Hello Ceph folks, when running latest Stable 64 Bit Google Chrome under Linux with home dirs on cephfs, I get the attached kernel trace and cephfs access hangs afterwards. Umounting cephfs no longer works, only a hard system reset helps. Since it only happens on cephfs, I hope to get some help here. Kernel versions tested are 4.9.67 and 4.9.68, each with RSBAC patched in. However, RSBAC does not touch these funktions. Since 4.9.68 just had a fix related to page_mkwrite(), it might be related. The trace can be reproduced with every start of Chrome, but we have not seen anything similar with other applications, e.g. Firefox. Ceph backend lives on xfs or btrfs on the same system, crash happens with both. Thanks for all your good work, Amon Ott -- Dr. Amon Ott m-privacy GmbH Tel: +49 30 24342334 Werner-Voß-Damm 62 Fax: +49 30 99296856 12101 Berlin http://www.m-privacy.de Amtsgericht Charlottenburg, HRB 84946 Geschäftsführer: Dipl.-Kfm. Holger Maczkowsky, Roman Maczkowsky GnuPG-Key-ID: 0x2DD3A649
Dec 11 11:54:47 tgkvm14 0000000653 kernel: BUG: unable to handle kernel NULL pointer dereference at (null) Dec 11 11:54:47 tgkvm14 0000000654 kernel: IP: [<ffffffff812aa5a2>] jbd2__journal_start+0x38/0x1a2 Dec 11 11:54:47 tgkvm14 0000000655 kernel: PGD 0 Dec 11 11:54:47 tgkvm14 0000000656 kernel: Dec 11 11:54:47 tgkvm14 0000000657 kernel: Oops: 0000 [#1] PREEMPT SMP Dec 11 11:54:47 tgkvm14 0000000658 kernel: Modules linked in: cbc ceph libceph xfs libcrc32c 8139cp mii intel_agp intel_gtt parport_pc i2c_piix4 agpgart floppy button evdev parport rtc_cmos i2c_core virtio_net virtio_balloon Dec 11 11:54:47 tgkvm14 0000000659 kernel: CPU: 1 PID: 6661 Comm: TaskSchedulerBa Not tainted 4.9.68-rsbac64 #1 Dec 11 11:54:47 tgkvm14 0000000660 kernel: Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 Dec 11 11:54:47 tgkvm14 0000000661 kernel: task: ffff95c5f3790b80 task.stack: ffff9ad7468a0000 Dec 11 11:54:47 tgkvm14 0000000662 kernel: RIP: 0010:[<ffffffff812aa5a2>] [<ffffffff812aa5a2>] jbd2__journal_start+0x38/0x1a2 Dec 11 11:54:47 tgkvm14 0000000663 kernel: RSP: 0018:ffff9ad7468a3900 EFLAGS: 00010286 Dec 11 11:54:47 tgkvm14 0000000664 kernel: RAX: ffff95c5f4c9ee00 RBX: ffff95c5f3aa7548 RCX: 0000000002400040 Dec 11 11:54:47 tgkvm14 0000000665 kernel: RDX: 0000000000000000 RSI: 0000000000000015 RDI: ffff95c5f566c000 Dec 11 11:54:47 tgkvm14 0000000666 kernel: RBP: ffff95c5f566c000 R08: 0000000000000002 R09: 000000000000167e Dec 11 11:54:47 tgkvm14 0000000667 kernel: R10: 0000000000000040 R11: fffff532c1045c40 R12: fffff532c1045c40 Dec 11 11:54:47 tgkvm14 0000000668 kernel: R13: ffff95c5f3aa7520 R14: ffff9ad7468a3998 R15: ffff95c5f3aa76b8 Dec 11 11:54:47 tgkvm14 0000000669 kernel: FS: 00007fd921124700(0000) GS:ffff95c5fcc80000(0000) knlGS:0000000000000000 Dec 11 11:54:47 tgkvm14 0000000670 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Dec 11 11:54:47 tgkvm14 0000000671 kernel: CR2: 0000000000000000 CR3: 0000000073ec4000 CR4: 00000000000006e0 Dec 11 11:54:47 tgkvm14 0000000672 kernel: Stack: Dec 11 11:54:47 tgkvm14 0000000673 kernel: c350512689f7b37d ffff95c5f3aa7548 c350512689f7b37d ffff95c5f3aa7548 Dec 11 11:54:47 tgkvm14 0000000674 kernel: ffff95c5b39c4958 fffff532c1045c40 ffffffff8126729c ffffffff812601ba Dec 11 11:54:47 tgkvm14 0000000675 kernel: ffff95c5b39c4958 0000000000000000 c350512689f7b37d fffff532c1045c40 Dec 11 11:54:47 tgkvm14 0000000676 kernel: Call Trace: Dec 11 11:54:47 tgkvm14 0000000677 kernel: [<ffffffff8126729c>] ? ext4_page_mkwrite+0x307/0x52b Dec 11 11:54:47 tgkvm14 0000000678 kernel: [<ffffffff812601ba>] ? _ext4_get_block+0xd8/0xd8 Dec 11 11:54:47 tgkvm14 0000000679 kernel: [<ffffffff8117ff78>] ? do_page_mkwrite+0x6e/0xd8 Dec 11 11:54:47 tgkvm14 0000000680 kernel: [<ffffffff811859a6>] ? handle_mm_fault+0x686/0xf9b Dec 11 11:54:47 tgkvm14 0000000681 kernel: [<ffffffff811ec9bb>] ? mntput_no_expire+0x1f/0x21e Dec 11 11:54:47 tgkvm14 0000000682 kernel: [<ffffffff8104f566>] ? __do_page_fault+0x21d/0x465 Dec 11 11:54:47 tgkvm14 0000000683 kernel: [<ffffffff811e1b2c>] ? dput+0x4a/0x2f7 Dec 11 11:54:47 tgkvm14 0000000684 kernel: [<ffffffff819c1bc2>] ? page_fault+0x22/0x30 Dec 11 11:54:47 tgkvm14 0000000685 kernel: [<ffffffff8131cabc>] ? copy_user_generic_string+0x2c/0x40 Dec 11 11:54:47 tgkvm14 0000000686 kernel: [<ffffffff813270d8>] ? copy_page_to_iter+0x8c/0x2b8 Dec 11 11:54:47 tgkvm14 0000000687 kernel: [<ffffffff81153e3f>] ? generic_file_read_iter+0x26e/0x845 Dec 11 11:54:47 tgkvm14 0000000688 kernel: [<ffffffff81318e51>] ? timerqueue_del+0x31/0x90 Dec 11 11:54:47 tgkvm14 0000000689 kernel: [<ffffffffc0426b7b>] ? ceph_read_iter+0x697/0xa33 [ceph] Dec 11 11:54:47 tgkvm14 0000000690 kernel: [<ffffffff8110bf48>] ? hrtimer_cancel+0x23/0x41 Dec 11 11:54:47 tgkvm14 0000000691 kernel: [<ffffffff8111ee85>] ? futex_wait+0x1c8/0x24d Dec 11 11:54:47 tgkvm14 0000000692 kernel: [<ffffffff8111eb2c>] ? get_futex_key+0x32c/0x39a Dec 11 11:54:47 tgkvm14 0000000693 kernel: [<ffffffff811c3912>] ? __vfs_read+0xe0/0x130 Dec 11 11:54:47 tgkvm14 0000000694 kernel: [<ffffffff811c43fb>] ? vfs_read.part.1+0x6c/0x123 Dec 11 11:54:47 tgkvm14 0000000695 kernel: [<ffffffff81185b51>] ? handle_mm_fault+0x831/0xf9b Dec 11 11:54:47 tgkvm14 0000000696 kernel: [<ffffffff811e8f99>] ? __fget+0x7e/0xbf Dec 11 11:54:47 tgkvm14 0000000697 kernel: [<ffffffff811c6020>] ? SyS_read+0x4d/0xb5 Dec 11 11:54:47 tgkvm14 0000000698 kernel: [<ffffffff819c0560>] ? entry_SYSCALL_64_fastpath+0x13/0x94 Dec 11 11:54:47 tgkvm14 0000000699 kernel: Code: 28 00 00 00 48 89 44 24 10 31 c0 65 48 8b 04 25 80 c4 00 00 48 8b 80 d0 06 00 00 48 85 ff 74 31 48 89 fd 48 85 c0 74 34 48 8b 10 <48> 3b 3a 75 2a 83 40 14 01 48 8b 5c 24 10 65 48 33 1c 25 28 00 Dec 11 11:54:47 tgkvm14 0000000700 kernel: RIP [<ffffffff812aa5a2>] jbd2__journal_start+0x38/0x1a2 Dec 11 11:54:47 tgkvm14 0000000701 kernel: RSP <ffff9ad7468a3900> Dec 11 11:54:47 tgkvm14 0000000702 kernel: CR2: 0000000000000000 Dec 11 11:54:47 tgkvm14 0000000703 kernel: ---[ end trace e3db716dfebb89ae ]---
