008 crw-rw-r-- 1 root root 189, 391 Oct 10 15:32 ./bus/usb/004/008I'll install the recommended rules, put the phone back to fastboot mode and restart the installation process from the beginning, I guess.
-
@phandersson
Even weirder: the installer now tells me that the Mi A2 "jasmine_sprout" is not supported!As said in the previous post, I've now installed the required custom udev rules on my PC, I've created the "plugdev" group that doesn't exist by default on openSUSE and added my user account to it, then i restarted the PC.
I've brought the phone back to the "fastboot" splash screen and I've restarted the installer. The phone is initially detected as "jasmine", which it says is not supported, and asks me to select the model manually (as I already did last time, BTW), so I select the Xiaomi Mi A2 in the list, and it tells me "Device not supported / The device jasmine_sprout is not supported":
$ ./ubports-installer info: Welcome to the UBports Installer version 0.8.9-beta! warn: Please update: https://devices.ubuntu-touch.io/installer/?package=AppImage info: device detected: jasmine warn: The device jasmine is not supported! info: device selected: jasmine_sprout warn: The device jasmine_sprout is not supported!This time, the USB device ownership correctly reflects the new udev rules, though:
$ lsusb [...] Bus 004 Device 003: ID 18d1:d00d Google Inc. [...] $ ll /dev/bus/usb/004/003 crw-rw-rw-+ 1 root plugdev 189, 386 Oct 10 16:14 /dev/bus/usb/004/003 -
@phandersson
Just retried by only connecting the phone when the installer asks for it -- same behaviour.Even though the installer doesn't issue an explicit error message about it, I wonder whether the problem doesn't result from an inability to locate / download the 'jasmine_sprout'-specific configuration file. The first time I ran the installer, that step succeeded, though.
-
@phandersson
OK -- found a way around it: do NOT plug the phone in when the installer requests it. Select model manually, select OS and channel, let the installer download the required files. Only then plug the phone in and click "Install".So, for a 2nd time, the first phase of flashing went well. I've rebooted the phone to the recovery system, and tried to proceed with the 2nd phase. This time, the phone is correctly detected and installation continues, but it soon fails with the following error:
[...] info: Downloaded file 13 of 14 info: Downloaded file 14 of 14 error: Error: adb:preparesystemimage: Error: {"error":{"code":1,"cmd":"adb -P 5037 shell mkdir -p /cache/recovery"},"stderr":"mkdir: '/cache/recovery': Operation not permitted"} stack trace: Error: {"error":{"code":1,"cmd":"adb -P 5037 shell mkdir -p /cache/recovery"},"stderr":"mkdir: '/cache/recovery': Operation not permitted"} at /tmp/.mount_ubporthyGI7P/resources/app.asar.unpacked/node_modules/promise-android-tools/lib/module.cjs:117:20 at ChildProcess.exithandler (child_process.js:319:5) at ChildProcess.emit (events.js:315:20) at maybeClose (internal/child_process.js:1021:16) at Socket.<anonymous> (internal/child_process.js:443:11) at Socket.emit (events.js:315:20) at Pipe.<anonymous> (net.js:674:12)In the dialog box titled "Yikes!", there is a "Try again" button. I pressed it but it resulted in the same error.
-
@phandersson : I don't own this device, nor have any knowledge of Open Suse. I'm on Ubuntu, but in my case studying the logfiles in .cache/ubports (on PC) proved useful. There were more errors.
-
@alter
Thanks for the tip.
Here is the relevant section from the log file:[...] {"level":"info","message":"Downloaded file 14 of 14"} {"level":"verbose","message":"running core action write"} {"level":"verbose","message":"running adb action wait"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"wait-for-any-any\"]}"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"get-state\"],\"stdout\":\"recovery\"}"} {"level":"verbose","message":"running adb action preparesystemimage"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"shell\",\"mount -a\"],\"error\":{\"message\":\"Command failed: adb -P 5037 shell mount -a\\nmount: '/dev/block/platform/soc/c0c4000.sdhci/by-name/vendor_b'->'/vendor': No such file or directory\\nmount: '/devices/soc/c084000.sdhci/mmc_host*'->'/storage/sdcard1': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/modem_b'->'/vendor/firmware_mnt': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/bluetooth_b'->'/vendor/bt_firmware': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/dsp_b'->'/vendor/dsp': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/persist'->'/mnt/vendor/persist': No such file or directory\\nmount: '/devices/soc/a800000.ssusb/a800000.dwc3/xhci-hcd.0.auto*'->'/storage/usbotg': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/boot_b'->'/boot': No such file or directory\",\"code\":1},\"stderr\":\"mount: '/dev/block/platform/soc/c0c4000.sdhci/by-name/vendor_b'->'/vendor': No such file or directory\\nmount: '/devices/soc/c084000.sdhci/mmc_host*'->'/storage/sdcard1': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/modem_b'->'/vendor/firmware_mnt': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/bluetooth_b'->'/vendor/bt_firmware': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/dsp_b'->'/vendor/dsp': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/persist'->'/mnt/vendor/persist': No such file or directory\\nmount: '/devices/soc/a800000.ssusb/a800000.dwc3/xhci-hcd.0.auto*'->'/storage/usbotg': No such file or directory\\nmount: '/dev/block/bootdevice/by-name/boot_b'->'/boot': No such file or directory\"}"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"shell\",\"cat /etc/recovery.fstab\"],\"stdout\":\"# Android fstab file.\\n# The filesystem that contains the filesystem checker binary (typically /system) cannot\\n# specify MF_CHECK, and must come before any filesystems that do specify MF_CHECK\\n\\n#TODO: Add 'check' as fs_mgr_flags with data partition.\\n# Currently we dont have e2fsck compiled. So fs check would failed.\\n\\n# A/B fstab.qcom variant\\n#<src> <mnt_point> <type> <mnt_flags and options> <fs_mgr_flags>\\n/dev/block/bootdevice/by-name/system / ext4 ro,barrier=1,discard wait,slotselect\\n/dev/block/bootdevice/by-name/userdata /data ext4 nodev,noatime,nosuid,barrier=1,noauto_da_alloc,discard,lazytime wait,check\\n/devices/soc/c084000.sdhci/mmc_host* /storage/sdcard1 vfat nodev,nosuid wait,voldmanaged=sdcard1:auto,encryptable=footer\\n/dev/block/bootdevice/by-name/misc /misc emmc defaults defaults\\n/dev/block/bootdevice/by-name/modem /vendor/firmware_mnt vfat ro,shortname=lower,uid=0,gid=1000,dmask=227,fmask=337,context=u:object_r:firmware_file:s0 wait,slotselect\\n/dev/block/bootdevice/by-name/bluetooth /vendor/bt_firmware vfat ro,shortname=lower,uid=1002,gid=3002,dmask=227,fmask=337,context=u:object_r:bt_firmware_file:s0 wait,slotselect\\n/dev/block/bootdevice/by-name/dsp /vendor/dsp ext4 ro,nodev,nosuid,barrier=1 wait,slotselect\\n/dev/block/bootdevice/by-name/persist /mnt/vendor/persist ext4 nodev,noatime,nosuid,barrier=1 wait\\n\\n/devices/soc/a800000.ssusb/a800000.dwc3/xhci-hcd.0.auto* /storage/usbotg vfat nodev,nosuid wait,voldmanaged=usbotg:auto\\n\\n/dev/block/bootdevice/by-name/boot /boot ext4 ro,barrier=1,discard wait,slotselect\"}"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"shell\",\"rm -rf /cache/*\"]}"} {"level":"command","message":"exec: {\"cmd\":[\"adb\",\"-P\",5037,\"shell\",\"mkdir -p /cache/recovery\"],\"error\":{\"message\":\"Command failed: adb -P 5037 shell mkdir -p /cache/recovery\\nmkdir: '/cache/recovery': Operation not permitted\",\"code\":1},\"stderr\":\"mkdir: '/cache/recovery': Operation not permitted\"}"} {"level":"debug","message":"attempting to handle Error: {\"error\":{\"code\":1,\"cmd\":\"adb -P 5037 shell mkdir -p /cache/recovery\"},\"stderr\":\"mkdir: '/cache/recovery': Operation not permitted\"}"} {"level":"error","message":"Error: adb:preparesystemimage: Error: {\"error\":{\"code\":1,\"cmd\":\"adb -P 5037 shell mkdir -p /cache/recovery\"},\"stderr\":\"mkdir: '/cache/recovery': Operation not permitted\"}\nstack trace: Error: {\"error\":{\"code\":1,\"cmd\":\"adb -P 5037 shell mkdir -p /cache/recovery\"},\"stderr\":\"mkdir: '/cache/recovery': Operation not permitted\"}\n at /tmp/.mount_ubporthyGI7P/resources/app.asar.unpacked/node_modules/promise-android-tools/lib/module.cjs:117:20\n at ChildProcess.exithandler (child_process.js:319:5)\n at ChildProcess.emit (events.js:315:20)\n at maybeClose (internal/child_process.js:1021:16)\n at Socket.<anonymous> (internal/child_process.js:443:11)\n at Socket.emit (events.js:315:20)\n at Pipe.<anonymous> (net.js:674:12)"} {"level":"warn","message":"restart after error"} [...]The last entry shown in the extract ("restart after error") corresponds to the moment I pressed on the "Try again" button.
Two lines before the 'mkdir' command resulting in an "operation not permitted", we can also see that a "mount -a" ends in error, although that one is not considered fatal, apparently.
-
@phandersson : I don't know exactly how to read the file. The first error at command mount -a says “no such file or directory”. That already doesn't sound right. I hope someone here knows a solution for it!
-
@alter Thinking back about the whole sequence, I'm a bit concerned that my first failed attempt at flashing UT (before adapting the udev rules) may have left the phone in an "unexpected" state, which in turn caused the 2nd flashing attempt to fail with these ominous error messages.
It is perhaps best at this point that use TWRP again to put Android9 back on the phone (in both slots), make sure the phone is bootable again, and retry the whole 'ubports-installer' process again.
-
@phandersson : I think I would also start over by doing just that. It seems errors keep popping up. Maybe also try the 0.9.0 version of installer then.
-
@alter Looking at its Changelog, release 0.9.0 seems to be the one that introduced the redesigned "File open" dialogue:
"Implement XDG desktop portals to enable file picker dialogs for manual file downloads in the snap build"
Also, if I understand their respective Changelogs correctly, neither 0.9.0 or 0.9.1 changed anything to the actual phone installation procedure. But please correct me if I'm wrong (or if the Changelog doesn't cover everything).
Xiaomi Mi A2 (16.04 OTA-25/stable) with 2 SIMs
Daily driver: Google Pixel 3a (20.04 OTA-8/stable) [was: Nokia N900 (Maemo) from 2009]. -
@phandersson : I'm not a developer, but from what I've heard the version with problems is 0.9.1.
-
@alter I can certainly check v0.9.0 as well, since with 0.9.1 I could reproduce the crash without even connecting the phone
Thanks for the suggestion.
-
@phandersson : You're welcome! I hope the UT install will go well for you this time!
-
@phandersson If you feel the need to run through the installer again try with 0.8.8 or 0.8.9. That seems to be solving most of the installer issues at present. https://github.com/ubports/ubports-installer/releases
-
@lakotaubp So just a simple question: As current installer version (0.9.1) seems definitely problematic, should one revert to 0.8.9 in Ubuntu repositories - including Ubuntu Software, Gnome Software, etc...? (Pending a future more stable release).
