rkc@ looks like a cryptohome issue, can you route?

Is quite possibly ext4 related, especially given 53 -> 61 (with ext4 migration somewhere in between).
We had a few issues leading to "Fatal decryption error, but unable to remove cryptohome." caused by fs corruption or bugs. E.g.  issue 699389 . Or can be a variation of  issue 760007 . Iirc, they should be fixed by 61, but not sure. +gwendal / +hashimoto for expert opinion re ext4 and migration to it.
We also have  issue 772940  and issue 773816 with similar symptoms.


I don't see the logs at the link from the description. Could you please provide a link / attach them here?

Apologies for delay with response to #3
The debug log for Glimmer: https://drive.google.com/open?id=154YtjhZH2yz22mW9Gj7lnkGfn12QH9gn


But we have another customer (case # 15308655), having this issue. 
This customer is running Chrome OS 65.0.3325.209 on Viglen Chromebook 11 
User don't get any error and just dropped to login screen:

###
message.log:

2018-04-19T09:11:30.291498+01:00 WARNING cryptohomed[1159]: No valid keysets on disk for 0acae8229e1c946b87bf825bba87966dde5ce00e
2018-04-19T09:11:30.292058+01:00 ERR cryptohomed[1159]: Failed to decrypt any keysets for 0acae8229e1c946b87bf825bba87966dde5ce00e
2018-04-19T09:11:30.292592+01:00 ERR cryptohomed[1159]: Error, cryptohome must be re-created because of fatal error.
2018-04-19T09:11:31.110911+01:00 CRIT kernel: [  208.285067] EXT4-fs error (device mmcblk0p1): ext4_xattr_block_get:332: inode #390446: comm MountThread: bad block 1581732
2018-04-19T09:11:31.110947+01:00 CRIT kernel: [  208.286456] EXT4-fs error (device mmcblk0p1): ext4_xattr_block_get:332: inode #390446: comm MountThread: bad block 1581732
2018-04-19T09:11:31.111592+01:00 ERR cryptohomed[1159]: Fatal decryption error, but unable to remove cryptohome.
2018-04-19T09:11:31.112017+01:00 WARNING cryptohomed[1159]: PKCS#11 initialization requested but cryptohome is not mounted.

chrome.log:

[1147:1147:0419/091130.206295:VERBOSE1:login_performer.cc(288)] Online login completion started.
[1147:1147:0419/091130.206978:VERBOSE1:cryptohome_authenticator.cc(776)] Resolved state to: 0
[1147:1147:0419/091131.112885:ERROR:device_event_log_impl.cc(156)] [09:11:31.112] Login: homedir_methods.cc:274 HomedirMethods MountEx error (CryptohomeErrorCode): 5
[1147:1147:0419/091131.113100:ERROR:device_event_log_impl.cc(156)] [09:11:31.113] Login: cryptohome_authenticator.cc:932 Cryptohome failure: state(AuthState)=2, code(cryptohome::MountError)=1
[1147:1147:0419/091131.113215:VERBOSE1:cryptohome_authenticator.cc(776)] Resolved state to: 2
[1147:1147:0419/091131.113413:ERROR:device_event_log_impl.cc(156)] [09:11:31.113] Login: cryptohome_authenticator.cc:710 Login failed: Could not mount cryptohome.
[1147:1147:0419/091131.113588:ERROR:login_performer.cc(63)] Login failure, reason=1, error.state=0
[1147:1147:0419/091131.113868:VERBOSE1:existing_user_controller.cc(1421)] Could not mount cryptohome.
[1147:1147:0419/091131.113987:VERBOSE1:webui_login_display.cc(109)] Show error, error_id: 1989, attempts:1, help_topic_id: 188036
[1147:1147:0419/091131.114104:ERROR:core_oobe_handler.cc(195)] CoreOobeHandler::ShowSignInError: error_text=Sorry, your email or password could not be verified. Please try again.
.......
[1147:1147:0419/091131.366959:INFO:signin_screen_handler.cc(1412)] Login WebUI >> active: 0, source: gaia-signin
[1147:1147:0419/091131.367196:INFO:signin_screen_handler.cc(1412)] Login WebUI >> active: 1, source: gaia-signin

###

Full debug log: https://drive.google.com/open?id=1QgrRkEwa9FbSRFQ80GwL3mqzogU5CV4V
 

Same for peppy for M-66 (66.0.3359.158)
Case: 15768607 
Storage: KINGSTON SNS4151S316GD

...
2018-05-09T20:06:41.096389+00:00 NOTICE kernel: [  101.332694] Sense Key : Medium Error [current] [descriptor]
2018-05-09T20:06:41.096390+00:00 NOTICE kernel: [  101.332713] Descriptor sense data with sense descriptors (in hex):
2018-05-09T20:06:41.096392+00:00 INFO kernel: [  101.332725]         72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00 
2018-05-09T20:06:41.096402+00:00 INFO kernel: [  101.332783]         01 44 5f 38 
2018-05-09T20:06:41.096403+00:00 INFO kernel: [  101.332805] sd 0:0:0:0: [sda]  
2018-05-09T20:06:41.096404+00:00 NOTICE kernel: [  101.332813] Add. Sense: Unrecovered read error - auto reallocate failed
2018-05-09T20:06:41.096405+00:00 INFO kernel: [  101.332826] sd 0:0:0:0: [sda] CDB: 
2018-05-09T20:06:41.096406+00:00 NOTICE kernel: [  101.332833] Read(10): 28 00 01 44 5f 38 00 00 08 00
2018-05-09T20:06:41.096407+00:00 INFO kernel: [  101.332892] ata1: EH complete
2018-05-09T20:06:41.110422+00:00 ERR kernel: [  101.347348] ata1.00: exception Emask 0x0 SAct 0x7 SErr 0x0 action 0x0
2018-05-09T20:06:41.110435+00:00 ERR kernel: [  101.347362] ata1.00: irq_stat 0x40000008
2018-05-09T20:06:41.110436+00:00 ERR kernel: [  101.347372] ata1.00: failed command: READ FPDMA QUEUED
2018-05-09T20:06:41.110437+00:00 ERR kernel: [  101.347386] ata1.00: cmd 60/08:00:60:5f:44/00:00:01:00:00/40 tag 0 ncq 4096 in
2018-05-09T20:06:41.110438+00:00 ERR kernel: [  101.347386]          res 41/40:08:60:5f:44/00:00:01:00:00/00 Emask 0x409 (media error) <F>
2018-05-09T20:06:41.110440+00:00 ERR kernel: [  101.347406] ata1.00: status: { DRDY ERR }
2018-05-09T20:06:41.110468+00:00 ERR kernel: [  101.347414] ata1.00: error: { UNC }
2018-05-09T20:06:41.111802+00:00 ERR cryptohomed[1339]: Fatal decryption error, but unable to remove cryptohome.
...

https://drive.google.com/corp/drive/folders/15Gmx2m2YrHXd1qvKe8Lzjzi-VubKVX5M

Hi team, we received cases related to this issue. 2 cases were resolved by Chrome OS Recovery. 1 case got resolved by deprovisioning the device > Wipe > Login to Chromebook then perform Powerwash.
Customer stated that wiping doesn't resolve the issue but Powerwash does.

I would like to confirm, does the policy set to each device is the one who's causing the issue? Not sure about it but any insight will be very much appreciated. Thank you!

[rephrased deleted #9]
Re #8: What device models the issues are reported for? What are the symptoms - same as here?
Here we have inability to sign-in for a single user with "Fatal decryption error, but unable to remove cryptohome" in the feedback logs, while other users login ok on that chromebook.

Gwendal, can you please take a look what's wrong with ext4 in those cases?

#9 - an issue with a specific user that couldn't login to the affected device (while other users can). And when this specific user will login to a different device, he/she can login successful. When we say unable to login, the login page just loops.

Viglen Chromebook 11 - 65.0.3325.209 (case from comment #5)
ACER C720 - 66.0.3359.158 
Acer Chromebook 11 - 66.0.3359.181

Do we have logs for 
ACER C720 - 66.0.3359.158 
Acer Chromebook 11 - 66.0.3359.181?

Based on the log snippets I see in #5 and #6: the issue is that the user directory on this specific machine is not mountable (and user won't be able to sign in in that case), and on top of that we fail to delete it to recreate from scratch. In both cases there are some underlying filesystem issues. 

That issue doesn't affect other machines, so the user can sign in there. That issue also affect other users on that machine who have separate directories.

Going through powerwash recreates the whole partition where user directories are stored, and the issue goes away.

The middle paragraph from #14 should have been (not the added "doesn't" for the 2nd sentence):

That issue doesn't affect other machines, so the user can sign in there. That issue also doesn't affect other users on that machine, who have separate directories.

Sorry for the typos :/

This bug has an owner, thus, it's been triaged. Changing status to "assigned".

Hello!
This bug is receiving this notice because there has been no acknowledgment of its existence in quite a bit of time
- If you are currently working on this bug, please provide an update.
- If you are currently affected by this bug, please update with your current symptoms and relevant logs.

If there has been no updates provided by EOD Wednesday, 12/19/18 (5pm EST), this bug will be archived and can be re-opened at any time deemed necessary.

Thank you!

Due to lack of action this bug has been Archived. If work is still being done on this issue or you are still experiencing this issue please feel free to re-open with the appropriate information.