NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery

[natemcnutty]

Originally Posted by pts69666

I ran the provisioning built into the ROM (it had Sprint provisioning in it) and I ran a provisioning cab file. Both to no avail. Government computers don't like USB devices, so I can't try flashing stock ROM until I get off.

Although you may have no way to get it on there, you can actually extract the NBH from the stock ROM's EXE and flash that from the sd card as well.

[[ACL]]

finally got word about the autobuild server. Looks like we had a power supply issue.

I'm going to back up some stuff tonight and have an alternate download link. Wistlitl is going to upgrade us to a new server so make sure to thank him if yall see him on IRC.

[natemcnutty]

Originally Posted by [ACL]

finally got word about the autobuild server. Looks like we had a power supply issue.

I'm going to back up some stuff tonight and have an alternate download link. Wistlitl is going to upgrade us to a new server so make sure to thank him if yall see him on IRC.

That's a bummer about the power supply. Glad that Wiltilt is able to help us out. If you have a backup of the autobuild stuff, can you shoot me the details? I'd love to see how that all works and maybe set my server up to do something similar

[pts69666]

Originally Posted by natemcnutty

Although you may have no way to get it on there, you can actually extract the NBH from the stock ROM's EXE and flash that from the sd card as well.

SD reader is still a USB device. =P If any USB device is plugged in, it forwards it over to the sysadmins...

Off work now, but too tired/lazy to flash atm. Will flash after nap and gym. =P

[[ACL]]

Originally Posted by natemcnutty

That's a bummer about the power supply. Glad that Wiltilt is able to help us out. If you have a backup of the autobuild stuff, can you shoot me the details? I'd love to see how that all works and maybe set my server up to do something similar

Ill send you over the script i got from Netripper. Its very flexible incase you wanna run your own autobuilds.Its good because it will always use a clean pull from a separate machine thats not affected by dirty code. But i'll shoot you over the email.

[natemcnutty]

Originally Posted by [ACL]

nate/lmiller:

ok so the damn autobuild server is down. So i cant post some new stuff yet. I did manage to commit everything to tinboot so thats all uptodate.

I also commited some stuff to bootenv. Mostly the portion about the modules. Now we are deploying the modules as an sqsh and it gets mounted to the correct folder in system. This way we can wipe and restore with no issues. Actually thanks to this new setup, ive been able to install and restore my data in the same recovery session. All without losing my modules. Only downside is you will need mksquashfs version 3.4 to make this file.

Check out the commits and yall will see. Once autobuild comes up you guys can download it and dissect it

Was getting an invalid mount, so I committed loop3 thinking that would fix it. Then got another error that it was an invalid argument. Finally figured out that I simply did sudo apt-get install squashfs-tools and ended up with v4, not 3.4...

Anyway, I did get system.ext2 loop mount off sdcard then cp -a everything over to /system working perfectly. I tried using dd to do a low level copy of everything, but that failed on me for some reason. I'll commit that change to the script in recovery shortly.

[[ACL]]

Originally Posted by natemcnutty

Was getting an invalid mount, so I committed loop3 thinking that would fix it. Then got another error that it was an invalid argument. Finally figured out that I simply did sudo apt-get install squashfs-tools and ended up with v4, not 3.4...

Anyway, I did get system.ext2 loop mount off sdcard then cp -a everything over to /system working perfectly. I tried using dd to do a low level copy of everything, but that failed on me for some reason. I'll commit that change to the script in recovery shortly.

yeah i updated the code right before your commit. I actually moved it to use loop0.

I'm actually just using the 3.4 mksquasfs binary so i dont have to mess with the versioning for squash fs tools.

Autobuild should spit out an image shortly. I was able to install fresh, restore without reboot and then after a reboot get back to android with a new kernel and no module issues so i think we are set for now. So everyone should be able to now continue without losing anything. Even after a full upgrade

ive been using the phone for about 4 hours and i do notice some small bugs that need fixing.. I killed my batt on purpose to work on that charging issue.

[natemcnutty]

Wirelessly posted (Opera/9.80 (Android; Opera Mini/6.0.24212/24.743; U; en) Presto/2.5.25 Version/10.54)

Originally Posted by [ACL

]

yeah i updated the code right before your commit. I actually moved it to use loop0.

I'm actually just using the 3.4 mksquasfs binary so i dont have to mess with the versioning for squash fs tools.

Autobuild should spit out an image shortly. I was able to install fresh, restore without reboot and then after a reboot get back to android with a new kernel and no module issues so i think we are set for now. So everyone should be able to now continue without losing anything. Even after a full upgrade

ive been using the phone for about 4 hours and i do notice some small bugs that need fixing.. I killed my batt on purpose to work on that charging issue.

That sounds awesome. I am still having issues getting FRX06 working properly. I can get it to boot without issues, but I cannot get data working for the life of me. The method of copying directly out of system.ext2 and then using the androidupdate.tgz to fix the rest seems to work great though

[arrrghhh]

Originally Posted by [ACL]

yeah i updated the code right before your commit. I actually moved it to use loop0.

I'm actually just using the 3.4 mksquasfs binary so i dont have to mess with the versioning for squash fs tools.

Autobuild should spit out an image shortly. I was able to install fresh, restore without reboot and then after a reboot get back to android with a new kernel and no module issues so i think we are set for now. So everyone should be able to now continue without losing anything. Even after a full upgrade

ive been using the phone for about 4 hours and i do notice some small bugs that need fixing.. I killed my batt on purpose to work on that charging issue.

Good news bro! That is very good news indeed... No more need to constantly reflash!

[Lmiller1708]

Wirelessly posted (TP2: Mozilla/5.0 (Linux; U; Android 2.2.2; en-us; MSM Build/FRX05) AppleWebKit/533.1 (KHTML, like Gecko) Version/4.0 Mobile Safari/533.1)

Originally Posted by [ACL

]

yeah i updated the code right before your commit. I actually moved it to use loop0.

I'm actually just using the 3.4 mksquasfs binary so i dont have to mess with the versioning for squash fs tools.

Autobuild should spit out an image shortly. I was able to install fresh, restore without reboot and then after a reboot get back to android with a new kernel and no module issues so i think we are set for now. So everyone should be able to now continue without losing anything. Even after a full upgrade

ive been using the phone for about 4 hours and i do notice some small bugs that need fixing.. I killed my batt on purpose to work on that charging issue.

That's great news acl!
Can't wait to test it out...