New: kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c: xlog_recover_do_reg_buffer() when mount a corrupted image
https://bugzilla.kernel.org/show_bug.cgi?id=215783
Bug ID: 215783
Summary: kernel NULL pointer dereference and general protection
fault in fs/xfs/xfs_buf_item_recover.c:
xlog_recover_do_reg_buffer() when mount a corrupted image
Product: File System
Version: 2.5
Kernel Version: 5.17.1, 5.15.32
Hardware: All
OS: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: XFS
Assignee: [email protected]
Reporter: [email protected]
Regression: No
Created attachment 300671
--> https://bugzilla.kernel.org/attachment.cgi?id=300671&action=edit
corrupted image and .config
- Overview
kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c:xlog_recover_do_reg_buffer() when mount a corrupted image, sometimes cause kernel hang
- Reproduce
tested on kernel 5.17.1, 5.15.32
$ mkdir mnt $ unzip tmp7.zip $ ./mount.sh xfs 7 ##NULL pointer derefence or $ sudo mount -t xfs tmp7.img mnt ##general protection fault
- Kernel dump
##1
[ 433.401124] loop0: detected capacity change from 0 to 32768 [ 433.416943] XFS (loop0): Deprecated V4 format (crc=0) will not be supported after September 2030. [ 433.417005] XFS (loop0): Mounting V10 Filesystem [ 433.418206] XFS (loop0): Starting recovery (logdev: internal) [ 433.418261] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ 433.418288] #PF: supervisor read access in kernel mode [ 433.418299] #PF: error_code(0x0000) - not-present page [ 433.418310] PGD 0 P4D 0 [ 433.418317] Oops: 0000 [#1] PREEMPT SMP NOPTI [ 433.418327] CPU: 0 PID: 977 Comm: mount Not tainted 5.17.1 #1 [ 433.418339] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-1ubuntu1.1 04/01/2014 [ 433.418356] RIP: 0010:memcpy_erms+0x6/0x10 [ 433.418368] Code: fe ff ff cc eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe [ 433.418401] RSP: 0018:ffffa985c0217990 EFLAGS: 00010287 [ 433.418412] RAX: ffff97c382232400 RBX: 0000000000000000 RCX: ffffffff913f5d00 [ 433.418425] RDX: ffffffff913f5d00 RSI: 0000000000000000 RDI: ffff97c382232400 [ 433.418439] RBP: ffff97c391472b20 R08: ffff97c391472b34 R09: 0000000000000000 [ 433.418452] R10: 0000000000000010 R11: fefefefefefefeff R12: 0000000000000008 [ 433.418465] R13: ffff97c391472b34 R14: ffff97c38ff1b958 R15: 00000000ff227eba [ 433.418479] FS: 00007f7490a8d080(0000) GS:ffff97c575c00000(0000) knlGS:0000000000000000 [ 433.418494] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 433.418505] CR2: 0000000000000000 CR3: 0000000105e2c004 CR4: 0000000000370ef0 [ 433.418521] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 433.418535] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 433.418573] Call Trace: [ 433.418580] <TASK> [ 433.418587] xlog_recover_do_reg_buffer.isra.11+0xf0/0x1e0 [xfs] [ 433.418768] xlog_recover_buf_commit_pass2+0x476/0x650 [xfs] [ 433.418844] xlog_recover_items_pass2+0x4e/0xc0 [xfs] [ 433.418921] xlog_recover_commit_trans+0x2de/0x300 [xfs] [ 433.418996] xlog_recovery_process_trans+0x8e/0xc0 [xfs] [ 433.419070] xlog_recover_process_data+0xab/0x130 [xfs] [ 433.419144] xlog_do_recovery_pass+0x2d5/0x5c0 [xfs] [ 433.419250] xlog_do_log_recovery+0x7f/0xb0 [xfs] [ 433.419323] xlog_do_recover+0x34/0x190 [xfs] [ 433.419402] xlog_recover+0xbc/0x170 [xfs] [ 433.419477] xfs_log_mount+0x125/0x2d0 [xfs] [ 433.419549] xfs_mountfs+0x4e0/0xa50 [xfs] [ 433.419621] ? kmem_alloc+0x88/0x140 [xfs] [ 433.419691] ? xfs_filestream_get_parent+0x70/0x70 [xfs] [ 433.419759] xfs_fs_fill_super+0x69f/0x880 [xfs] [ 433.419829] ? sget_fc+0x1be/0x230 [ 433.419839] ? xfs_fs_inode_init_once+0x70/0x70 [xfs] [ 433.419911] get_tree_bdev+0x16a/0x280 [ 433.419920] vfs_get_tree+0x22/0xc0 [ 433.419929] path_mount+0x59b/0x9a0 [ 433.419938] do_mount+0x75/0x90 [ 433.419946] __x64_sys_mount+0x86/0xd0 [ 433.420302] do_syscall_64+0x37/0xb0 [ 433.420634] entry_SYSCALL_64_after_hwframe+0x44/0xae [ 433.420966] RIP: 0033:0x7f749034e15a [ 433.421290] Code: 48 8b 0d 31 8d 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fe 8c 2c 00 f7 d8 64 89 01 48 [ 433.421964] RSP: 002b:00007fff34b29268 EFLAGS: 00000202 ORIG_RAX: 00000000000000a5 [ 433.422310] RAX: ffffffffffffffda RBX: 000055de17f34420 RCX: 00007f749034e15a [ 433.422676] RDX: 000055de17f34600 RSI: 000055de17f36320 RDI: 000055de17f3dc40 [ 433.423024] RBP: 0000000000000000 R08: 0000000000000000 R09: 000055de17f34620 [ 433.423377] R10: 00000000c0ed0000 R11: 0000000000000202 R12: 000055de17f3dc40 [ 433.423713] R13: 000055de17f34600 R14: 0000000000000000 R15: 00007f749086f8a4 [ 433.424043] </TASK> [ 433.424359] Modules linked in: input_leds serio_raw joydev iscsi_tcp
libiscsi_tcp libiscsi scsi_transport_iscsi qemu_fw_cfg xfs autofs4 raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor hid_generic async_tx raid1 usbhid raid0 multipath linear hid qxl drm_ttm_helper ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm crct10dif_pclmul crc32_pclmul ghash_clmulni_intel psmouse aesni_intel crypto_simd cryptd
[ 433.425796] CR2: 0000000000000000 [ 433.426147] ---[ end trace 0000000000000000 ]--- [ 433.426497] RIP: 0010:memcpy_erms+0x6/0x10 [ 433.426856] Code: fe ff ff cc eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe [ 433.427595] RSP: 0018:ffffa985c0217990 EFLAGS: 00010287 [ 433.427960] RAX: ffff97c382232400 RBX: 0000000000000000 RCX: ffffffff913f5d00 [ 433.428358] RDX: ffffffff913f5d00 RSI: 0000000000000000 RDI: ffff97c382232400 [ 433.428739] RBP: ffff97c391472b20 R08: ffff97c391472b34 R09: 0000000000000000 [ 433.429125] R10: 0000000000000010 R11: fefefefefefefeff R12: 0000000000000008 [ 433.429510] R13: ffff97c391472b34 R14: ffff97c38ff1b958 R15: 00000000ff227eba [ 433.429897] FS: 00007f7490a8d080(0000) GS:ffff97c575c00000(0000) knlGS:0000000000000000 [ 433.430291] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 433.430708] CR2: 0000000000000000 CR3: 0000000105e2c004 CR4: 0000000000370ef0 [ 433.431110] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 433.431507] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 ======================================================== ##2 [ 35.767636] loop0: detected capacity change from 0 to 32768 [ 35.782283] XFS (loop0): Deprecated V4 format (crc=0) will not be supported after September 2030. [ 35.782334] XFS (loop0): Mounting V10 Filesystem [ 35.783545] XFS (loop0): Starting recovery (logdev: internal) [ 35.783660] general protection fault, maybe for address 0xffff9c0c8b530600: 0000 [#1] SMP NOPTI [ 35.783699] CPU: 3 PID: 943 Comm: mount Not tainted 5.15.32 #1 [ 35.783735] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-1ubuntu1.1 04/01/2014 [ 35.783764] RIP: 0010:memcpy_erms+0x6/0x10 [ 35.783783] Code: cc cc cc cc eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe [ 35.783845] RSP: 0018:ffffbb6f0095f990 EFLAGS: 00010287 [ 35.783866] RAX: ffff9c0c8b530600 RBX: 0000000000000000 RCX: ffffffff914fb600 [ 35.783891] RDX: ffffffff914fb600 RSI: 30a131db6e24bf76 RDI: ffff9c0c8b530600 [ 35.783916] RBP: ffff9c0c83c8b6e0 R08: ffff9c0c83c8b6f4 R09: 30a131db6e24bf76 [ 35.783941] R10: ffffbb6f0095f858 R11: 0000000000000000 R12: 0000000000000008 [ 35.783964] R13: ffff9c0c83c8b6f4 R14: ffff9c0c91095118 R15: 00000000ff229f6c [ 35.783989] FS: 00007fd72fc59080(0000) GS:ffff9c0e75d80000(0000) knlGS:0000000000000000 [ 35.784017] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.784038] CR2: 000055e0eb070ee8 CR3: 0000000111f0c004 CR4: 0000000000370ee0 [ 35.784067] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 35.784093] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 35.784118] Call Trace: [ 35.784130] <TASK> [ 35.784142] xlog_recover_do_reg_buffer.isra.11+0xf0/0x1d0 [xfs] [ 35.784291] xlog_recover_buf_commit_pass2+0x449/0x5f0 [xfs] [ 35.784393] xlog_recover_items_pass2+0x4a/0xa0 [xfs] [ 35.784490] xlog_recover_commit_trans+0x2c3/0x2e0 [xfs] [ 35.784590] xlog_recovery_process_trans+0x8e/0xc0 [xfs] [ 35.784689] xlog_recover_process_data+0xa3/0x110 [xfs] [ 35.784804] xlog_do_recovery_pass+0x2d5/0x5c0 [xfs] [ 35.784909] xlog_do_log_recovery+0x7f/0xb0 [xfs] [ 35.785011] xlog_do_recover+0x34/0x180 [xfs] [ 35.785129] xlog_recover+0xbc/0x170 [xfs] [ 35.785229] xfs_log_mount+0x125/0x2d0 [xfs] [ 35.785331] xfs_mountfs+0x4ad/0xa50 [xfs] [ 35.785431] ? kmem_alloc+0x74/0x110 [xfs] [ 35.785530] ? xfs_filestream_get_parent+0x70/0x70 [xfs] [ 35.785631] xfs_fs_fill_super+0x6a4/0x890 [xfs] [ 35.785733] ? sget_fc+0x1bd/0x230 [ 35.785749] ? set_bdev_super+0x90/0x90 [ 35.785766] ? xfs_fs_statfs+0x200/0x200 [xfs] [ 35.785869] get_tree_bdev+0x16a/0x280 [ 35.785872] vfs_get_tree+0x22/0xc0 [ 35.785901] path_mount+0x5a0/0x9a0 [ 35.785918] do_mount+0x75/0x90 [ 35.785932] __x64_sys_mount+0x86/0xd0 [ 35.785949] do_syscall_64+0x37/0xb0 [ 35.785966] entry_SYSCALL_64_after_hwframe+0x44/0xae [ 35.785987] RIP: 0033:0x7fd72f51a15a [ 35.786657] Code: 48 8b 0d 31 8d 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 49 89 ca b8 a5 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fe 8c 2c 00 f7 d8 64 89 01 48 [ 35.788084] RSP: 002b:00007ffe8d0e64b8 EFLAGS: 00000206 ORIG_RAX: 00000000000000a5 [ 35.788808] RAX: ffffffffffffffda RBX: 000055e0eb063420 RCX: 00007fd72f51a15a [ 35.789532] RDX: 000055e0eb063600 RSI: 000055e0eb065320 RDI: 000055e0eb06cc40 [ 35.790220] RBP: 0000000000000000 R08: 0000000000000000 R09: 000055e0eb063620 [ 35.790906] R10: 00000000c0ed0000 R11: 0000000000000206 R12: 000055e0eb06cc40 [ 35.791562] R13: 000055e0eb063600 R14: 0000000000000000 R15: 00007fd72fa3b8a4 [ 35.792208] </TASK> [ 35.792828] Modules linked in: iscsi_tcp libiscsi_tcp libiscsi
scsi_transport_iscsi joydev input_leds serio_raw qemu_fw_cfg xfs autofs4 raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx raid1 raid0 multipath linear qxl drm_ttm_helper ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm psmouse hid_generic crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel usbhid crypto_simd hid cryptd
[ 35.795726] ---[ end trace 876aa8f34409ae4f ]--- [ 35.796447] RIP: 0010:memcpy_erms+0x6/0x10 [ 35.797153] Code: cc cc cc cc eb 1e 0f 1f 00 48 89 f8 48 89 d1 48 c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48 89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38 fe [ 35.798627] RSP: 0018:ffffbb6f0095f990 EFLAGS: 00010287 [ 35.799357] RAX: ffff9c0c8b530600 RBX: 0000000000000000 RCX: ffffffff914fb600 [ 35.800098] RDX: ffffffff914fb600 RSI: 30a131db6e24bf76 RDI: ffff9c0c8b530600 [ 35.800843] RBP: ffff9c0c83c8b6e0 R08: ffff9c0c83c8b6f4 R09: 30a131db6e24bf76 [ 35.801614] R10: ffffbb6f0095f858 R11: 0000000000000000 R12: 0000000000000008 [ 35.802395] R13: ffff9c0c83c8b6f4 R14: ffff9c0c91095118 R15: 00000000ff229f6c [ 35.803158] FS: 00007fd72fc59080(0000) GS:ffff9c0e75d80000(0000) knlGS:0000000000000000 [ 35.803950] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.804759] CR2: 000055e0eb070ee8 CR3: 0000000111f0c004 CR4: 0000000000370ee0 [ 35.805568] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 35.806404] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
Signature · 4 lines
You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
Quoted 9 lines · On Thu, Mar 31, 2022 at 08:07:08PM +0000, [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=215783
- Overview
kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c:xlog_recover_do_reg_buffer() when mount a corrupted image, sometimes cause kernel hang- Reproduce
tested on kernel 5.17.1, 5.15.32
Quoted 7 lines
$ mkdir mnt $ unzip tmp7.zip $ ./mount.sh xfs 7 ##NULL pointer derefence or $ sudo mount -t xfs tmp7.img mnt ##general protection fault- Kernel dump
You've now raised 4 bugs that all look very similar and are quite possibly all caused by the same corruption vector. Please do some triage on the failure to identify the source of the corruption that trigger this failure.
If you are going to run some scripted tool to randomly corrupt the filesystem to find failures, then you have an ethical and moral responsibility to do some of the work to narrow down and identify the cause of the failure, not just throw them at someone to do all the work.
You can automate this - track the corruptions you add to the filesystem image, then when you have an image that reproduces a problem, iterate over it removing corruptions until you have just the minimum set of changes in the image that reproduce the issue. Then you can cull all the images that trip over the same corruptions and only report the actual corruption that causes the problem.
Then list those corruptions in the bug report so that we don't have to do all this triage ourselves to weed out all the duplicates and noise that all the random corruptions that don't cause crashes induce.
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
[Bug 215783] kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c: xlog_recover_do_reg_buffer() when mount a corrupted image
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=215783
--- Comment #1 from Dave Chinner ([email protected]) ---
Quoted 9 lines · On Thu, Mar 31, 2022 at 08:07:08PM +0000, [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=215783
- Overview
kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c:xlog_recover_do_reg_buffer() when mount a corrupted image, sometimes cause kernel hang- Reproduce
tested on kernel 5.17.1, 5.15.32
Quoted 7 lines
$ mkdir mnt $ unzip tmp7.zip $ ./mount.sh xfs 7 ##NULL pointer derefence or $ sudo mount -t xfs tmp7.img mnt ##general protection fault- Kernel dump
You've now raised 4 bugs that all look very similar and are quite possibly all caused by the same corruption vector. Please do some triage on the failure to identify the source of the corruption that trigger this failure.
If you are going to run some scripted tool to randomly corrupt the filesystem to find failures, then you have an ethical and moral responsibility to do some of the work to narrow down and identify the cause of the failure, not just throw them at someone to do all the work.
You can automate this - track the corruptions you add to the filesystem image, then when you have an image that reproduces a problem, iterate over it removing corruptions until you have just the minimum set of changes in the image that reproduce the issue. Then you can cull all the images that trip over the same corruptions and only report the actual corruption that causes the problem.
Then list those corruptions in the bug report so that we don't have to do all this triage ourselves to weed out all the duplicates and noise that all the random corruptions that don't cause crashes induce.
-Dave.
Signature · 4 lines
You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
Quoted 16 lines · Dave Chinner
Quoted 15 lines · On Thu, Mar 31, 2022 at 08:07:08PM +0000, [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=215783
- Overview
kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c:xlog_recover_do_reg_buffer() when mount a corrupted image, sometimes cause kernel hang- Reproduce
tested on kernel 5.17.1, 5.15.32$ mkdir mnt $ unzip tmp7.zip $ ./mount.sh xfs 7 ##NULL pointer derefence or $ sudo mount -t xfs tmp7.img mnt ##general protection fault
Quoted 6 lines
Quoted 1 line
- Kernel dump
You've now raised 4 bugs that all look very similar and are quite possibly all caused by the same corruption vector. Please do some triage on the failure to identify the source of the corruption that trigger this failure.
Ok, the log has been intentionally corrupted in a way that does not happen in the real world. i.e. The iclog header at the tail of the log has had the CRC zeroed, so CRC checking for media bit corruption has been intentionally bypassed by the tool that corrupted the log.
The first item is a superblock buffer item, which contains 2 regions; a buf log item and a 384 byte long region containing the logged superblock data.
However, the buf log item has been screwed with to say that it has 8 regions rather than 2, and so when recovery goes to recovery the third region that doesn't exist, it falls off the end of the allocated transaction buffer.
We only ever write iclogs with CRCs in them (except for mkfs when it writes an unmount record to intialise the log), so bit corruptions like this will get caught before we even started log recovery in production systems.
We've got enough issues with actual log recovery bugs that we don't need to be overloaded by being forced to play whack-a-mole with malicious corruptions that *will not happen in the real world* because "security!".
Looking at the crash locations for the other bugs, they are all going to be the same thing - you've corrupted the vector index in the log item and so they all fall off the end of the buffer because the index no longer matches the actual contents of the log item.
vvvv THIS vvvv
Quoted 5 lines
If you are going to run some scripted tool to randomly corrupt the filesystem to find failures, then you have an ethical and moral responsibility to do some of the work to narrow down and identify the cause of the failure, not just throw them at someone to do all the work.
^^^^ THIS ^^^^^
Please confirm your other reports have the same root cause and close them if they are. If not, please point us to the unique corruption in the log that causes the failure.
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
[Bug 215783] kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c: xlog_recover_do_reg_buffer() when mount a corrupted image
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=215783
--- Comment #2 from Dave Chinner ([email protected]) ---
Quoted 16 lines · Dave Chinner
Quoted 15 lines · On Thu, Mar 31, 2022 at 08:07:08PM +0000, [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=215783
- Overview
kernel NULL pointer dereference and general protection fault in fs/xfs/xfs_buf_item_recover.c:xlog_recover_do_reg_buffer() when mount a corrupted image, sometimes cause kernel hang- Reproduce
tested on kernel 5.17.1, 5.15.32$ mkdir mnt $ unzip tmp7.zip $ ./mount.sh xfs 7 ##NULL pointer derefence or $ sudo mount -t xfs tmp7.img mnt ##general protection fault
Quoted 6 lines
Quoted 1 line
- Kernel dump
You've now raised 4 bugs that all look very similar and are quite possibly all caused by the same corruption vector. Please do some triage on the failure to identify the source of the corruption that trigger this failure.
Ok, the log has been intentionally corrupted in a way that does not happen in the real world. i.e. The iclog header at the tail of the log has had the CRC zeroed, so CRC checking for media bit corruption has been intentionally bypassed by the tool that corrupted the log.
The first item is a superblock buffer item, which contains 2 regions; a buf log item and a 384 byte long region containing the logged superblock data.
However, the buf log item has been screwed with to say that it has 8 regions rather than 2, and so when recovery goes to recovery the third region that doesn't exist, it falls off the end of the allocated transaction buffer.
We only ever write iclogs with CRCs in them (except for mkfs when it writes an unmount record to intialise the log), so bit corruptions like this will get caught before we even started log recovery in production systems.
We've got enough issues with actual log recovery bugs that we don't need to be overloaded by being forced to play whack-a-mole with malicious corruptions that *will not happen in the real world* because "security!".
Looking at the crash locations for the other bugs, they are all going to be the same thing - you've corrupted the vector index in the log item and so they all fall off the end of the buffer because the index no longer matches the actual contents of the log item.
vvvv THIS vvvv
Quoted 5 lines
If you are going to run some scripted tool to randomly corrupt the filesystem to find failures, then you have an ethical and moral responsibility to do some of the work to narrow down and identify the cause of the failure, not just throw them at someone to do all the work.
^^^^ THIS ^^^^^
Please confirm your other reports have the same root cause and close them if they are. If not, please point us to the unique corruption in the log that causes the failure.
-Dave.
Signature · 4 lines
You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.