Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Too many devices mounted on XperiaXZ2c@SFOS4.0.1 #41

Closed
Olf0 opened this issue May 28, 2021 · 25 comments
Closed

Too many devices mounted on XperiaXZ2c@SFOS4.0.1 #41

Olf0 opened this issue May 28, 2021 · 25 comments
Assignees
Labels
enhancement New feature or request

Comments

@Olf0
Copy link
Owner

Olf0 commented May 28, 2021

Environment

  • Device: Xperia XZ2 Compact ("8314"?)
  • SFOS version: "clean flash 4.0.1.48, no modifications."

Issue description

"Storage is mounted normally until installing mount-sdcard, then the issue apears." from this OpenRepos thread, which refers back to this FSO thread.

Open question

Which version of mount-sdcard was used (might have been v1.6.1 and / or v1.7.2)?

Steps for a first analysis

Please execute the following commands:
devel-su # become root
ls -l /dev/mmcblk[1-9]* /dev/sd[a-z]* /dev/sr* | tee ls-l_dev.txt
Then upload the resulting file ls-l_dev.txt here, or copy & paste its whole content here.

@levone1
Copy link

levone1 commented May 29, 2021

open question answer: yes, both versions.

@levone1
Copy link

levone1 commented May 29, 2021

output:

brw-rw---- 1 root disk 8, 0 May 29 14:24 /dev/sda
brw-rw---- 1 root disk 8, 1 May 29 14:25 /dev/sda1
brw-rw---- 1 root disk 8, 10 May 29 14:24 /dev/sda10
brw-rw---- 1 root disk 8, 11 May 29 14:24 /dev/sda11
brw-rw---- 1 root disk 8, 12 May 29 14:24 /dev/sda12
brw-rw---- 1 root disk 8, 13 May 29 14:24 /dev/sda13
brw-rw---- 1 root disk 8, 14 May 29 14:24 /dev/sda14
brw-rw---- 1 root disk 8, 15 May 29 14:24 /dev/sda15
brw-rw---- 1 root disk 259, 0 May 29 14:24 /dev/sda16
brw-rw---- 1 root disk 259, 1 May 29 14:24 /dev/sda17
brw-rw---- 1 root disk 259, 2 May 29 14:24 /dev/sda18
brw-rw---- 1 root disk 259, 3 May 29 14:24 /dev/sda19
brw-rw---- 1 root root 8, 2 May 29 14:24 /dev/sda2
brw-rw---- 1 root disk 259, 4 May 29 14:24 /dev/sda20
brw-rw---- 1 root disk 259, 5 May 29 14:24 /dev/sda21
brw-rw---- 1 root disk 259, 6 May 29 14:24 /dev/sda22
brw-rw---- 1 root disk 259, 7 May 29 14:24 /dev/sda23
brw-rw---- 1 root disk 259, 8 May 29 14:24 /dev/sda24
brw-rw---- 1 root disk 259, 9 May 29 14:24 /dev/sda25
brw-rw---- 1 root disk 259, 10 May 29 14:24 /dev/sda26
brw-rw---- 1 root disk 259, 11 May 29 14:24 /dev/sda27
brw-rw---- 1 root disk 259, 12 May 29 14:24 /dev/sda28
brw-rw---- 1 root disk 259, 13 May 29 14:24 /dev/sda29
brw-rw---- 1 root disk 8, 3 May 29 14:24 /dev/sda3
brw-rw---- 1 root disk 259, 14 May 29 14:24 /dev/sda30
brw-rw---- 1 root disk 259, 15 May 29 14:24 /dev/sda31
brw-rw---- 1 root disk 259, 16 May 29 14:24 /dev/sda32
brw-rw---- 1 root root 259, 17 May 29 14:24 /dev/sda33
brw-rw---- 1 root root 259, 18 May 29 14:24 /dev/sda34
brw-rw---- 1 root root 259, 19 May 29 14:24 /dev/sda35
brw-rw---- 1 root disk 259, 20 May 29 14:24 /dev/sda36
brw-rw---- 1 root disk 259, 21 May 29 14:24 /dev/sda37
brw-rw---- 1 root disk 259, 22 May 29 14:24 /dev/sda38
brw-rw---- 1 root disk 259, 23 May 29 14:24 /dev/sda39
brw-rw---- 1 root disk 8, 4 May 29 14:24 /dev/sda4
brw-rw---- 1 root root 259, 24 May 29 14:24 /dev/sda40
brw-rw---- 1 root root 259, 25 May 29 14:24 /dev/sda41
brw-rw---- 1 root root 259, 26 May 29 14:24 /dev/sda42
brw-rw---- 1 root root 259, 27 May 29 14:24 /dev/sda43
brw-rw---- 1 root root 259, 28 May 29 14:24 /dev/sda44
brw-rw---- 1 root root 259, 29 May 29 14:24 /dev/sda45
brw-rw---- 1 root disk 259, 30 May 29 14:24 /dev/sda46
brw-rw---- 1 root disk 259, 31 May 29 14:24 /dev/sda47
brw-rw---- 1 root disk 259, 32 May 29 14:24 /dev/sda48
brw-rw---- 1 root disk 259, 33 May 29 14:24 /dev/sda49
brw-rw---- 1 root root 8, 5 May 29 14:24 /dev/sda5
brw-rw---- 1 root root 259, 34 May 29 14:24 /dev/sda50
brw-rw---- 1 root root 259, 35 May 29 14:24 /dev/sda51
brw-rw---- 1 root root 259, 36 May 29 14:24 /dev/sda52
brw-rw---- 1 root disk 259, 37 May 29 14:24 /dev/sda53
brw-rw---- 1 root disk 259, 38 May 29 14:24 /dev/sda54
brw-rw---- 1 root disk 259, 39 May 29 14:24 /dev/sda55
brw-rw---- 1 root disk 259, 40 May 29 14:24 /dev/sda56
brw-rw---- 1 root disk 259, 41 May 29 14:24 /dev/sda57
brw-rw---- 1 root disk 259, 42 May 29 14:24 /dev/sda58
brw-rw---- 1 root disk 259, 43 May 29 14:24 /dev/sda59
brw-rw---- 1 root disk 8, 6 May 29 14:24 /dev/sda6
brw-rw---- 1 root disk 259, 44 May 29 14:24 /dev/sda60
brw-rw---- 1 root disk 259, 45 May 29 14:24 /dev/sda61
brw-rw---- 1 root disk 259, 46 May 29 14:24 /dev/sda62
brw-rw---- 1 root disk 259, 47 May 29 14:24 /dev/sda63
brw-rw---- 1 root disk 259, 48 May 29 14:24 /dev/sda64
brw-rw---- 1 root disk 259, 49 May 29 14:24 /dev/sda65
brw-rw---- 1 root disk 259, 50 May 29 14:24 /dev/sda66
brw-rw---- 1 root disk 259, 51 May 29 14:24 /dev/sda67
brw-rw---- 1 root root 259, 52 May 29 14:24 /dev/sda68
brw-rw---- 1 root root 259, 53 May 29 14:24 /dev/sda69
brw-rw---- 1 root disk 8, 7 May 29 14:24 /dev/sda7
brw-rw---- 1 root root 259, 54 May 29 14:24 /dev/sda70
brw-rw---- 1 root disk 259, 55 May 29 14:24 /dev/sda71
brw-rw---- 1 root disk 259, 56 May 29 14:24 /dev/sda72
brw-rw---- 1 root disk 259, 57 May 29 14:24 /dev/sda73
brw-rw---- 1 root disk 259, 58 May 29 14:24 /dev/sda74
brw-rw---- 1 root disk 8, 8 May 29 14:24 /dev/sda8
brw-rw---- 1 root disk 8, 9 May 29 14:24 /dev/sda9
brw-rw---- 1 root disk 8, 16 May 29 14:24 /dev/sdb
brw-rw---- 1 root disk 8, 17 May 29 14:24 /dev/sdb1
brw-rw---- 1 root disk 8, 18 May 29 14:24 /dev/sdb2
brw-rw---- 1 root disk 8, 32 May 29 14:24 /dev/sdc
brw-rw---- 1 root disk 8, 33 May 29 14:24 /dev/sdc1
brw-rw---- 1 root disk 8, 34 May 29 14:24 /dev/sdc2

@Olf0
Copy link
Owner Author

Olf0 commented May 29, 2021

Your command line output shows that the mapping of partition on internal eMMC on this community port of SailfishOS is quite different to all officially supported devices.

@Olf0
Copy link
Owner Author

Olf0 commented May 29, 2021

[OT] Side note & off topic

Also, one thing I was interested in installing it for is the info that says it will enable sd card access on Android, but not working.

Supposedly you are addressing the statement "Ensure, that AlienDalvik begins starting after mounting succeeded, to allow for android_storage on SD-card." (from the third bullet point at https://openrepos.net/content/olf/mount-sdcard).
If so, you heavily misread it ("to allow for" points to a prerequisite) and did not follow its link or did not read the content there (including the comments).
If not, please point me to the sentence, which created this impression for you.

TL;DR: Currently that does not work with AlienDalvik 8, 9 or 10, because Jolla changed something (again), and nobody has traced what and how to workaround it, yet.

@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

Back on topic, again

Was your output generated with or without an SD-card inserted?

Well, what I really want to know is: To which devices names does (partitions on) an SD-card become assigned on your XperiaXZ2c@SFOS4.0.1?
You might look at the difference of above output with and without an SD-card inserted and report that "delta" here.

@rinigus
Copy link

rinigus commented May 30, 2021

On Sony Tama (XZ2{c}/3), system partitions are on /dev/sd{a,b} and maybe few more. SD card is mounted on mmcblk0. Same, ass far as I know, goes for Pro1 and OnePlus 5 or 6 (or both?). Issue has been also exposed in systems settings and was "fixed" there

We have been asking few times for getting some proper fix using udev, but nothing happened so far.

@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

@rinigus, O.K., I think I can handle that ("system partitions are on /dev/sd{a,b}"), but "and maybe few more" might be tricky.
(Side note: Specifically on @levone1's device /dev/sdc was also there.)

But "SD card is mounted on mmcblk0" collides with the system partitions on all officially supported devices, which are all at some mmcblk0p*!
The crucial point here is the inability to distinguish a internal, fixed eMMC from an SD- or MMC-card inserted into some internal slot, because they all are attached via an MMC/SD-controller (as a udev SUBSYSTEMS=).

And which device names does USB-attached storage become assigned to?
If the answer is "on some /dev/sd?, after the last system partitions [...] on /dev/sd{a,b} and maybe few more", this scheme is very awkward and hard to utilise (but seems feasible here, in contrast to mmcblk0, because one can test for SUBSYSTEMS="usb").

@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

After thinking about it a little, I believe this device naming scheme has the potential to cause big trouble (up to bricking devices non-manually; mount-sdcard currently just enables one to do that manually), because it more or less conflicts with what Jolla chose for their device adaptations.
Insofar I am glad that @levone1 discovered this.

Hence I will consider to limit the /dev/sd[a-z]* mount-sdcard and crypto-sdcard care about to those with SUBSYSTEMS="usb", but I think the proper fix is to move those community device adaptations to Jolla's device naming scheme.

@rinigus
Copy link

rinigus commented May 30, 2021

The naming scheme for devices comes from Android layer/SOC and cannot be changed. The devices in question are using SD 835, 845 and I suspect that the later SDs as well (don't know though). So, in my eyes, real solution would be to instruct UDEV to attach some label to removable devices and use that label to handle mounting.

I agree, it is a dangerous situation and users should be aware of it.

@rinigus
Copy link

rinigus commented May 30, 2021

PS: I have not checked specific devices as I didn't have time for it. But, in principle, we should not look for solution where each app or subsystem has to write down all possible configs. Hence I wasn't documenting it here.

@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

The naming scheme for devices comes from Android layer/SOC and cannot be changed.

That is unfortunate, tough I was expecting this to be reason.

So, in my eyes, real solution would be to instruct UDEV to attach some label to removable devices and use that label to handle mounting.

Oh, that would be the perfect solution. But only Jolla and the udev project are in the position to change that for SailfishOS (or all distros using udev) in general.
Side note: Likely the path via the udev project is the quicker one. Maybe @levone1 has the energy to draft a suggestion for the udev project (e.g., starting as a Git{hub|lab} "Gist" or a separate, new TMO thread; actually a new FSO thread for this would be automatically bring this on Jolla's radar), and after some some discussion and potentially enhancing the draft, this suggestion can be filed as an "issue" at the udev project.

But, in principle, we should not look for solution where each app or subsystem has to write down all possible configs.

I fully agree, but what else can be done to fix this clash between mount-sdcard and some community adaptations soon?
Which implies "on our own behalf" (your, mine or together; i.e., without having to involve others).

Olf0 added a commit that referenced this issue May 30, 2021
@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

O.K., I drafted a workaround per commit 89bc3d1.
I will try to release a new version within the next couple of days, which will be named mount-sdcard 1.8.0
@levone1, I would be glad, if you test it then, for that I will explicitly notify you of the Github release, here (ususally one to two days before I release at Openrepos).

But as I wrote

But "SD card is mounted on mmcblk0" collides with the system partitions on all officially supported devices, which are all at some mmcblk0p*!
The crucial point here is the inability to distinguish a internal, fixed eMMC from an SD- or MMC-card inserted into some internal slot, because they all are attached via an MMC/SD-controller

@rinigus, is it possible to assign the sd-card slot to mmcblk1, or to put it more generically:
Can you block mmcblk0 from being used for the device enumeration on your SFOS port(s)?
Side note: Mind that SD- and MMC-cards in USB-attached card readers also become assigned to mmcblk* names.

@rinigus
Copy link

rinigus commented May 30, 2021

I am massively swamped right now and will probably be next week as well. After that I can have a meaningful discussion on it. Sorry for it.

Don't count on us to be able to block mmcblk0 from appearing. However, maybe we could tag it somehow as removable device. I will have to pull my notes regarding it, hopefully I wrote it down in the corresponding issue of Tama port. But that will have to wait.

@Olf0
Copy link
Owner Author

Olf0 commented May 30, 2021

Sorry for it.

No, don't! ;-)
That is and always has to be O.K., for everybody spending his / her spare time on this.
Just keep in mind that the few handful of Jolla employees in technical jobs have no problem to delay various things indefinitely, to leave unfinished task in the state they are forever etc. etc. without ever saying a word: They are getting paid for that work, still you will not hear an excuse, or even an apology from them.

TL;DR: Ping us here, when you have some time to deal with this.

Side note:
If "we" (i.e., you for your ports) are able to "tag" (should not be a real udev tag, rather just a udev variable) all mmcblk0* as removable, I can (and am willing to) evaluate that in mount-sdcard and crypto-sdcard, but then we are back at the very individual solution you initially wanted to avoid (IMO, correctly so).
This is why I considered and suggested to check, if it is possible to force the device enumeration for SD- and MMC-cards & -ICs to start at mmcblk1 (not 0), because that would be compatible to the naming scheme the officially supported devices use (i.e., mmcblk0* implicitly and always is on internal, fixed mass storage.)
OTOH, I do agree that Jolla should address this, because they will have to sooner or later anyway (as these are device naming conventions of the Linux kernel depend on which driver / subsystem is used; maybe this is one of the reasons, why they always preferred Sonys midrange devices with Qualcomm SDM6xx SoCs). But I do not want to wait "Jolla time", often just to receive "Jolla quality" after waiting for long.

@Olf0
Copy link
Owner Author

Olf0 commented May 31, 2021

Please test mount-sdcard-1.8.0-1.sfos340.noarch.rpm on XperiaXZ[23]{c}, Pro¹ or another affected community device adaptation with SFOS4.x.y, @levone1.

  • It should now only take care of /dev/sd*, which are attached via USB.
  • As long as there is no proper way to determine if /dev/mmcblk0* is "removable", mount-sdcard will not take care of these devices (assuming they are on a fixed, internal eMMC).

Does it now work as expected / indented (by me)?


Suggested procedures (either one):

  • Start at the command line as regular user, preferably in /tmp
    curl -LO https://github.com/Olf0/mount-sdcard/releases/download/1.8.0-1.sfos340/mount-sdcard-1.8.0-1.sfos340.noarch.rpm
    devel-su
    pkcon install-local mount-sdcard-1.8.0-1.sfos340.noarch.rpm
    
  • Download the *.noarch.rpm in your web-browser from https://github.com/Olf0/mount-sdcard/releases/tag/1.8.0-1.sfos340
    Open the native Filemanager app, go to the download directory, select the file mount-sdcard-1.8.0-1.sfos340.noarch.rpm and install it per top pulley entry. Note that this action needs "Untrusted sources" (at least temporarily) enabled in the SailfishOS' Settings app.

@Olf0 Olf0 closed this as completed May 31, 2021
@Olf0 Olf0 reopened this May 31, 2021
@Olf0 Olf0 self-assigned this May 31, 2021
@Olf0 Olf0 added enhancement New feature or request help wanted Extra attention is needed question Further information is requested labels May 31, 2021
@Olf0
Copy link
Owner Author

Olf0 commented May 31, 2021

[OT] continued

@levone1, as Jolla has been fixing stuff lately and had explicitly broken SD-card access for the AlienDalvik runtime environment in SailfishOS 3.3.0 - 3.4.0, you may just retry the suggestions in this sub-thread (please read it twice in its completeness, which is just a few messages, before starting to test / try things): https://together.jolla.com/question/203539/guide-externalising-android_storage-and-other-directories-files-to-sd-card/?answer=206350#post-id-206350

Please report you success or lack thereof somewhere and point me to that.

@Olf0
Copy link
Owner Author

Olf0 commented Jun 1, 2021

@levone1, as you were so keen on having this quickly addressed and fixed:
Can you please test and report back here!

P.S.: I would like to release v1.8.0 at Openrepos today, but preferably after knowing that this fixed your issue.

@Olf0
Copy link
Owner Author

Olf0 commented Jun 1, 2021

Closing, as there is no feedback by the original reporter and because mount-sdcard 1.8.0 is likely working well on these niche devices.
A bit unfortunate to receive no feedback at all after all this upheaval in many places (FSO, TMO, OpenRepos and here), but this did not come completely unexpected.

@rinigus, you may utilise this (closed) thread (or by some other means) to notify me, when you have found a way to somehow mark a "removable" mmcblk0* as such.

@Olf0 Olf0 closed this as completed Jun 1, 2021
@Olf0 Olf0 added enhancement New feature or request and removed enhancement New feature or request help wanted Extra attention is needed question Further information is requested labels Jun 2, 2021
@levone1
Copy link

levone1 commented Jun 2, 2021

Update installed and working fine - no additional "removable media" devices. Thanks

p.s. Sorry you feel so put-out by it. It really wasn't a big problem for me; I thought I was being helpful to you.

@Olf0
Copy link
Owner Author

Olf0 commented Jun 2, 2021

@levone1, thanks for your feedback.

As you asked, two things were problematic for me:

  • [Technical] If it comes to my knowledge that a tool I provide enables users to wreck / "brick" their devices, I feel some responsibility to fully comprehend the issue and (if necessary, feasible, able to) do something about it.
    Despite the fact, that all my tools are FLOSS licensed (i.e., I have no liability at all).
  • [Procedural] If you would have filed an issue report here (as I suggested early in our extensively distributed communication), let me (or in general: the maintainer) handle your report, simply answer the question(s) and ultimately test a new release, this would have been far less tedious for me.
    It is O.K., that you might not see a couple of non-technical aspects your "spreading the news of something going pretty wrong in many places" has, but that is definitely not helpful!
    There is usually a single best place for reporting bugs: The primary issue tracker of the original project (and rather not general forums as e.g. FSO and TMO).

@Olf0
Copy link
Owner Author

Olf0 commented Jun 2, 2021

P.S.: And as stated, @levone1, if you like to try out things / contribute to a long lost open question / serve your original interest of having android_storage on SD-card, take a look at the two "Off Topic (OT)" postings above.

@levone1
Copy link

levone1 commented Jun 3, 2021

Externalizing seems interesting ... I'll probably mess with it at some point to test. My current sd card is nearly full, but I'll pull one out to work with. Are yo saying that as of now, there is no possible way for AD to access the card directly? My interest was more in that realm, but workarounds work...

@Olf0
Copy link
Owner Author

Olf0 commented Jun 3, 2021

???
Please try to precisely describe what you mean, especially because we all are (presumably) non-native English speakers and writers:

  1. Externalizing seems interesting ...

    Does that address "externalising android_storage to SD-card"?

  2. Are yo saying that as of now, there is no possible way for AD to access the card directly?

    I don't know, because I fail to comprehend, what you mean with "directly".
    And also what "as of now" is supposed to mean: "with SFOS 4.1.0"?

    I wrote "Jolla [...] had explicitly broken SD-card access for the AlienDalvik runtime environment in SailfishOS 3.3.0 - 3.4.0, [...]", hence "externalising android_storage to SD-card" cannot work on SailfishOS 3.3.0 - 3.4.0. It seems that everybody has given up with this back then, so nobody ever tried again (or did not report their result, which is basically equivalent).
    This is all I know, any further research (i.e. trying out and analysing things, reading documentation, etc.: in short, "researching") is up to you.

P.S.: It was reported at FSO that simply accessing an SD-card from Android apps works again with SFOS ≥ 4.0.1. I have not tried that yet, because my "production phone" is still on SFOS 3.2.1 due to the massive amount of bugs introduced in each following release.

@levone1
Copy link

levone1 commented Jun 4, 2021

My initial fiddling ...
I went through section 2 (Externalising /home/nemo/android_storage) of your guide, and seemed to be working properly, (replacing 'nemo' with 'defaultuser'), up to section 2.2, (Integrating an externalised android_storage when booting), where I dead-ended, since there is no longer such a thing as /opt/alien/system/script, (the only thing in .../alien is system.img). I did unsquash syste img to poke around, thinking maybe there was some correspondent script in there somewhere, but didn't find anything.

Then I tested the guide by mynameisnotimportant posted under yours, and I found a significant difference in my config file compared to his. I tried the tweaks he posted anyway, just to see, but it caused AD to not start, (tried a few different paths/syntaxes).

Here is the content of my AD config file:
`lxc.rootfs.path = loop:/opt/alien/system.img
lxc.uts.name = aliendalvik
lxc.arch = aarch64

lxc.init.cmd = /init

lxc.mount.auto = cgroup:ro sys:ro

necessary dev nodes (renamed)

lxc.mount.entry = /dev/puddlejumper dev/binder none bind,create=file 0 0
lxc.mount.entry = /dev/vndpuddlejumper dev/vndbinder none bind,create=file 0 0

necessary device nodes (same name)

lxc.mount.entry = /dev/ashmem dev/ashmem none bind,create=file 0 0

lxc.mount.entry = /dev/fuse dev/fuse none bind,create=file 0 0
lxc.mount.entry = /dev/ion dev/ion none bind,create=file,optional 0 0

Mount /data - always /home/.android now

lxc.mount.entry = tmpfs mnt tmpfs mode=0755,uid=0,gid=1000

useful device properties from the host

lxc.mount.entry = /tmp/aliendalvik/generated_props system/vendor/alien.prop none bind,create=file 0 0

Xkb - required for on screen keyboard

lxc.mount.entry = /usr/share/X11/xkb usr/share/X11/xkb none bind 0 0

lxc.include = /var/lib/lxc/aliendalvik/bsp_config
lxc.include = /var/lib/lxc/aliendalvik/extra_config

`
It's true that 'accessing' sd card from Android works, but read-only, (at least for me)...

@Olf0
Copy link
Owner Author

Olf0 commented Jun 6, 2021

My initial fiddling ...

... is still off-topic to this issue thread: Please open a new one.
Please do choose the title and issue description well, so that any new reader is able to comprehend it easily. I.e., provide context or well labelled links to the necessary context in the issue description (you might reuse and transform the two messages I posted here labelled [off-topic / OT] to an issue description).

I went through section 2 (Externalising /home/nemo/android_storage) of your guide, and seemed to be working properly, (replacing 'nemo' with 'defaultuser'), up to section 2.2, (Integrating an externalised android_storage when booting), where I dead-ended, since there is no longer such a thing as /opt/alien/system/script, (the only thing in .../alien is system.img).

Well, my advice is still to wholly read threads, before starting to experiment / trying to reproduce.
As clearly stated, this guide was created for 1st and 2nd generation AlienDalvik (the ASOP 4 based ones), only.

I did unsquash syste img to poke around, thinking maybe there was some correspondent script in there somewhere, but didn't find anything.

Structurally I expected that outcome, but thanks for checking.

Then I tested the guide by mynameisnotimportant posted under yours,

... where I pointed you to, twice.

and I found a significant difference in my config file compared to his.

Which ones?

I tried the tweaks he posted anyway, just to see, but it caused AD to not start, (tried a few different paths/syntaxes).

Here is the content of my AD config file:

The original or patched content?

Also, please upload larger (> ~300 Bytes) files as a file-attachment to a post.
If you post file-content inline, protect it from being interpreted as markdown by quoting (i.e., framed in ` for a single line, and ``` for a block of lines).

It's true that 'accessing' sd card from Android works, but read-only, (at least for me)...

Thanks for this result.
Yes, that makes sense to me, when I look at Jolla's AlienDalvik-LXC configuration.
Which is unfortunate, because it once was read-write.
And this has to be resolved first, before trying to put android_storage onto SD-card.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants