IssueTracker

what is in update.zip ? can you share that file to triage this issue

MHC19Q for Nexus 5X (bullhead) - https://dl.google.com/dl/android/aosp//bullhead-ota-mhc19q-8fe67a2b.zip
N4F26T for Nexus 5X (bullhead) - https://dl.google.com/dl/android/aosp/bullhead-ota-n4f26t-648ce802.zip
WW-12.2.5.23 for Asus ZenFone Go (ASUS_X014D) - http://dlcdnet.asus.com/pub/ASUS/ZenFone/ZB452KG/UL-ASUS_X014D-WW-12.2.5.23-user.zip

Look like, this happens only with a big images.

On Windows 10 (x64)

related to (and probably worked around for now by) internal bug http://b/36046324

i do think we should also start shipping a 64-bit Windows platform tools package.

we should also consider changing android::base::ReadFdToString to call fstat and pre-size the vector. not worthwhile for reading things like /proc/uptime, but the default expansion behavior for a huge file like a full OTA update is going to result in substantial overhead. (which will explain the regression between .3 and .4, since the former will have allocated exactly the necessary number of bytes.)

lastly (and maybe not worth doing at all, depending on how soon we can get folks to use a 64-bit platform tools on Windows, since all the other platforms are already 64-bit-only), we could consider rewriting the adb sideload code to (a) mmap/munmap rather than actually read in to physical memory or (b) pread. i'm not sure what the performance impact of pread would be (especially on Windows where there is no pread equivalent).

Correction to #5: Windows does support pread() - it's in ReadFile()'s OVERLAPPED parameter.

@7: we'd still need to write the pread wrapper, though. (similar if we went with mmap/munmap.)

https://android-review.googlesource.com/#/c/355481/ pre-sizes the vector, which should be enough to undo the regression from '.3 to '.4...

https://android-review.googlesource.com/355818 switches "adb sideload" over to lseek/read. the time difference isn't measurable, but the memory requirements are then down to just 64KiB.

Hi I'm getting the same issue on Windows 10 when I try to sideload a TWRP backup.
- adb devices gives my device
- adb usb gives error: closed
- adb sideload 6.13.771.4_ckpv5.zip gives
loading: '6.13.771.4_ckpv5.zip'...
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.

Please can somebody help with this issue I've been trying to sort this for 3 days. Thanks.

I also get this problem. Is there a workaround?

I also get this error message when trying to update Android OS 7.1.1 from N6F26R to N6F26U security patch for Nexus 6. Please help. Thank you.

It happens when i try to flash any rom to my OnePlus 3, how can i solve it?

Trying to sideload via adb on Win10 gives me this as well. Nexus 6 trying to sideload from 7.0 to 7.1.1 (N6F26U, Mar 2017)

My Nexus 9 is on a bootloop and is locked so this was my way out of the bootloop but I've got the same error std::bad_alloc...
If someone has found a way to suppress this error let us know !

As this bug is in 25.0.4 only, simply revert to platform tools 25.0.3 to successfully sideload the image. This solved both my 'std::bad_alloc' error and bootloop error (hopefully).

You can find the earlier platform tools version here: https://dl.google.com/android/repository/platform-tools_r25.0.3-windows.zip

update: bootloop still looping, but will boot successfully every few boots, which is better than I had.

Same issue

Reverted to earlier platform-tools and it works like a charm. Thanks for the link.

Same problem, OnePlus 3, Open Beta 13 (http://downloads.oneplus.net/oneplus-3/oneplus_3_oxygenos_openbeta_13/), reverting to the platform-tools in #17 and using Large Address Aware 2.0.4 (https://www.techpowerup.com/forums/threads/large-address-aware.112556/) let me sideload with vanilla Oxygen Recovery (without LAA flag, it just says cannot read rom.zip)

full log:
C:\Android\SDK\platform-tools>adb sideload ../rom.zip
loading: '../rom.zip'...
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.

the same result by using latest adb version delivered by SDK Tools(Preview Channel) 26rc1.

use this tool to allocate more memory for your adb https://www.techpowerup.com/forums/threads/large-address-aware.112556/ 

IMP:
1. LAA your ADB
2. Reboot PC
3. sideload

Having same issue with Android SDK Platform tools 25.0.4 on Windows 10 with all latest updates. Was trying to install the April OTA patch on my Pixel.

same issue here with adb and pixel

Nikita gave the solution.

Download the LLA, go to Mode -> Advanced. Then, go to Add -> Files in folder recursive ... Select the folder where you have platform-tools installed (or unzipped). Select all the files in LLA then go to With selected -> Force Large Address Aware.

Tested on Win10, ran adb sideload <image> from cmd.exe with Administrative privilege. Installed 7.1.1 OTA on Nexus 6 without any issues.

same error for me flashing android 7.1.2 with latest adb version.

Same problem. Win10 x64 latest of win updates and adb. Not going to give you more info. Same as everyone else. This issue is on n9 and n6 with april updates. Solution from " #17 m...@ajmorley.com" about using " https://dl.google.com/android/repository/platform-tools_r25.0.3-windows.zip" works great.

Hey guys use the earlier tool https://dl.google.com/android/repository/platform-tools_r25.0.3-windows.zip 

Worked fine for me!! Thanks for solution bro!

Same issue on r25.0.4.
On old versions .3 and .1 I get cannot read "file".
Am I doing something wrong? It was a long time since I tried to sideload something.

Nexus 5X, I reflashed with ADB (25.0.3) but no luck! guess it's doomed

Following #16,#23 and #26's suggestion (sorry I cannot view their full name), it works for me.
win10 x64
platform-tools r25.0.3 (link at #16)
oneplus 3t Oxygen OS

Following #16,#23 and #26's suggestion (sorry I cannot view their full name), it works for me.
win10 x64
platform-tools r25.0.3 (link at #16)
oneplus 3t Oxygen OS

yes, using the previous version (available at https://dl.google.com/android/repository/platform-tools_r25.0.3-windows.zip) is the best workaround for now. 25.0.5 should be released at 10:00 America/Los_Angeles time on Monday, at which point adb.exe will only need 64KiB to sideload a package no matter how large it is.

25.0.5 was released this morning. available from https://developer.android.com/studio/releases/platform-tools.html now.

Still receiving this error on 25.0.5 with Nexus 6p 7.1.2 OTA image. angler-ota-n2g47h-f3bffaba.zip

@38: you're almost certainly not running the new adb.

deleted old platform-tools dir, downloaded 25.0.5, extracted, still receiving this error

output of "adb --version"?

c:\> adb --version
Android Debug Bridge version 1.0.39
Revision 3db08f2c6889-android

downloaded last night from:
https://developer.android.com/studio/releases/platform-tools.html

screenshot of the full error output?

screenshot attached

huh. given the "falling back to older sideload method" message, this kinda makes sense --- i didn't touch the old sideload codepath.

but it doesn't really make sense: you really shouldn't be on that codepath talking to a 6P. i think JellyBean was the last release that used the older sideload method! it looks like we'll give the old method a go regardless of how the new method fails. so my suspicion is that your real problem is that adb isn't able to talk to your 6P and falls back to the old method just in case (even though we know that won't work because the 6P is too new, adb doesn't).

the output from

adb kill-server
set ADB_TRACE=all
adb sideload <whatever you've been doing>

might be informative.

Hope this helps!

yeah, the "no devices/emulators found" is pretty much what i was expecting to see. what does "adb devices" say? i'm assuming that can't see your device either.

(have you ever sideloaded before? you know that you have to get the device ready to receive before "adb sideload" will work?)

I had no devices connected. Had to unplug and reconnect my phone. Then it worked. You should add that step to the official instructions here: https://developers.google.com/android/ota

C:\WINDOWS\system32>adb devices
List of devices attached
8XV7N15B24004979        device

Is that my device?

I have never sideloaded. I haven't messed with a rom since Samsung S3 days.
I would think that it sees my device since I can do the "adb reboot recovery"

@daniel: i've no idea whether that's your device :-) but what i can tell you is that if your device was actually in sideload mode, it would say "sideload" rather than "device". so that seems like your problem. see step 4 of the instructions at https://developers.google.com/android/ota.

@miguel: that's actually a 6P bug where it drops off USB when it reboots. i'm assuming that's fixed in O, but you're right that it might be worth adding an "adb devices" step to the instructions. i'll see what i can do.

i've uploaded https://android-review.googlesource.com/387532/ to improve the error messages. a future version of adb asked to sideload without any devices connected will instead say:

  $ adb sideload mysid-ota-424425.zip
  adb: sideload connection failed: no devices/emulators found
  adb: trying pre-KitKat sideload method...
  adb: pre-KitKat sideload connection failed: no devices/emulators found

i ran adb devices with phone just plugged in. previous time i ran the sideload i had followed step 4.
i unplugged and replugged in my phone and got a installing drivers screen. looks like it is working now???!!
I'm at the serving :~2% so far.

I can confirm I no longer experience bad_alloc with 25.0.5 on Windows 10 x64

Confirmed as well. Once i unplugged and replugged in my phone and got an installing drivers screen while in recovery mode.

I am still getting the crash on the most recent version with a 6P. How do I check the version of adb that I am running? I only downloaded it from Google today.

Follow up to #54, I also rebooted into recovery, selected update from adb, while on the screen waiting for the package, I unplugged and replugged in my phone, adb status changed to sideload, sent the package, loaded as usual.

adb --version will tell you what version you're using. if you include the screenshot of your error message (or just look for "  falling back to older sideload method..." and you're not trying to sideload to a pre-KitKat device) we can see if it's the same "adb can't see your device" problem as the other two folks. "adb devices" before "adb sideload" will show you whether (a) adb can see your device and (b) whether it's actually showing up in "sideload" mode.

+1 same issue on Windows 10 with Android Debug Bridge version 1.0.39, platform-tools 25.0.5, Pixel XL.

opening '.\marlin-ota-n2g47o-61bf55a3.zip'...
connecting...
falling back to older sideload method...
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

I tried unplugging and plugging back in once recovery was up, no change.

Update to #57, looks like I wasn't in sideload mode. Even with that, I think adb shouldn't crash but rather exit with an error message like  'no devices in sideload found' or something to that effect.

@57: that's what we've done for a future release. see my comment #50.

i've also updated the public instructions at https://developers.google.com/android/ota to recommend folks check that their device is actually in sideload mode.

Now unlock your device and confirm the backup operation...
PS C:\Users\wohen> adb reboot sideload "D:\download\Compressed\update.zip"
PS C:\Users\wohen> adb  sideload "D:\download\Compressed\update.zip"
loading: 'D:\download\Compressed\update.zip'...
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
...........................................................

Android lolipop 5.1.1

Lyf wind 7

All best  tq

Heloo semua

#santosrudy05

Hizmet

On 26 Sep 2017 00:52, <buganizer-system@google.com> wrote:

Brinton

Fucker

0633601601

Cheng my device

Hi

https://docs.google.com/spreadsheets/d/1ffwgqhvNrHuVKLEOTc10RrqkrsIs2zBEetZiVo8uyuo/edit?usp=drive_web (Expense report)clash of clans

https://drive.google.com/file/d/17NgW6N5DfIFhh2IFJtfsJrmA2qYkKR7f/view?usp=drive_web (1f449-1f3fc.svg)

Check out the one who's been trying to destroy people me and my children in particular  and hurt little kids...ka....@gmail  or LH@gmail. Now....Bo...at gmail. I'm Bo but the way the one who lost all three of my babys due to her and jr. Dunns decipt along with her little online friends she was my wife of 8 years moved me and the kids here after setting my up in my company by saying she'd do the books and I'd do the work not only did she do that ....talk about clash of clans...she molested a 14 year old girl.  Raped her with her boyfriend I never new anything about ...the state tool all my children and both KAYLA BOAK AND JEFFREY GABBERD ARE IN MANATEE COUNTY JAIL. FACING LIFE FOR KIDNAPING RAPE ECT....JUST LOOK UP THEIR NAMES kayla boak Bradenton florida.....They supposedly ran a land escaping company together called the landscape connection 9412509643. And kayla's was 9412414667....she made flyers for this man and at least gave him a shot to be a business man. All I have now is deep depression fro me and my children and have nothing to show for 8 years....it was most likely because she didn't but in the effort she just wanted to set me up watch me fail and get 3 kids out of me ....I'm in florida now doing everything I can to get my kids and go home I'll never stop fighting...all I can do now I'd even if its just 100 a day if I can feed my kids and there with me and not some sick freaks I will be more thankful than ever ....don't expect kayla and Lowe and jr to hang on to much longer unless there just plain hatefull....the stare of florida is offering them life in prison...he has like a 500,000 bond and hers is pathetically incredible too...check out Google and type I'm kayla boak bradenton itll pull up. And if you want the real details of these sick fucks please feel free to google manatee cleark of courts web page put in her first and last name it tells all....I'm so sorry for that little girl and her family I hope to god I never see them again....she has shamed our whole family so she could use drugs and cheet on a good husband....oh well I'm a great person and so are my babys someone will love is and be loyal one day. I wont loose hope or stop fighting. Anyhow tha I  you for the time ...if you do the research it's worth the read....ypu never really know someone I suppose. Please pray for me and my kids it's all i have to keep going foward. Tha is again . BO Edwards

Ps . Please ask jr Dunn and the person who kayla set me up years ago in 2015 to remove that stupid repository on me ..I'd be .I'm not a bad man at all I gave it my all with her but she lied and cheated us out of our heard work . I have nothing to show in my life but a rented bedroom and a broke down SUV and a car I bought my 16year old son I'm lonely broken down and hurt as are my kids. So if you have any kinda hearts please leave us in peace now. Me and my kids. We would be very thankful
..and JR of you still want to be with some that evil to hurt little 14 and 12 year old girls go for it. And be sure and bring your gun like you promised in your encrypted letter I decoded.....I wasnt scared when I drove to Texas for kayla and my front door is open every night . So if that's who you are so be it I'm scared of no man! I just hope your not that stupid in both manners frankly but do as ye will. And I'll be waiting.

Although the webcam now is functional, the performance after loading the camera is unbearable. Is there a fix in the works for this related to this fix?

I am forced to use other devices as the performance is so poor after the camera opens. Unfortunately it is not possible to use just macOS any longer to work on Android/Flutter projects because of this.

It's been a couple months so I am very hopeful I am just missing the fix?

Thank you for the hard work.

+1, performance are horrible.

Thanks for the link, as for fixing it myself, heh.........

I wish this was my area of expertise. Hopefully someone with the capability will work on this, it's extremely significant for Android Studio on macOS obviously.

Tested the most recent emulator version 34.2.5, no apparent changes in performance, still poor performance after opening camera.

Unfortunate.

+1, performance is horrible. after close camera, emulator still slow

Latest emulator build (35.1.20) still has the same webcam performance issue on Sonoma and above. Not only that but it also locks up the entire emulator itself (upon closing the camera app).
Downgrading to Ventura solves this problem, but it's not an ideal solution.