As the title says, the /data partition keeps getting corrupted on my SGH-i317 (AT&T model). It manifests itself as random lockups and reboots, eventually ending up in a no-boot condition. At this point I boot into Philz recovery and use an adb shell to run e2fsck, which fixes a bunch of things and then I can boot normally. This has happened three times in the last month. The first time it happened I was running CleanROM JB, and at that point I upgraded to dicksteele's DS-NE5v1, so I don't think the ROM is to blame. Any idea what could be doing this and/or how to prevent it? Could it be that the FLASH ram is failing? Is there a way to force an fsck at every boot?
Perhaps I spoke too soon...the phone locked up on me last night while we were both asleep (i.e. I woke up, the phone didn't, so I had to yank the battery). The first thing I did was an e2fsck, and it was clean, which leads me to think that the lockups are causing the corruption instead of the other way around.
A little more history: when I was on CleanROM there were no random lockups or reboots, it was a particular app that caused them when I would run it. I tried to uninstall that app, which caused more lockups/reboots, then becoming reboots just opening the Application Manager or trying to uninstall it with Titanium Backup, eventually ending up in a no-boot scenario. After a couple day's worth of pulling my hair out trying to install various recoveries and flashing new ROMs, I determined that it was a corrupted /data partition, found an article on how to e2fsck it and ended up on DS-NE5v1 (the last ROM I tried in the previous step). At that point I had random lockups/reboots until it eventually went no-boot again after a couple weeks. I've since gone through the same cycle twice, having to run an e2fsck to get it booting again.
Any ideas how to diagnose the random lockups?
I am no longer amused: the thing locked up on me again last night, but rebooted just fine after I yanked the battery. Just now it rebooted while sitting untouched on my desk, then rebooted again and locked up at the animated Samsung logo. The e2fsck said this:
Code:
e2fsck -fy /dev/block/mmcblk0p16
e2fsck 1.41.14 (22-Dec-2010)
Pass 1: Checking inodes, blocks, and sizes
Deleted inode 377887 has zero dtime. Fix? yes
Inodes that were part of a corrupted orphan linked list found. Fix? yes
Inode 378413 was part of the orphaned inode list. FIXED.
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(1433114--1433116) -1509879
Fix? yes
Free blocks count wrong for group #43 (22311, counted=22314).
Fix? yes
Free blocks count wrong for group #46 (11952, counted=11953).
Fix? yes
Free blocks count wrong (1075855, counted=1075859).
Fix? yes
Inode bitmap differences: -377887 -378413
Fix? yes
Free inodes count wrong for group #46 (5787, counted=5789).
Fix? yes
Free inodes count wrong (712361, counted=712363).
Fix? yes
/dev/block/mmcblk0p16: ***** FILE SYSTEM WAS MODIFIED *****
/dev/block/mmcblk0p16: 24917/737280 files (3.6% non-contiguous), 1871209/2947068
blocks
Has anyone seen this behaviour before in DS-NE5v1 or any other ROM?
Mostlyhrmlss said:
I am no longer amused: the thing locked up on me again last night, but rebooted just fine after I yanked the battery. Just now it rebooted while sitting untouched on my desk, then rebooted again and locked up at the animated Samsung logo. The e2fsck said this:
Code:
e2fsck -fy /dev/block/mmcblk0p16
e2fsck 1.41.14 (22-Dec-2010)
Pass 1: Checking inodes, blocks, and sizes
Deleted inode 377887 has zero dtime. Fix? yes
Inodes that were part of a corrupted orphan linked list found. Fix? yes
Inode 378413 was part of the orphaned inode list. FIXED.
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(1433114--1433116) -1509879
Fix? yes
Free blocks count wrong for group #43 (22311, counted=22314).
Fix? yes
Free blocks count wrong for group #46 (11952, counted=11953).
Fix? yes
Free blocks count wrong (1075855, counted=1075859).
Fix? yes
Inode bitmap differences: -377887 -378413
Fix? yes
Free inodes count wrong for group #46 (5787, counted=5789).
Fix? yes
Free inodes count wrong (712361, counted=712363).
Fix? yes
/dev/block/mmcblk0p16: ***** FILE SYSTEM WAS MODIFIED *****
/dev/block/mmcblk0p16: 24917/737280 files (3.6% non-contiguous), 1871209/2947068
blocks
Has anyone seen this behaviour before in DS-NE5v1 or any other ROM?
Click to expand...
Click to collapse
More than likely, there is a way to run the command as a script at boot. You can run just about anything as a script at boot using script manager & terminal emulator.
However, have you checked your chip? Some of them had SDS (Sudden Death Syndrome, aka Brick Bug) as I'm sure you're aware. There's an app called eMMC Brickbug Check you can use to check your chip.
Also, when was the last time you formatted data in recovery? Are you getting any errors like UIDs inconsistent or anything?
I don't know much about the e2fsck, but I'm sure if I were to run it right now on my perfectly running phone, it would return similar results. Honestly, IMO, random reboots, crashes, and boot failure sounds like a kernel problem to me. I don't know if you've tried yet, but it might be worth it to try a different kernel.
Without exception, the e2fsck comes back clean when the thing will boot, comes back dirty when it won't, and always boots afterwards. The check needs to be done with the partition unmounted. Since any script would be run from the /system tree, the /data partition would have to be mounted in order to read the script in the first place. Reminds me of something about a chicken and an egg
I formatted the partition when I installed DS-NE5v1, a couple months ago, right before the random problems started.
The eMMC Brickbug came back clean with a sane chip. I did notice that it reported the model as t0lteuc, it is a t0lteatt. Is that anything to be worried about?
The kernel is AGNi 3.0.101-India-v4.2.3_R2-OC-LTE-KitKat. Is there something better?
Mostlyhrmlss said:
Without exception, the e2fsck comes back clean when the thing will boot, comes back dirty when it won't, and always boots afterwards. The check needs to be done with the partition unmounted. Since any script would be run from the /system tree, the /data partition would have to be mounted in order to read the script in the first place. Reminds me of something about a chicken and an egg
I formatted the partition when I installed DS-NE5v1, a couple months ago, right before the random problems started.
The eMMC Brickbug came back clean with a sane chip. I did notice that it reported the model as t0lteuc, it is a t0lteatt. Is that anything to be worried about?
The kernel is AGNi 3.0.101-India-v4.2.3_R2-OC-LTE-KitKat. Is there something better?
Click to expand...
Click to collapse
Well as for agni, 4.23 is kinda old. I know he's since put out 4.24 and 4.25 for our note 2. A new onewith multiroom support was released today so I'll be testing that. To me the most stable versions are 4.24 and 4.25. But if you flash a new kernel be sure to wipe your init.d folder or you will have issues. I do it even when updating the same kernel just to be sure.
As for the build prop I'm not sure what tolteuc is, that is clearly a different country code... but i dont think it's a problem. International would be toltexx and at&t is tolteatt. But that's a rom thing, like mine currently has an s5 build prop so mine says klteatt and kltexx. If you're concerned about it, you can dig into your build prop and the files in your CSC folder. Sometimes these have conflicting information in them which can be a problem (but usually isn't)... sometimes people need to get correct CSC for their country or device and if they're having problems it can help to set the build prop to your original defaults (after making a backup, of course).
Running that command fixes whatever problems your having temporarily. Ok well that's good but you don't want to be running it all the time either. The best way in this case to test your memory & chip without replacing it or running a command constantly would be to Odin back to stock for a little while to see if it's still a problem.
Imo, you can start seeing problems on one rom and despite wiping & wiping, it can still carry over to any subsequent roms you flash. When you see a certain problem consistently, you need to take it back to stock because that's the only way to reset everything back to a clean slate... and if you still have the problem there & then you know it's probably hardware. However, I think it'll resolve whatever is causing these issues and you will be free to flash whatever you want after that. It's worth a try, better than trying to run a script at boot every boot.
This is all just my opinion and what I would do. I just saw that no one had helped you yet and that shouldn't happen here when you have a legitimate concern and you post in the correct section as you did.
I put in AGNi 4.2.8 today, and it crashed within an hour. I see a stock flash in my future...
BTW, perusing the AGNi Control app I found an option for FileSystems Checking on Boot, which is off by default. I don't suppose it could hurt, am going to try it.
Oh, and I really do appreciate the help...was beginning to think I was talking to myself
Mostlyhrmlss said:
I put in AGNi 4.2.8 today, and it crashed within an hour. I see a stock flash in my future...
BTW, perusing the AGNi Control app I found an option for FileSystems Checking on Boot, which is off by default. I don't suppose it could hurt, am going to try it.
Oh, and I really do appreciate the help...was beginning to think I was talking to myself
Click to expand...
Click to collapse
Lol I was surprised no one had helped yet! Yeah try the check file system setting, if you still have issues, try a stock flash. It's always helped me in the past. Too bad the new agni didn't solve it on its own... but hey maybe that check setting will help. I'm sure you'll get it sorted!
I did the stock flash and everything seems good, no crashes for a couple of days now. Thanks for your help!
Mostlyhrmlss said:
I did the stock flash and everything seems good, no crashes for a couple of days now. Thanks for your help!
Click to expand...
Click to collapse
That's excellent news! Glad to hear it!
Related
This is the thread for us A100 owners to have a general Q&A, chatter, thanks, non specific bug reports, or anything else you feel like posting related to the A100 CM10 builds by waydownsouth.
You can find his preview build thread here
Use this gapps for now.
Please keep his Dev thread clean of the usual thanks, you rock, take my wife please, shut up and take my money pictures, etc, and post them here. Please keep the Dev thread clean and Dev related only.
Specific backed up suggestions for fixes, logs, and other development related posting can be directed to his release thread in the dev forum.
Some general information
Bugs
The first thing you may notice, the boot animation will blank out, just a black screen for a bit halfway through the CM9 animation. Don't freak out, it isn't wrecked. It only occurs on boot animations that have 2 stages, it blanks when the loop kicks in. If this bothers you, replace the boot animation with either one that is always a loop, or doesn't have a second stage, so animation, then stops.
The stock launcher is always landscape, I'm not sure why this is, the rest of the system will work either in portrait or landscape. 2 ways around this, first is use something like Apex launcher or Nova launcher (or any launcher really) OR force Tablet UI, see the FAQ for information on this. This isn't really a bug, its just confusion caused by Google adding a 3rd UI, and our screens mess it up.
There is two gallery apps and two camera apps, one gallery is normal, the other is the 3D version, I'm not clear on why there is 2 camera apps, it may be from the GApps. It'll eventually calm down, and if it annoys you, you can uninstall through TiBu or whatever app you choose, or manually renaming, removing or deleting the apk in /system/app.
FAQ
Yup, I already have some questions about this, so here's some answered here.
Q: Should I do a full wipe before flashing this?
A: It's always recommended to do a full wipe moving from one base to another, or upgrading from one version to the next. IE, CM9 to CM10. You can always feel free to flash anyways and see how it goes, but if you have problems, you'll know why.
Q: What can I restore?
A: As usual, you can restore whatever apps you wish, however I suggest not restoring System apps (red in TiBu) or their data. User apps and data (green in TiBu) are generally safe. Restore in steps, a few at a time, and check for any issues, reboot, then try some more. This is the careful route, of course. Issues from User apps and data are pretty rare unless the app is outdated and doesn't work with Jellybean.
Q: What works and what doesn't?
A: Please view the Dev thread for this information, it will be more current and up to date. The link is in the first post of this thread.
Q: I hate this new 7" Jellybean UI, how do I get the combined bar back ("Tablet UI")?
A: Setting density to 130 or lower will force the normal Tablet UI back to ICS style. This of course comes at the cost of some apps not working, play store issues, and everything being smaller on the screen. Up to you if it's worth it in the end.
Q: Why is there still the CM9 Boot Animation? Isn't this CM10?
A: Yes, it's CM10, however CM hasn't included a new boot animation for it, even CM9 went for awhile before one was introduced, and it was later changed to the current version. It'll happen in time as ideas are tossed about and the new animation is voted in, then gets merged for the appropriate device types and resolutions.
Q: If this is just a preview, when will we see an Alpha, Beta, or RC?
A: Don't ask. waydownsouth is only 1 man, and he can only do so much on his own. As the device tree gets cleaned up and the rest of the Iconia-Dev guys can get to cracking, it'll move along quickly. Until then, sit tight, and don't ask for an ETA. Or, just take this as your ETA "It'll be done when it's done."
Q: I use the Crossix Mount Swap script and now everything is broken, what gives?
A: The Crossix Mount Swap was intended for HC then ICS, it hasn't been updated for JB yet. Please stop using it. If you use it, please stop reporting storage issues as bugs, its not the ROM, its the script. If you choose to use it, consider your CM10 warranty void.
Q: Does the EzT20 Kernel work in CM10?
A: Reports say yes, it does, however WDS changed the audio hal from nVidia's to something far better (notice CM9/CM10 doesn't pop like stock based ROMs? Thank him for that) and the EzT20 kernel still uses the nVidia audio system. While it appears to not cause any outright issues, I would not suggest using it. If you choose to use it (your choice of course, have fun and experiment) please don't report any bugs AT ALL. Consider your CM10 warranty void.
Q: WTF did my overclock go?
A: Overclocking was removed in the later builds (preview builds) for stability. Bug reports coming in due to overclocking and running a destabilized ROM are hard to track, and are generally not a real bug, just system instability due to overclocking or undervolting. If you just HAVE to have your overclocking, see the Q above.
Q: Will this work on my A101?
A: It will if you don't mind losing 3G. The modem hasn't been integrated yet by waydownsouth, and while I'm sure he'll get it going, he hasn't yet. If you want to play around with CM10, just make a nandroid backup, wipe, flash CM10, play with it, then go back to recovery, full wipe, restore the backup, then resume normal usage with 3G.
Q: Where is the device tree?
A: Right now WDS is keeping it private, no one has access to it, and he'll make it public when he feels it's ready to be available to everyone, please be patient and respect his wishes on this.
Q: Does linux society's kernel work with this?
A: Yes, it does, however if you have issues please don't report in the CM10 threads, but his kernel thread, if you wish to report an issue please use the stock kernel first.
I'll keep adding things as I come across them in this or other threads (even other forums if I happen across them) and PMs I receive.
Last updated FAQ 8.17.12
Flashing now...
Awesome! Flashing now...
:laugh:
Boot Animation is still CM9 (no biggie)
When the tab is unlocked it goes to Landscape mode but when you go into settings for example it turns to portrait mode but when you exit settings it goes to landscape again. Is there a setting for this that I am missing? (Yes, rotation lock is off.) Apps list is also stuck in landscape mode...
All for the moment... looks AWESOME!
Originally I did NOT restore Data, gonna live dangerously and try it to just see how it reacts. Will probably regret it but thats the fun in this. Will report back later today (Tuesday) time for bed here on the coast of Maine...
Just as mentioned above by pio_masaki, dont restore data... it will not like it. LoL Just had to try it though... now to wipe and flash... maybe I'll sleep tonight... maybe...
aszelinski said:
Awesome! Flashing now...
:laugh:
Boot Animation is still CM9 (no biggie)
When the tab is unlocked it goes to Landscape mode but when you go into settings for example it turns to portrait mode but when you exit settings it goes to landscape again. Is there a setting for this that I am missing? (Yes, rotation lock is off.) Apps list is also stuck in landscape mode...
All for the moment... looks AWESOME!
Click to expand...
Click to collapse
A lot of the weird things in Jellytime will appear here as well, namely the stock launcher always being landscape. In the pulldown there is a rotation lock button, if you choose. A lot of the odd behavior is also just Jellybean vs ICS.
The recent apps list in landscape is swipe up to remove, and in portrait swipe left or right.
2nd Go taking a more subtle approach...
2nd Go taking a more subtle approach...
Recovery - TWRP 2.2.0
Installed - pio_masaki Blackhole_System_Wipe_1.5a_A100.zip
cm-10-20120723-Unofficial-a100.zip (THANK YOU DEVS!!!)
gapps-jb-20120717-signed.zip (THANK YOU DEVS!!!)
Rebooted into system (blank screen for a few moments after boot animation stopped,known issue) let tab sit for 10 minutes. Signed in.
Loaded apps only from TiBu. Nothing screamed
GPS working: GPS Test reports 10 sats in view / 10 in use. / took maybe 15 seconds
GPS Status reports 10 sats in view / fix on 8-9 / took maybe 15 seconds
Cameras works : Front and out facing. ICS Camera app is still better that standard camera app.
Gonna play with it more as the day progresses...
Not understanding the flash process?
Greetings.
I must be missing something in the flash process. I downloaded cm-10-20120723-UNOFFICIAL-z100.zip, did
adb push cm-10-20120723-UNOFFICIAL-z100.zip /mnt/sdcard
to get it onto my a100. Ran ROM Manager and installed from the zip file. The tablet says
Code:
ROM Manager Version 5.0.0.8
July 24, 2012
Preparing to install ROM...
about to run program [/cache/dowipedalvikcache.sh] with 1 args
mount: mounting /dev/block/mmcblk0p4 on /cache failed: Device or resource busy
mount: can't find /sd-ext in /etc/fstab
umount: can't umount /sd-ext: Invalid argument
run_program: child exited with status 1
Formatting /cache...
Creating filesystem with parameters:
Size: 1283457024
Block size: 4096
Blocks per group: 32768
Inodes per group: 7840
Inode size: 256
Journal blocks: 4896
Label:
Blocks: 313344
Block groups: 10
Reserved block group size: 79
Created filesystem with 11/78400 inodes and 10305/313344 blocks
Formatting /data...
Skipping erase of /data/media.
Formatting /sd-ext...
-- Installing: /sdcard/cm-10-20120723-UNOFFICIAL-a100.zip
Finding update package...
I:Update location: /sdcard/cm-10-20120723-UNOFFICIAL-a100.zip
Opening update package...
Installing update...
Installation aborted.
result was NULL, message is: assert failed: install_zip("/sdcard/cm-10-20120723-UNOFFICIAL-a100.zip")
Suggestions anyone?
croldham said:
Greetings.
I must be missing something in the flash process. I downloaded cm-10-20120723-UNOFFICIAL-z100.zip, did
adb push cm-10-20120723-UNOFFICIAL-z100.zip /mnt/sdcard
to get it onto my a100. Ran ROM Manager and installed from the zip file. The tablet says
Code:
ROM Manager Version 5.0.0.8
July 24, 2012
Preparing to install ROM...
about to run program [/cache/dowipedalvikcache.sh] with 1 args
mount: mounting /dev/block/mmcblk0p4 on /cache failed: Device or resource busy
mount: can't find /sd-ext in /etc/fstab
umount: can't umount /sd-ext: Invalid argument
run_program: child exited with status 1
Formatting /cache...
Creating filesystem with parameters:
Size: 1283457024
Block size: 4096
Blocks per group: 32768
Inodes per group: 7840
Inode size: 256
Journal blocks: 4896
Label:
Blocks: 313344
Block groups: 10
Reserved block group size: 79
Created filesystem with 11/78400 inodes and 10305/313344 blocks
Formatting /data...
Skipping erase of /data/media.
Formatting /sd-ext...
-- Installing: /sdcard/cm-10-20120723-UNOFFICIAL-a100.zip
Finding update package...
I:Update location: /sdcard/cm-10-20120723-UNOFFICIAL-a100.zip
Opening update package...
Installing update...
Installation aborted.
result was NULL, message is: assert failed: install_zip("/sdcard/cm-10-20120723-UNOFFICIAL-a100.zip")
Suggestions anyone?
Click to expand...
Click to collapse
Which recovery do you run? Usually that assert failed means your recovery is reporting the old way, picasso / a500. The fix is update recovery, the work around is to remove the assert lines from the updater script. Has this occured with with any of the CM9 builds (if you tried any)? If its failing this but not say Flex Reaper, its because that doesn't check which device you are on.
croldham said:
Suggestions anyone?
Click to expand...
Click to collapse
You may want to boot into recovery manually do everything. Usually Rom Manager would give errors like that (even though nothing's wrong).
Installed fine with CMW recovery. Only bugs I've seen so far is the ones listed already.
I tried to use the SD card int/ext switch by copying to /etc but that did not work. No surprise there.
Used script manager to run it with SU and boot and it seems to work. However, I was not able to access anything in the SD card. It boots up fine and the folders are there but not accessible. Anyone have better luck? Did I miss a step somewhere?
pio_masaki said:
Which recovery do you run? Usually that assert failed means your recovery is reporting the old way, picasso / a500. The fix is update recovery, the work around is to remove the assert lines from the updater script. Has this occured with with any of the CM9 builds (if you tried any)? If its failing this but not say Flex Reaper, its because that doesn't check which device you are on.
Click to expand...
Click to collapse
CWM-based Recovery v5.5.0.x (thor2002ro rev1.7)
I haven't tried any CM9 based builds, I was going right from stock 4.0.3 to Jelly Bean.
You may want to boot into recovery manually do everything. Usually Rom Manager would give errors like that (even though nothing's wrong).
Click to expand...
Click to collapse
Tried this too, booted into recovery and selected 'install from zip file'. I get finding...opening...Installing update...Installation aborted. Is there a way to do it from adb?
croldham said:
CWM-based Recovery v5.5.0.x (thor2002ro rev1.7)
I haven't tried any CM9 based builds, I was going right from stock 4.0.3 to Jelly Bean.
Tried this too, booted into recovery and selected 'install from zip file'. I get finding...opening...Installing update...Installation aborted. Is there a way to do it from adb?
Click to expand...
Click to collapse
Have you tried downloading again to make sure its not a bad download?
croldham said:
CWM-based Recovery v5.5.0.x (thor2002ro rev1.7)
I haven't tried any CM9 based builds, I was going right from stock 4.0.3 to Jelly Bean.
Tried this too, booted into recovery and selected 'install from zip file'. I get finding...opening...Installing update...Installation aborted. Is there a way to do it from adb?
Click to expand...
Click to collapse
It's your recovery, thats the old recovery, updating with the new CWM or TWRP will correct this error for any zip you will try to install. Mine all have that check on them, and do all CM builds.
Here is TWRP, my personal favorite
Here is CWM, also a very solid recovery
Either of those two will offer a fix for this issue on any ROM you flash in the future.
Theonew said:
You may want to boot into recovery manually do everything. Usually Rom Manager would give errors like that (even though nothing's wrong).
Click to expand...
Click to collapse
pio_masaki said:
It's your recovery, thats the old recovery, updating with the new CWM or TWRP will correct this error for any zip you will try to install. Mine all have that check on them, and do all CM builds.
Here is TWRP, my personal favorite
Here is CWM, also a very solid recovery
Either of those two will offer a fix for this issue on any ROM you flash in the future.
Click to expand...
Click to collapse
pio_masaki: You are right, I flashed to TWRP and was able to successfully flash Jelly Bean. I had to turn off archive integrity checking to do it, but the flash went fine and I'm now "munching them beans."
I have a bunch of desk work to do today so I'm going to be watching Netflix as a test.
here are my first impressions:
-device is stuck at landscape during home screens and app drawer as previously reported.
-HDMI audio is definitely not working for me. Tested with MX Video.
Will edit and add more issues as I find them.
On a side note, is Chrome being problematic for everyone else since the last update? (ICS too) Sometimes it won't render websites or freezes itself, all it takes is a swipe on the task manager and open it again to get it to work, but it still seems unreliable and inconvenient.
littleemp said:
here are my first impressions:
-device is stuck at landscape during home screens and app drawer as previously reported.
-HDMI audio is definitely not working for me. Tested with MX Video.
Will edit and add more issues as I find them.
On a side note, is Chrome being problematic for everyone else since the last update? (ICS too) Sometimes it won't render websites or freezes itself, all it takes is a swipe on the task manager and open it again to get it to work, but it still seems unreliable and inconvenient.
Click to expand...
Click to collapse
Chrome has been problematic for awhile, I think we figured out the version that they started in either the aoikaze thread or the cm9 nightly thread. Also I think the tegra drivers has a memory leak issue that causes issues with chrome.
Thank you for reporting the HDMI audio it was something I could not test as I don't have a cable for it. I'll forward it to wds.
Tapatalked from my Galaxy S II.
Agreed on Chrome--it was flaky on ICS for me a well.
---------- Post added at 11:26 AM ---------- Previous post was at 11:10 AM ----------
So far so good with 4.1. Couple of things I've noticed--the keyboard is much more sensitive than under ICS and repeats keys often (like the old double-bounce problem with the first electronic keyboards). Second, as has already been noted, Chrome isn't stable really.
I have another question somewhat unrelated--I have several email addresses but I route them all to one account. Can the stock email client allow me to change the From: address for an email? It shows it in the compose/reply screens but tapping on it doesn't do anything. I haven't found another client that makes this easy.
Need to do more testing, but I think the light sensor/automatic Brightness might not be working properly.
littleemp said:
Need to do more testing, but I think the light sensor/automatic Brightness might not be working properly.
Click to expand...
Click to collapse
Ah yeah forgot that one. It does work, however resume while on auto makes it 100% brightness. Wds is taking a look at the brightness levels as they are currently incorrect. It didn't bother ICS but JB seems to care.
Tapatalked from my Galaxy S II.
I remember a way to backup all your apps into a zip, will it work for jelleytime?
sent from my sexy-a$$-iconia tab
das_webos_transformer said:
I remember a way to backup all your apps into a zip, will it work for jelleytime?
sent from my sexy-a$$-iconia tab
Click to expand...
Click to collapse
I don't see why not, TiBu etc all work fine. Haven't tested that personally though, so maybe report back if it works or not?
Tapatalked from my Galaxy S II.
Anybody else had this issue? I tried it on Shiny Rom and now CM 10.2 nightly. Get in one time with no issues, if I reboot I get stuck at the Google screen. The one way to get the phone usable is to do a completely factory data wipe and start all over.
I'm using TWRP, I've flashed tons of roms and I'm doing it no different than any of them. I do a complete wipe, install CM, GApps, reboot I get in one time. Do a reboot and then get stuck at Google screen.
This happened with both ShinyROM and 10.2 both are 4.3. They both had a valid MD5.
I have flashed PRIMEMD04_Bootloader-signed - One that came from Shiny ROM bootloader-toro-primelc03 - Stock (I believe)
Nothing worked, got the same issue.
GiZiM said:
Anybody else had this issue? I tried it on Shiny Rom and now CM 10.2 nightly. Get in one time with no issues, if I reboot I get stuck at the Google screen. The one way to get the phone usable is to do a completely factory data wipe and start all over.
I'm using TWRP, I've flashed tons of roms and I'm doing it no different than any of them. I do a complete wipe, install CM, GApps, reboot I get in one time. Do a reboot and then get stuck at Google screen.
This happened with both ShinyROM and 10.2 both are 4.3. They both had a valid MD5.
I have flashed PRIMEMD04_Bootloader-signed - One that came from Shiny ROM bootloader-toro-primelc03 - Stock (I believe)
Nothing worked, got the same issue.
Click to expand...
Click to collapse
The only time I get persistent Google screen is when I have a corrupt /userdata partition.
Try running e2fsck on mmcblk0p12 if you have one of those devices that get emmc corruption by changes in sdcard free space.
In case you need more information on this :
1. In TWRP Mount options, unmount Data. The command won't work on mounted partition.
2. In Advanced > Terminal Command, select the /dev/block folder. Then type in
Code:
e2fsck -n mmcblk0p12
("-n" option will cause no changes to your filesystem, it just shows a simulation of the file system correction tool's result.) If the result says the filesystem is "clean", then you have something else to worry about and its not corrupted /userdata.
3. If it starts checking for errors (there's a KILL button while running the command), your filesystem has errors. You can kill it, and then run
Code:
e2fsck -f -C 0 -y mmcblk0p12
(-f is for forced run, -C 0 will give you progress of completion of task, -y will say yes to all prompts (necessary), and mmcblk0p12 is the /userdata partition)
4. Let the command run until you don't see the KILL button (shouldn't be more than a couple of minutes). On running
Code:
e2fsck -n mmcblk0p12
again, it must show you a "clean" filesystem.
5. Just reboot to System. If your /userdata was corrupt, running this will mostly solve your problem.
Damn, I wish I had found such instructions to get over my horrendous time with corrupt partitions over six months from November 2012-April 2013.
Thank you! I will look into this but I would love to fix the cause.
I'm sure this is the issue because I cannot get the phone back into a working state until after doing a full factory wipe.
What would be causing 4.3 roms to corrupt my /userdata? Whatever it is it is happening after the first reboot after I get into the rom.
desktopfusion said:
The only time I get persistent Google screen is when I have a corrupt /userdata partition.
Try running e2fsck on mmcblk0p12 if you have one of those devices that get emmc corruption by changes in sdcard free space.
In case you need more information on this :
1. In TWRP Mount options, unmount Data. The command won't work on mounted partition.
2. In Advanced > Terminal Command, select the /dev/block folder. Then type in
Code:
e2fsck -n mmcblk0p12
("-n" option will cause no changes to your filesystem, it just shows a simulation of the file system correction tool's result.) If the result says the filesystem is "clean", then you have something else to worry about and its not corrupted /userdata.
3. If it starts checking for errors (there's a KILL button while running the command), your filesystem has errors. You can kill it, and then run
Code:
e2fsck -f -C 0 -y mmcblk0p12
(-f is for forced run, -C 0 will give you progress of completion of task, -y will say yes to all prompts (necessary), and mmcblk0p12 is the /userdata partition)
4. Let the command run until you don't see the KILL button. On running
Code:
e2fsck -n mmcblk0p12
again, it must show you a "clean" filesystem.
5. Just reboot to System. If your /userdata was corrupt, running this will mostly solve your problem.
Damn, I wish I had found such instructions to get over my horrendous time with corrupt partitions over six months from November 2012-April 2013.
Click to expand...
Click to collapse
Whatever my phone faces is not limited to 4.3. It has been happening since 4.1. Something to do with the eMMC of the phone itself. Gets corrupt too easily when the decreasing free sdcard space approaches 3 GB. Sometimes a battery pull can also cause it, sometimes a random reboot can cause it. Technically, 4.3 is supposed to solve this issue with TRIM execution in the background, but it hasn't helped me definitively yet.
It has been documented on these forums for this to happen to phones produced in August, September, October 2012.
In my case, I am quite sure that adding and/or deleting files larger than say 500 MB (for example Nandroid backups) causes this problem to happen at any time afterwards without notice (preceded by random reboot straight into stuck Google screen). I guess new ROMs can cause such issues if there is a large amount of app data deleted everytime, but this is just uninformed speculation.
I've setup Tasker profiles to autorun e2fsck through TWRP every morning before I wake up so I don't have these issues while awake. And one more profile to tell me every 8 hours if my sdcard has changed by a lot.
I hope this is all there is to your phone's problem!
Thanks this is very helpful, I will work on this tonight after work and see what I can come up with.
Right now it is showing as CLEAN but i'm running 4.2.2. Is there anything special to go from 4.2.2 to 4.3?
desktopfusion said:
Whatever my phone faces is not limited to 4.3. It has been happening since 4.1. Something to do with the eMMC of the phone itself. Gets corrupt too easily when the decreasing free sdcard space approaches 3 GB. Sometimes a battery pull can also cause it, sometimes a random reboot can cause it. Technically, 4.3 is supposed to solve this issue with TRIM execution in the background, but it hasn't helped me definitively yet.
It has been documented on these forums for this to happen to phones produced in August, September, October 2012.
In my case, I am quite sure that adding and/or deleting files larger than say 500 MB (for example Nandroid backups) causes this problem to happen at any time afterwards without notice (preceded by random reboot straight into stuck Google screen). I guess new ROMs can cause such issues if there is a large amount of app data deleted everytime, but this is just uninformed speculation.
I've setup Tasker profiles to autorun e2fsck through TWRP every morning before I wake up so I don't have these issues while awake. And one more profile to tell me every 8 hours if my sdcard has changed by a lot.
I hope this is all there is to your phone's problem!
Click to expand...
Click to collapse
Nothing special really.
Was the result Clean after a stuck Google logo?
You could pull battery at the stuck logo and go to TWRP and then check if it is clean. If it is indeed clean, then there's something else wrong that is completely different!
I'm going to give it a go tonight when I get home.
1. So pretty much backup everything, do a complete factory wipe of all data.
2. I'll do a adb push of the .zip for the rom and everything.
3. Flash rom and any other packages needed
4. Run that command see if it comes back clean
5. Reboot go into the rom
6. Reboot, if i get stuck at the Google screen run that script if it's not clean just keep running it?
desktopfusion said:
Nothing special really.
Was the result Clean after a stuck Google logo?
You could pull battery at the stuck logo and go to TWRP and then check if it is clean. If it is indeed clean, then there's something else wrong that is completely different!
Click to expand...
Click to collapse
It's running, it says "clone multiply-claimed blocks? <yes>"
How long does this normally run?
Edit: Running now for 4hrs
GiZiM said:
It's running, it says "clone multiply-claimed blocks? <yes>"
How long does this normally run?
Edit: Running now for 4hrs
Click to expand...
Click to collapse
Hey, sorry.
4 hours?! I've not had it run for more than 2 minutes, ever!
Did you run the -n option and let it say whether there is a corrupt partition? If yes then there may be a problem with the eMMC since its taking so long.
As a last resort, you could go completely stock via fastboot. [Formats /userdata partition, erases SDcard as well - take backup]
Lifesaving string of code.
desktopfusion said:
The only time I get persistent Google screen is when I have a corrupt /userdata partition.
Try running e2fsck on mmcblk0p12 if you have one of those devices that get emmc corruption by changes in sdcard free space.
In case you need more information on this :
1. In TWRP Mount options, unmount Data. The command won't work on mounted partition.
2. In Advanced > Terminal Command, select the /dev/block folder. Then type in
Code:
e2fsck -n mmcblk0p12
("-n" option will cause no changes to your filesystem, it just shows a simulation of the file system correction tool's result.) If the result says the filesystem is "clean", then you have something else to worry about and its not corrupted /userdata.
3. If it starts checking for errors (there's a KILL button while running the command), your filesystem has errors. You can kill it, and then run
Code:
e2fsck -f -C 0 -y mmcblk0p12
(-f is for forced run, -C 0 will give you progress of completion of task, -y will say yes to all prompts (necessary), and mmcblk0p12 is the /userdata partition)
4. Let the command run until you don't see the KILL button. On running
Code:
e2fsck -n mmcblk0p12
again, it must show you a "clean" filesystem.
5. Just reboot to System. If your /userdata was corrupt, running this will mostly solve your problem.
Damn, I wish I had found such instructions to get over my horrendous time with corrupt partitions over six months from November 2012-April 2013.
Click to expand...
Click to collapse
Holy mother of god.
This was a lifesaving string of code. I was convinced that I was going to have to buy a brand new phone. Kudos to you DesktopFusion. Thanks a lot!
Explanation as to why a corrupt userdata partition causes this issue
So, in case anybody was wondering, I would like to explain how I see the issue at hand (corrupt userdata partition).
Upon first boot, the ROM is getting set up and doesn't have any orders to check /userdata because well, there is none on the device. So why waste precious time being redundant (smart coders, right?). So the userdata partition is not checked.
However, now that the second boot is being performed, there is, by default in the system's repertoir, data put in by the user, so the bootloader is told to check the partition containing user data. Since the partition is corrupt, and the free space is technically "modified", it errors up and gets stuck.
I could be wrong, I'm not a developer of any sorts. Just a dude giving my point of view.
I got this trouble too last night. I'm flashing elia's 4.3 custom rom with under 5 gb free space. Got it boot twice,then got stuck at google's logo after trying flashing slim bean gapps. Flashing whole 4.3 stock firmware finally solve this. Maybe there is correlation between upgrading from 4.2.2 without upgrading the bootloader with this problem.
Sent from my Galaxy Nexus using Tapatalk 4
It ended up running for 6hrs and never finished, I did do the -n and it said it was CLEAN but still could not boot into the 4.3 ROM.
So ran it any way for 6hrs and no change as of right now I am unable to boot into ANY 4.3 ROM.
If eMMC was screwed up why can 4.2.2 roms work fine but not 4.3 roms? what's the difference?
desktopfusion said:
Hey, sorry.
4 hours?! I've not had it run for more than 2 minutes, ever!
Did you run the -n option and let it say whether there is a corrupt partition? If yes then there may be a problem with the eMMC since its taking so long.
As a last resort, you could go completely stock via fastboot. [Formats /userdata partition, erases SDcard as well - take backup]
Click to expand...
Click to collapse
BUMP still no 4.3 love
GiZiM said:
BUMP still no 4.3 love
Click to expand...
Click to collapse
Still not resolved?
Have you tried flashing completely stock factory image via fastboot?
Consider checking your flash memory with this app, "eMMC Brickbug Check"
Could you also put in here what Type and Date the eMMC chip was made as per the application?
desktopfusion said:
Still not resolved?
Have you tried flashing completely stock factory image via fastboot?
Consider checking your flash memory with this app, "eMMC Brickbug Check"
Could you also put in here what Type and Date the eMMC chip was made as per the application?
Click to expand...
Click to collapse
Sane chip
Ran memory check no issues
Type: SEM32G
Date: 10/2011
FW: 0x90
GiZiM said:
Sane chip
Ran memory check no issues
Type: SEM32G
Date: 10/2011
FW: 0x90
Click to expand...
Click to collapse
Rats. Falling short of ideas.
I might have missed your answer on whether you've tried flashing a stock Nexus image through fastboot, though.
I had basically the same problem coming from 4.2.2 to 4.3. I'm not sure if it was the bootloader upgrade, 4.3 CM alpha from early August, or the small amount of free space under 5GB, but it was definitely userdata corruption like desktopfusion is saying. I was finally able to fix the problem after a "fastboot -w" userdata and cache*format. This has to be done in fastboot, so hook your computer to your phone in twrp, run "adb pull /sdcard "drive:\pc_backup_location"" to backup your sdcard data (grab what you can if errors come up from badly named or corrupted files). Make sure you got md5verified recovery & bootloader images renamed to recovery.img & bootloader.img, and a rom.zip in your adb folder on the computer. Reboot to bootloader, still connected. **And make sure you're not forgetting any sdcard backup, or twrp backup if still functional, before this erase and format. Run "fastboot -w". Then "fastboot flash recovery recovery.img" and you should be good to reboot, adb transfer ROM.zip and other files to fresh sdcard, etc. There's info on this around to double check, and I just posted a little more detailed walkthrough on this a couple days ago. This definitely saved me though with an identical problem, so if the emmc error check doesn't pan out, I would give this a try. May the fastboot gods be with you.
@desktopfusion -Thanks for your post on the previous page! Copied and saved to mygiant android useful infos text file ... which will one day end up corrupting my userdata partition with my luck.
7175 said:
I had basically the same problem coming from 4.2.2 to 4.3. I'm not sure if it was the bootloader upgrade, 4.3 CM alpha from early August, or the small amount of free space under 5GB, but it was definitely userdata corruption like desktopfusion is saying. I was finally able to fix the problem after a "fastboot -w" userdata and cache*format. This has to be done in fastboot, so hook your computer to your phone in twrp, run "adb pull /sdcard "drive:\pc_backup_location"" to backup your sdcard data (grab what you can if errors come up from badly named or corrupted files). Make sure you got md5verified recovery & bootloader images renamed to recovery.img & bootloader.img, and a rom.zip in your adb folder on the computer. Reboot to bootloader, still connected. **And make sure you're not forgetting any sdcard backup, or twrp backup if still functional, before this erase and format. Run "fastboot -w". Then "fastboot flash recovery recovery.img" and you should be good to reboot, adb transfer ROM.zip and other files to fresh sdcard, etc. There's info on this around to double check, and I just posted a little more detailed walkthrough on this a couple days ago. This definitely saved me though with an identical problem, so if the emmc error check doesn't pan out, I would give this a try. May the fastboot gods be with you.http://forum.xda-developers.com/images/smilies/highfive.gif
@desktopfusion -Thanks for your post on the previous page! Copied and saved to mygiant android useful infos text file ... which will one day end up corrupting my userdata partition with my luck.
Click to expand...
Click to collapse
7175 nailed it. I tried flashing all sorts of different ROMs/gapps/Super SU combinations and tried different wipes/formats to no avail. I wish I would have found the instructions to "fastboot cleanse" the phone 4 or 5 hours ago. I have been reading through forums for several hours today and every suggested fix failed. 7175's fix was spot on for me. My phone was locking up at the Google screen with the lock every time I rebooted after installing a 4.3 ROM.
:highfive:
7175 said:
I had basically the same problem coming from 4.2.2 to 4.3. I'm not sure if it was the bootloader upgrade, 4.3 CM alpha from early August, or the small amount of free space under 5GB, but it was definitely userdata corruption like desktopfusion is saying. I was finally able to fix the problem after a "fastboot -w" userdata and cache*format. This has to be done in fastboot, so hook your computer to your phone in twrp, run "adb pull /sdcard "drive:\pc_backup_location"" to backup your sdcard data (grab what you can if errors come up from badly named or corrupted files). Make sure you got md5verified recovery & bootloader images renamed to recovery.img & bootloader.img, and a rom.zip in your adb folder on the computer. Reboot to bootloader, still connected. **And make sure you're not forgetting any sdcard backup, or twrp backup if still functional, before this erase and format. Run "fastboot -w". Then "fastboot flash recovery recovery.img" and you should be good to reboot, adb transfer ROM.zip and other files to fresh sdcard, etc. There's info on this around to double check, and I just posted a little more detailed walkthrough on this a couple days ago. This definitely saved me though with an identical problem, so if the emmc error check doesn't pan out, I would give this a try. May the fastboot gods be with you.
@desktopfusion -Thanks for your post on the previous page! Copied and saved to mygiant android useful infos text file ... which will one day end up corrupting my userdata partition with my luck.
Click to expand...
Click to collapse
Thanks for this...
First thought it was an issue with SlimRom, tried different kernels, problem wouldn't go away, then realised it was happening with every rom I flashed...
Hello, I am posting this as a potential service to other flash happy DE owners. It may be obvious to some but I had to compile a number of semi-related Google searches to come by a solution for my issue.
I have been happily playing with my new DE and flashing our one ROM along with Xposed and a few modules. Everything was going just fine until I started having the phone reboot after a minute or so of starting up. I found that if I had Airplane Mode enabled it would boot just fine. Long story short, my phone would reboot every time AdFree tried to update the hosts file (evidently within a minute of the phone starting up if there was any sort of internet connection).
While trying to debug this I disabled Xposed and its modules as they can cause all kinds of problems - to no avail. I also tried re-flashing the ROM in TWRP and TWRP crashed and rebooted just as AdFree was doing. I then re-downloaded the ROM and rechecked its MD5 signature. Once I knew I had a good download, I again tried to flash it in TWRP and again my phone rebooted.
Again, long story short (too late!), I finally realized that my phone was rebooting anytime something tried to alter the /system partition. Some more Googling led me to a suggestion to check that partition for errors. To do this you must first boot into TWRP. Then connect your phone to your computer. If the phone doesn't install properly, go to Device Manager, you'll see under "other devices" an Android ADB with the yellow exclamation mark next to it. You "update drivers" and choose the drivers to install - choose Samsung ADB or Samsung ADB Interface. Now the device is recognized from TWRP (thanks unkellsam for outlining this). Our /system partition is /dev/block/mmcblk0p24 but owners of other devices can run the following commands to discover their /system partition (thanks android927 for outlining this):
Code:
adb shell
ls -al /dev/block/platform/*/by-name
Once you know your system partition, run the following code (leave out the "adb shell" if you have already run it and replace "mmcblk0p24" with your own device's /system partition if you don't have a Note 4 DE):
Code:
adb shell
e2fsck -fv /dev/block/mmcblk0p24
For me this produced the following results:
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Entry 'feature.xml.bak' in /csc (8021) has deleted/unused inode 2671. Clear<y>?
Entry 'PowerOn.ogg.bak' in /media/audio/ui (2449) has deleted/unused inode 2672. Clear<y>?
Pass 3: Checking directory connectivity
/lost+found not found. Create<y>?
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong for group #2 (1613, counted=1618).
Fix<y>?
Free blocks count wrong for group #5 (325, counted=337).
Fix<y>?
Free blocks count wrong for group #8 (1382, counted=1377).
Fix<y>?
Free blocks count wrong for group #20 (895, counted=913).
Fix<y>?
Free blocks count wrong (395151, counted=395181).
Fix<y>?
Free inodes count wrong for group #0 (5345, counted=5346).
Fix<y>?
Free inodes count wrong for group #16 (6741, counted=6742).
Fix<y>?
Free inodes count wrong (252552, counted=252554).
Fix<y>?
/dev/block/mmcblk0p24: ***** FILE SYSTEM WAS MODIFIED *****
3958 inodes used (1.54%)
48 non-contiguous files (1.2%)
4 non-contiguous directories (0.1%)
# of inodes with ind/dind/tind blocks: 0/0/0
Extent depth histogram: 3869
653395 blocks used (62.31%)
0 bad blocks
0 large files
3692 regular files
176 directories
0 character device files
0 block device files
0 fifos
0 links
80 symbolic links (80 fast symbolic links)
0 sockets
--------
3948 files
I said yes to each of the repair questions then ran the file scan again confirming no further errors. Booted back up and AdFree wrote the hosts file just fine.
I hope no one else runs into this kind of error but I hope this is helpful if someone else has similar issues.
I am a little concerned about the potential for a bad batch of memory in my phone but the scan did say no bad blocks so I am hoping it was just a bit of file corruption from my flash-happy weekend. Anyone else want to chime in on whether I should be worried about actual hardware failure?
Thanks
azumandias said:
Hello, I am posting this as a potential service to other flash happy DE owners. It may be obvious to some but I had to compile a number of semi-related Google searches to come by a solution for my issue.
I have been happily playing with my new DE and flashing our one ROM along with Xposed and a few modules. Everything was going just fine until I started having the phone reboot after a minute or so of starting up. I found that if I had Airplane Mode enabled it would boot just fine. Long story short, my phone would reboot every time AdFree tried to update the hosts file (evidently within a minute of the phone starting up if there was any sort of internet connection).
While trying to debug this I disabled Xposed and its modules as they can cause all kinds of problems - to no avail. I also tried re-flashing the ROM in TWRP and TWRP crashed and rebooted just as AdFree was doing. I then re-downloaded the ROM and rechecked its MD5 signature. Once I knew I had a good download, I again tried to flash it in TWRP and again my phone rebooted.
Again, long story short (too late!), I finally realized that my phone was rebooting anytime something tried to alter the /system partition. Some more Googling led me to a suggestion to check that partition for errors. Our /system partition is /dev/block/mmcblk0p24 so I booted up TWRP, connected my phone to my computer (forced it to install as a Samsung ADB Interface as it didn't automatically recognize it) and ran the following commands:
Code:
adb shell
e2fsck -fv /dev/block/mmcblk0p24
This produced the following results:
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Entry 'feature.xml.bak' in /csc (8021) has deleted/unused inode 2671. Clear<y>?
Entry 'PowerOn.ogg.bak' in /media/audio/ui (2449) has deleted/unused inode 2672. Clear<y>?
Pass 3: Checking directory connectivity
/lost+found not found. Create<y>?
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong for group #2 (1613, counted=1618).
Fix<y>?
Free blocks count wrong for group #5 (325, counted=337).
Fix<y>?
Free blocks count wrong for group #8 (1382, counted=1377).
Fix<y>?
Free blocks count wrong for group #20 (895, counted=913).
Fix<y>?
Free blocks count wrong (395151, counted=395181).
Fix<y>?
Free inodes count wrong for group #0 (5345, counted=5346).
Fix<y>?
Free inodes count wrong for group #16 (6741, counted=6742).
Fix<y>?
Free inodes count wrong (252552, counted=252554).
Fix<y>?
/dev/block/mmcblk0p24: ***** FILE SYSTEM WAS MODIFIED *****
3958 inodes used (1.54%)
48 non-contiguous files (1.2%)
4 non-contiguous directories (0.1%)
# of inodes with ind/dind/tind blocks: 0/0/0
Extent depth histogram: 3869
653395 blocks used (62.31%)
0 bad blocks
0 large files
3692 regular files
176 directories
0 character device files
0 block device files
0 fifos
0 links
80 symbolic links (80 fast symbolic links)
0 sockets
--------
3948 files
I said yes to each of the repair questions then ran the file scan again confirming no further errors. Booted back up and AdFree wrote the hosts file just fine.
I hope no one else runs into this kind of error but I hope this is helpful if someone else has similar issues.
I am a little concerned about the potential for a bad batch of memory in my phone but the scan did say no bad blocks so I am hoping it was just a bit of file corruption from my flash-happy weekend. Anyone else want to chime in on whether I should be worried about actual hardware failure?
Thanks
Click to expand...
Click to collapse
Great thread, thank you. I'm not having re-boot issues but then again I haven't flashed the ROM you mention. I wonder, or do you have any idea, what caused the errors in memory?
Well, I am certainly hoping it is not bad memory hardware. In my effort to break in the battery there was at least one instance of the phone shutting down due to a drained battery which may have caused a file to not be written properly (although I don't know why system .ogg file would have been affected). It could also have been due to the Xposed modules. I guess I can't give a good answer to your question.
Thanks a lot for posting this! I'm having the same symptoms on my Note 4 every time I try to write to the hosts file even though I had no problems before. I get the crash/reboot in recovery as well. I will try your fix asap.
Hey I'm having a little trouble connecting through ADB from recovery. I can connect to ADB fine when the phone is on, but no devices are recongnized when I boot into TWRP.
I also tried "ADB sideload," but it just gets stuck on "Starting ADB sideload feature" indefinitely. Is there any other way to repair the file system?
I tried doing it with the phone on, but I get an error saying "...block mmcblk0p24 is in use."
I also tried the "Terminal Command" in TWRP but the repair command gives back: "need terminal for interactive repairs." Any help would be appreciated.
unkellsam said:
Hey I'm having a little trouble connecting through ADB from recovery. I can connect to ADB fine when the phone is on, but no devices are recongnized when I boot into TWRP.
I also tried "ADB sideload," but it just gets stuck on "Starting ADB sideload feature" indefinitely. Is there any other way to repair the file system?
I tried doing it with the phone on, but I get an error saying "...block mmcblk0p24 is in use."
I also tried the "Terminal Command" in TWRP but the repair command gives back: "need terminal for interactive repairs." Any help would be appreciated.
Click to expand...
Click to collapse
Bump. How did you force your phone to install as Samsung ADB, azumandias?
EDIT: I figured it out, and your fix works perfectly, thanks! I was not able to ADB from TWRP. Now I can flash stuff without the phone crashing!! The way to do it is to go to device manager, you'll see under "other devices" an Android ADB with the yellow exclamation mark next to it. You "update drivers" and choose the drivers to install - choose Samsung ADB. Now the device is recognized from TWRP.
unkellsam said:
Bump. How did you force your phone to install as Samsung ADB, azumandias?
EDIT: I figured it out, and your fix works perfectly, thanks! I was not able to ADB from TWRP. Now I can flash stuff without the phone crashing!! The way to do it is to go to device manager, you'll see under "other devices" an Android ADB with the yellow exclamation mark next to it. You "update drivers" and choose the drivers to install - choose Samsung ADB. Now the device is recognized from TWRP.
Click to expand...
Click to collapse
Sorry, was down with the stomach flu and didn't see your post. Thanks for laying out the step by step for getting the phone to connect in TWRP. Glad you are back up and running. Mind if I edit my first post to include your steps (with proper recognition of course)?
azumandias said:
Sorry, was down with the stomach flu and didn't see your post. Thanks for laying out the step by step for getting the phone to connect in TWRP. Glad you are back up and running. Mind if I edit my first post to include your steps (with proper recognition of course)?
Click to expand...
Click to collapse
Please do!
azumandias said:
Hello, I am posting this as a potential service to other flash happy DE owners. It may be obvious to some but I had to compile a number of semi-related Google searches to come by a solution for my issue.
I have been happily playing with my new DE and flashing our one ROM along with Xposed and a few modules. Everything was going just fine until I started having the phone reboot after a minute or so of starting up. I found that if I had Airplane Mode enabled it would boot just fine. Long story short, my phone would reboot every time AdFree tried to update the hosts file (evidently within a minute of the phone starting up if there was any sort of internet connection).
While trying to debug this I disabled Xposed and its modules as they can cause all kinds of problems - to no avail. I also tried re-flashing the ROM in TWRP and TWRP crashed and rebooted just as AdFree was doing. I then re-downloaded the ROM and rechecked its MD5 signature. Once I knew I had a good download, I again tried to flash it in TWRP and again my phone rebooted.
Again, long story short (too late!), I finally realized that my phone was rebooting anytime something tried to alter the /system partition. Some more Googling led me to a suggestion to check that partition for errors. To do this you must first boot into TWRP. Then connect your phone to your computer. If the phone doesn't install properly, go to Device Manager, you'll see under "other devices" an Android ADB with the yellow exclamation mark next to it. You "update drivers" and choose the drivers to install - choose Samsung ADB or Samsung ADB Interface. Now the device is recognized from TWRP (thanks unkellsam for outlining this). Our /system partition is /dev/block/mmcblk0p24 so run the following commands:
Code:
adb shell
e2fsck -fv /dev/block/mmcblk0p24
For me this produced the following results:
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Entry 'feature.xml.bak' in /csc (8021) has deleted/unused inode 2671. Clear<y>?
Entry 'PowerOn.ogg.bak' in /media/audio/ui (2449) has deleted/unused inode 2672. Clear<y>?
Pass 3: Checking directory connectivity
/lost+found not found. Create<y>?
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong for group #2 (1613, counted=1618).
Fix<y>?
Free blocks count wrong for group #5 (325, counted=337).
Fix<y>?
Free blocks count wrong for group #8 (1382, counted=1377).
Fix<y>?
Free blocks count wrong for group #20 (895, counted=913).
Fix<y>?
Free blocks count wrong (395151, counted=395181).
Fix<y>?
Free inodes count wrong for group #0 (5345, counted=5346).
Fix<y>?
Free inodes count wrong for group #16 (6741, counted=6742).
Fix<y>?
Free inodes count wrong (252552, counted=252554).
Fix<y>?
/dev/block/mmcblk0p24: ***** FILE SYSTEM WAS MODIFIED *****
3958 inodes used (1.54%)
48 non-contiguous files (1.2%)
4 non-contiguous directories (0.1%)
# of inodes with ind/dind/tind blocks: 0/0/0
Extent depth histogram: 3869
653395 blocks used (62.31%)
0 bad blocks
0 large files
3692 regular files
176 directories
0 character device files
0 block device files
0 fifos
0 links
80 symbolic links (80 fast symbolic links)
0 sockets
--------
3948 files
I said yes to each of the repair questions then ran the file scan again confirming no further errors. Booted back up and AdFree wrote the hosts file just fine.
I hope no one else runs into this kind of error but I hope this is helpful if someone else has similar issues.
I am a little concerned about the potential for a bad batch of memory in my phone but the scan did say no bad blocks so I am hoping it was just a bit of file corruption from my flash-happy weekend. Anyone else want to chime in on whether I should be worried about actual hardware failure?
Thanks
Click to expand...
Click to collapse
Thanks so much for this! I was having a similar issue on my htc one m9 and this fixed it. The only change i had to make was the mount point, as the m9 has its system partition mounted differently, but that was easy to figure out; all i did was put the phone in recovery, plug it into my computer, and then run the command "ls -al /dev/block/platform/*/by-name" from the adb shell in command prompt. That gave me a list of all the mount points and their associated partitions, so i just copied the mount point for /system and ran e2fsck using that mount point.
android927 said:
Thanks so much for this! I was having a similar issue on my htc one m9 and this fixed it. The only change i had to make was the mount point, as the m9 has its system partition mounted differently, but that was easy to figure out; all i did was put the phone in recovery, plug it into my computer, and then run the command "ls -al /dev/block/platform/*/by-name" from the adb shell in command prompt. That gave me a list of all the mount points and their associated partitions, so i just copied the mount point for /system and ran e2fsck using that mount point.
Click to expand...
Click to collapse
android927, I'm glad you found this and it got you back on your feet. Do you mind if I update the original post with your instructions on finding the system partition? It may help others get going quicker in the future.
Hi, a little background first. My phone is s-Off, bootloader is unlocked and I have a been running rooted custom roms since day 1 that I had this phone. I always ran good except that 2 months ago it started rebooting for no reason. I change my rom from InsertCoin to LeeDroid thinking that my help but no change. Now, yesterday was phone rebooted in the middle of playing a YouTube video right into safe mode asking for a password to Decrypt storage which I never encrypted.
I found a few threads about it. Here's what I tried:
- wipe user data: tried that with fastboot erase userdata and also fastboot format userdata, both failed.
- I went into TWRP to wipe data, said it ran successfully but it didn't since all my pictures and other personal files.
- I tried restoring a known working backup I had and it failed too
- fastboot oem lock fails
- fastboot rebootruu reboots in a blackscreen (no HTC logo in the background) then completely reboots back in safemode.
Any idea what I should try next?
Thanks a lot
anyone? I really desperate right now :crying:
stoned99 said:
- I went into TWRP to wipe data, said it ran successfully but it didn't since all my pictures and other personal files.
Click to expand...
Click to collapse
This is working correctly. You need to go into the advanced wipe options, and select Internal Storage to wipe that partition.
Try the "Format Data" option, and see if that helps.
stoned99 said:
- fastboot oem lock fails
Click to expand...
Click to collapse
I don't ever recommend doing this on an s-off device (for troubleshooting purposes). It doesn't enable anything that isn't already enabled under s-off (running RUUs is the most common reason); but can leave you stuck unable to do a heck of a lot of things (installing custom recovery, ROMs, etc.).
Does your version have an RUU that you can try?
redpoint73 said:
This is working correctly. You need to go into the advanced wipe options, and select Internal Storage to wipe that partition.
Try the "Format Data" option, and see if that helps.
I don't ever recommend doing this on an s-off device (for troubleshooting purposes). It doesn't enable anything that isn't already enabled under s-off (running RUUs is the most common reason); but can leave you stuck unable to do a heck of a lot of things (installing custom recovery, ROMs, etc.).
Does your version have an RUU that you can try?
Click to expand...
Click to collapse
thanks a lot for the reply. I tried the format option. Typed yes in the field. It said Updating Partitions details.... done. But when I go in the file manager I can see I didn't lose anything. Also if I mount my sdcard (data/media) and delete all the files from my PCs. all the files are being deleted. I rebooted the phone and they are all back.
Yes I have a RUU. I ran both using the exe which failed saying it could be an issue with the usb cable or something. Tried with the zip file renamed to img file (can't remember the exam name), hboot detected it then asked me if I wanted to update, I select yes by using volume 1, then I saw Update, 1 or 2 second later the phone rebooted in a black screen (not RUU), then rebooted in safe mode again asking to Decrypt Storage.
I tried the info found here http://android-revolution-hd.blogspot.ca/2014/09/how-to-fix-random-reboots-on-htc-one-m8.html but it didn't fix my data partition. Here's the output.
~ # ←[6n/tmp/fsck.ext4 -fn /dev/block/mmcblk0p47
/tmp/fsck.ext4 -fn /dev/block/mmcblk0p47
e2fsck 1.42.9 (28-Dec-2013)
Warning: skipping journal recovery because doing a read-only filesystem check.
Pass 1: Checking inodes, blocks, and sizes
Inodes that were part of a corrupted orphan linked list found. Fix? no
Inode 360574 was part of the orphaned inode list. IGNORED.
Deleted inode 368721 has zero dtime. Fix? no
Inode 426053 was part of the orphaned inode list. IGNORED.
Inode 426055 was part of the orphaned inode list. IGNORED.
Inode 696341 was part of the orphaned inode list. IGNORED.
Inode 712757 was part of the orphaned inode list. IGNORED.
Inode 745473 was part of the orphaned inode list. IGNORED.
Inode 1236999 was part of the orphaned inode list. IGNORED.
Inode 1237006 was part of the orphaned inode list. IGNORED.
Inode 1237021 was part of the orphaned inode list. IGNORED.
Inode 1237022 was part of the orphaned inode list. IGNORED.
Inode 1237036 was part of the orphaned inode list. IGNORED.
Inode 1237043 was part of the orphaned inode list. IGNORED.
Inode 1237046 was part of the orphaned inode list. IGNORED.
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(1159609--1159624) -(1170414--1170429) -1562594 -(15
65703--1565709) -1704549 -(2531003--2531013) -2743942 -(3047984--3047992) -34477
28 -(4796517--4796523) -6504126 -6504155 -6504168 -6504725
Fix? no
Free blocks count wrong (2592986, counted=2598704).
Fix? no
Inode bitmap differences: -360574 -368721 -426053 -426055 -696341 -712757 -7454
73 -1236999 -1237006 -(1237021--1237022) -1237036 -1237043 -1237046
Fix? no
Directories count wrong for group #52 (27, counted=26).
Fix? no
Free inodes count wrong (1653079, counted=1650912).
Fix? no
/dev/block/mmcblk0p47: ********** WARNING: Filesystem still has errors *********
*
/dev/block/mmcblk0p47: 9897/1662976 files (29.1% non-contiguous), 4058918/665190
4 blocks
Click to expand...
Click to collapse
stoned99 said:
thanks a lot for the reply. I tried the format option. Typed yes in the field. It said Updating Partitions details.... done. But when I go in the file manager I can see I didn't lose anything. Also if I mount my sdcard (data/media) and delete all the files from my PCs. all the files are being deleted. I rebooted the phone and they are all back.
Yes I have a RUU. I ran both using the exe which failed saying it could be an issue with the usb cable or something. Tried with the zip file renamed to img file (can't remember the exam name), hboot detected it then asked me if I wanted to update, I select yes by using volume 1, then I saw Update, 1 or 2 second later the phone rebooted in a black screen (not RUU), then rebooted in safe mode again asking to Decrypt Storage.
I tried the info found here http://android-revolution-hd.blogspot.ca/2014/09/how-to-fix-random-reboots-on-htc-one-m8.html but it didn't fix my data partition. Here's the output.
Click to expand...
Click to collapse
i would run a full ruu of the dev edition exe in fastboot usb mode and see if that fix the problem
toysoldierq said:
i would run a full ruu of the dev edition exe in fastboot usb mode and see if that fix the problem
Click to expand...
Click to collapse
Thanks. I already tried running the DEV RUU zip file by renaming it to that PXXXIMG.zip without a success. Since fastbootruu doesn't work I can tried it. Regarding the dev edition exe, I can't find it anywhere, only found the zip one. Do you know where I could find it?
stoned99 said:
Thanks. I already tried running the DEV RUU zip file by renaming it to that PXXXIMG.zip without a success. Since fastbootruu doesn't work I can tried it. Regarding the dev edition exe, I can't find it anywhere, only found the zip one. Do you know where I could find it?
Click to expand...
Click to collapse
Right here: http://www.htc.com/us/support/htc-one-m8/news/
But you need to change to the appropriate CID and MID. Plus, if your version's RUU exe didn't work, I don't see how this will be any different.
To be clear, you usually run the RUU in bootloader-fastboot mode, not RUU mode (it will boot to that itself once you run the RUU).
But if you have a fundamental USB connectivity issue, it really doesn't matter, and it still won't run. Try different cables, different USB ports, even another PC.
redpoint73 said:
Right here: http://www.htc.com/us/support/htc-one-m8/news/
But you need to change to the appropriate CID and MID. Plus, if your version's RUU exe didn't work, I don't see how this will be any different.
To be clear, you usually run the RUU in bootloader-fastboot mode, not RUU mode (it will boot to that itself once you run the RUU).
But if you have a fundamental USB connectivity issue, it really doesn't matter, and it still won't run. Try different cables, different USB ports, even another PC.
Click to expand...
Click to collapse
This is not a USB issue. The problem itself seems to be an ext4 corruption issue. When trying to restore my backups, it fails when trying to restore /data/media. I can't delete/format it either. I'll try the RUU exe but doubt it will work. Thanks for the help
stoned99 said:
This is not a USB issue.
Click to expand...
Click to collapse
You said in Post #4:
stoned99 said:
Yes I have a RUU. I ran both using the exe which failed saying it could be an issue with the usb cable or something.
Click to expand...
Click to collapse
I'm not saying the USB connectivity is causing the main Safe Mode/Decrypt Storage issue. But it does seem to be keeping the RUU from running, based on your previous post.
Also, I'm pretty sure I've seen this similar issue posted before (encrypted storage asking for password, when encryption wasn't selected by the user). From what I remember, the format storage option in TWRP (that you already tried) was the solution. But you might just search for those threads, to see if my recollection on the fix is incorrect; or if other fixes were found.
redpoint73 said:
You said in Post #4:
I'm not saying the USB connectivity is causing the main Safe Mode/Decrypt Storage issue. But it does seem to be keeping the RUU from running, based on your previous post.
Also, I'm pretty sure I've seen this similar issue posted before (encrypted storage asking for password, when encryption wasn't selected by the user). From what I remember, the format storage option in TWRP (that you already tried) was the solution. But you might just search for those threads, to see if my recollection on the fix is incorrect; or if other fixes were found.
Click to expand...
Click to collapse
Indeed, I looked at a few posts with that format storage issue and there was basically 2 solutions: 1) format data in TWRP. 2) Repair ext4 filesystem which didn't work either.
PS: Just saw that the link you gave me is for the Sense version, not the Google Edition. That's what I already tried and failed.
stoned99 said:
Indeed, I looked at a few posts with that format storage issue and there was basically 2 solutions: 1) format data in TWRP. 2) Repair ext4 filesystem which didn't work either.
PS: Just saw that the link you gave me is for the Sense version, not the Google Edition. That's what I already tried and failed.
Click to expand...
Click to collapse
You asked for Developer's Edition. Dev Edition and Google Play Edition (GPE) are 2 completely different things. Dev Edition is always Sense.
I've never seen an RUU.exe for GPE.
redpoint73 said:
You asked for Developer's Edition. Dev Edition and Google Play Edition (GPE) are 2 completely different things. Dev Edition is always Sense.
I've never seen an RUU.exe for GPE.
Click to expand...
Click to collapse
Oops my bad then. Sorry, probably my bad cold/sinusitis. I'm having a hard time concentrating.
Ok, so I tried that exe already and it didn't work. Any other ideas?
stoned99 said:
Oops my bad then. Sorry, probably my bad cold/sinusitis. I'm having a hard time concentrating.
Ok, so I tried that exe already and it didn't work. Any other ideas?
Click to expand...
Click to collapse
do you have teamviewer if so get it and let me help you, message private with the login and password.
stoned99 said:
Oops my bad then. Sorry, probably my bad cold/sinusitis. I'm having a hard time concentrating.
Click to expand...
Click to collapse
Heh, no worries. The last couple weeks have been rough for me (as far as illnesses), as well. And Dev Ed vs. GPE is a fairly common mistake.
Hopefully toysoldierq can help you out. You've tried everything I can think of.
redpoint73 said:
Heh, no worries. The last couple weeks have been rough for me (as far as illnesses), as well. And Dev Ed vs. GPE is a fairly common mistake.
Hopefully toysoldierq can help you out. You've tried everything I can think of.
Click to expand...
Click to collapse
I think he did not do the whole process going back to sense using this 1.54.401.5-W_Splash
toysoldierq said:
I think he did not do the whole process going back to sense using this 1.54.401.5-W_Splash
Click to expand...
Click to collapse
Sorry I'm back now. So what do you think I didn't try? Using the sense exe ruu? I tried it but it failed.
BTW regarding team viewer I'm not sure if that will really help since you won't be able to see my phone. I know how to use adb, fastboot, etc so if you have an idea of what I should try I would sure try it. Thanks a lot
stoned99 said:
Sorry I'm back now. So what do you think I didn't try? Using the sense exe ruu? I tried it but it failed.
BTW regarding team viewer I'm not sure if that will really help since you won't be able to see my phone. I know how to use adb, fastboot, etc so if you have an idea of what I should try I would sure try it. Thanks a lot
Click to expand...
Click to collapse
i have fix m8 phones over teamviewer already son im way ahead of you, if you want me fix the issue let me know or i can tell u what u need to do choice is yours.
step 1 well i reflash the gpe rooted rom
step 2 i change his mid to dev edtion one
step 3 change his cid to super cid
step 4 i flash 1.54.401.5-W_Splash and let him wipe data and cache in stock recovery
step 5 change his cid to dev
step 6 ran dev ruu the lateset
Ok we can try it over team viewer, give me a few minutes to install it and give you access
stoned99 said:
Ok we can try it over team viewer, give me a few minutes to install it and give you access
Click to expand...
Click to collapse
i gave you the instructions u can try that.
toysoldierq said:
i have fix m8 phones over teamviewer already son im way ahead of you, if you want me fix the issue let me know or i can tell u what u need to do choice is yours.
step 1 well i reflash the gpe rooted rom
step 2 i change his mid to dev edtion one
step 3 change his cid to super cid
step 4 i flash 1.54.401.5-W_Splash and let him wipe data and cache in stock recovery
step 5 change his cid to dev
step 6 ran dev ruu the lateset
Click to expand...
Click to collapse
Thanks much appreciated.
Would you reboot the phone between each steps? Maybe not because 2 and 3
A week ago I flashed the current (lineage-14.1-20170504-nightly-oneplus3-signed.zip) lineage os nightly to my phone. I did not pay that much attention til I realize the impact, so no guarantee for the temporal order. I am pretty shure, that in the first place, I tried to flash the image via the update menu (is this called OTA update?). The previous update two weeks ago already results in partial data los. By this I mean some (maybe all) apps disappears after the update, but after installing them again, all of them provided the pre-update data - only exception was signal. This time the phone was like in factory state. But when I try to reinstall the telegram app - the app store refuses to install the app because of the presence of a identical named app.
Using adb shell shows me, that all apps and data where still present in the /data subdirectories. Even those I didn't had reinstalled til that time . So I took advantage of momentum and backed up some files and decided to perform a full pre factory reset formatting everything twrp let me format. After flashing the lineage update, another full format because things seems still weird, second lineage nightly flash, realizing I have to first flash OOS-x.y.z and a final third lineage flash, I thought I finally obtain a running system. After initial configuration I decided to only copy my previous wpa_supplicant.conf file back to the system. Since overwriting this file while running android didn't led to anything, I tried to replace this file in recovery mode. Instead of remembering my password, goatish oneplus 3 desides to perform factory reset.
After what seems to be a lower six digit number of flashing attemps I came to the conclusion that performing any modification to the file system while the native kernel is not running results in factory reseting the device. In the mean time I also figured out what dm-verity means - an error message I got used to and maybe a nice feature as long as you don't try to take your old config files to a new system. And I know I am the millionth person facing that problem in a dedicated thread, but I found nothing which links this annoying boot loader message to my reset problem.
I made my previous recovery attemps between the toilet and my bed, so I hadn't any strategy at all. But thank god its friday! I read: dm-verity is a feature of the device mapper which guarantees data integrity. When I alter the data using any other than the native system, the device mapper notices the integrity violation and reports to the system. I am only guessing that this makes the system to reset itself - Am I right? Even if I appreciate this concept, I think this might be a little too ambitious. Can I alter androids behaviour in this point? In the best case I want to temporary disable this feature, do some root stuff and turn it on again. Is this possible?
I could follow one of that numerous tutorials to remove that dm-verity message, but I am not confortable with flashing zip files from dubious file hosters as well as copy-pasting commands without any idea of what I am doing. So maybe someone could provide a little more information than the seven step recipe of getting rid of that error message blog posts.
I really appreciate material which explains my problem. I am quite new to android and I didn't figured out how to maneuver around all these google results which tells me to install a specific app which may will hide my problem. So thank you in advance.
hinerk0815 said:
A week ago I flashed the current (lineage-14.1-20170504-nightly-oneplus3-signed.zip) lineage os nightly to my phone. I did not pay that much attention til I realize the impact, so no guarantee for the temporal order. I am pretty shure, that in the first place, I tried to flash the image via the update menu (is this called OTA update?). The previous update two weeks ago already results in partial data los. By this I mean some (maybe all) apps disappears after the update, but after installing them again, all of them provided the pre-update data - only exception was signal. This time the phone was like in factory state. But when I try to reinstall the telegram app - the app store refuses to install the app because of the presence of a identical named app.
Using adb shell shows me, that all apps and data where still present in the /data subdirectories. Even those I didn't had reinstalled til that time . So I took advantage of momentum and backed up some files and decided to perform a full pre factory reset formatting everything twrp let me format. After flashing the lineage update, another full format because things seems still weird, second lineage nightly flash, realizing I have to first flash OOS-x.y.z and a final third lineage flash, I thought I finally obtain a running system. After initial configuration I decided to only copy my previous wpa_supplicant.conf file back to the system. Since overwriting this file while running android didn't led to anything, I tried to replace this file in recovery mode. Instead of remembering my password, goatish oneplus 3 desides to perform factory reset.
After what seems to be a lower six digit number of flashing attemps I came to the conclusion that performing any modification to the file system while the native kernel is not running results in factory reseting the device. In the mean time I also figured out what dm-verity means - an error message I got used to and maybe a nice feature as long as you don't try to take your old config files to a new system. And I know I am the millionth person facing that problem in a dedicated thread, but I found nothing which links this annoying boot loader message to my reset problem.
I made my previous recovery attemps between the toilet and my bed, so I hadn't any strategy at all. But thank god its friday! I read: dm-verity is a feature of the device mapper which guarantees data integrity. When I alter the data using any other than the native system, the device mapper notices the integrity violation and reports to the system. I am only guessing that this makes the system to reset itself - Am I right? Even if I appreciate this concept, I think this might be a little too ambitious. Can I alter androids behaviour in this point? In the best case I want to temporary disable this feature, do some root stuff and turn it on again. Is this possible?
I could follow one of that numerous tutorials to remove that dm-verity message, but I am not confortable with flashing zip files from dubious file hosters as well as copy-pasting commands without any idea of what I am doing. So maybe someone could provide a little more information than the seven step recipe of getting rid of that error message blog posts.
I really appreciate material which explains my problem. I am quite new to android and I didn't figured out how to maneuver around all these google results which tells me to install a specific app which may will hide my problem. So thank you in advance.
Click to expand...
Click to collapse
Your story is too long to go through. But most likely your issue will be solved if you install blue sparks twrp. Check his thread in the unified section