ecryptfs not compatible ext4 filesystem

Bug #1763655 reported by YuYu on 2018-04-13
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
eCryptfs
Undecided
Unassigned

Bug Description

System: Cetnos 6.9

FileSystem: ext4

Kernel: 2.6.32-696.16.1.el6.x86_64

Ecryptfs: ecryptfs-utils-82-6.el6_1.3.x86_64

Execution:
mount.ecryptfs /data/game /data/game

After the mount, execute a command ( cp pip-7.1.2 /data/game/ ), overlay the existing files, kernel will prompt the error

Problem display:
Nov 22 04:15:13 ip-192-168-1-22 kernel: ------------[ cut here ]------------

Nov 22 04:15:13 ip-192-168-1-22 kernel: WARNING: at fs/ext4/inode.c:3969 ext4_flush_unwritten_io+0x74/0x80 [ext4]() (Tainted: G -- ------------ T)

Nov 22 04:15:13 ip-192-168-1-22 kernel: Hardware name: HVM domU

Nov 22 04:15:13 ip-192-168-1-22 kernel: Modules linked in: cbc xt_state xt_multiport iptable_filter ip_tables ecb drbg ansi_cprng nf_conntrack_ipv4 nf_conntrack nf_defrag_ipv4 ecryptfs(T) dm_crypt aesni_intel ablk_helper cryptd lrw gf128mul glue_helper aes_x86_64 aes_generic xen_netfront i2c_piix4 i2c_core ext4 jbd2 mbcache xen_blkfront pata_acpi ata_generic ata_piix dm_mirror dm_region_hash dm_log dm_mod [last unloaded: scsi_wait_scan]

Nov 22 04:15:13 ip-192-168-1-22 kernel: Pid: 14723, comm: cp Tainted: G -- ------------ T 2.6.32-696.3.2.el6.x86_64 #1

Nov 22 04:15:13 ip-192-168-1-22 kernel: Call Trace:

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff8107c811>] ? warn_slowpath_common+0x91/0xe0

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff8107c87a>] ? warn_slowpath_null+0x1a/0x20

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa0087d54>] ? ext4_flush_unwritten_io+0x74/0x80 [ext4]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa00afb27>] ? ext4_ext_truncate+0x37/0x1f0 [ext4]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff811468b9>] ? truncate_inode_pages_range+0x209/0x500

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa008df38>] ? ext4_truncate+0x4c8/0x6a0 [ext4]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa0168671>] ? ecryptfs_write_lower+0xa1/0xc0 [ecryptfs]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81158732>] ? unmap_mapping_range+0x72/0x140

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81146cb1>] ? truncate_pagecache+0x61/0x70

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81146d76>] ? vmtruncate+0x56/0x70

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa0165087>] ? ecryptfs_truncate+0x297/0x2f0 [ecryptfs]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff811a8990>] ? path_to_nameidata+0x20/0x60

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa01652a6>] ? ecryptfs_setattr+0x1c6/0x220 [ecryptfs]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff811b9438>] ? notify_change+0x168/0x340

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffffa01643eb>] ? ecryptfs_open+0x11b/0x280 [ecryptfs]

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81198004>] ? do_truncate+0x64/0xa0

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff8123bf8f>] ? security_inode_permission+0x1f/0x30

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff811ad301>] ? do_filp_open+0x861/0xd20

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff8119f964>] ? cp_new_stat+0xe4/0x100

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff812a8cda>] ? strncpy_from_user+0x4a/0x90

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff811ba312>] ? alloc_fd+0x92/0x160

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81196cb7>] ? do_sys_open+0x67/0x130

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff81196dc0>] ? sys_open+0x20/0x30

Nov 22 04:15:13 ip-192-168-1-22 kernel: [<ffffffff8100b0d2>] ? system_call_fastpath+0x16/0x1b

Nov 22 04:15:13 ip-192-168-1-22 kernel: ---[ end trace 3e7def58e59cc951 ]---

Tyler Hicks (tyhicks) wrote :

Hi YuYu - Thanks for the bug report! AFAIK, this is not a bug in the latest upstream linux kernel but please let me know if you are able to trigger this bug there.

I don't recall fixing this issue upstream so I'm wondering if it is due to changes to ext4 and/or ecryptfs in the CentOS kernel (it is quite an old version). I wish I had an upstream commit that I could point you to but I don't know of one off the top of my head.

I'm going to set this bug to invalid since this is the bug tracker for the upstream eCryptfs project and your bug is in the CentOS kernel. If you're able to reproduce the issue in a newer upstream kernel, please reopen this bug.

Changed in ecryptfs:
status: New → Invalid
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers