View Issue Details

IDProjectCategoryView StatusLast Update
0015141CentOS-7kernelpublic2018-08-16 17:48
Reportertrekkie 
PrioritynormalSeveritycrashReproducibilityalways
Status newResolutionopen 
Product Version7.5.1804 
Target VersionFixed in Version 
Summary0015141: on reboot after initial creation attempting to start VDO crashes system
DescriptionAfter creating a vdo device for dedup/compression and using it, any system reboot after that the device being mounted (automatically or manually) will cause a system abend.

Tried two hardware platforms. Lenovo TinyPC 710 with USB storage, and Lenovo TD with LSI Megaraid.

able to reproduce now attempting 5 times, every time, same error message.
Steps To Reproducecreate a vdo based device of at least 20TB in size using 32GB slab size (or other, doesn't seem to matter)

Before rebooting, mount and use the vdo device successfully, transfer data.

reboot system

manually start the VDO device (or automatically) and system will abend
Additional Informationvm-core-dmesg-txt contents at time of crash

[ 809.545864] kvdo0:dmsetup: starting device 'vdo_media' device instantiation 0 write policy auto
[ 809.545894] kvdo0:dmsetup: underlying device, REQ_FLUSH: not supported, REQ_FUA: not supported
[ 809.545899] kvdo0:dmsetup: Using mode sync automatically.
[ 809.545942] kvdo0:dmsetup: zones: 1 logical, 1 physical, 1 hash; base threads: 5
[ 810.158866] request botched: dev sda: type=1, flags=122d8001
[ 810.158869] sector 46883405817, nr/cnr 7/4294961333
[ 810.158870] bio ffff98dfe9721f00, biotail ffff98dfe9721f00, buffer 2ab5aadb45038583, len 3584
[ 810.158886] ------------[ cut here ]------------
[ 810.158889] kernel BUG at drivers/scsi/scsi_lib.c:1048!
[ 810.158891] invalid opcode: 0000 [#1] SMP
[ 810.158893] Modules linked in: rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace fscache sd_mod crc_t10dif crct10dif_generic sg uas usb_storage kvdo(OE) uds(OE) sunrpc vfat fat snd_hda_codec_hdmi intel_powerclamp coretemp intel_rapl snd_hda_codec_realtek snd_hda_codec_generic kvm snd_hda_intel irqbypass crc32_pclmul ghash_clmulni_intel snd_hda_codec mei_wdt iTCO_wdt iTCO_vendor_support aesni_intel ppdev snd_hda_core lrw snd_hwdep gf128mul glue_helper snd_seq ablk_helper cryptd snd_seq_device snd_pcm mei_me mei snd_timer pcspkr snd shpchp i2c_i801 soundcore parport_pc parport tpm_crb acpi_pad wmi ip_tables xfs libcrc32c i915 iosf_mbi i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ahci e1000e drm libahci libata crct10dif_pclmul crct10dif_common crc32c_intel nvme
[ 810.158929] nvme_core serio_raw ptp pps_core i2c_hid i2c_core video dm_mirror dm_region_hash dm_log dm_mod
[ 810.158935] CPU: 4 PID: 29 Comm: ksoftirqd/4 Kdump: loaded Tainted: G OE ------------ 3.10.0-862.9.1.el7.x86_64 #1
[ 810.158938] Hardware name: LENOVO 10MRCTO1WW/3111, BIOS M1AKT35A 03/21/2018
[ 810.158940] task: ffff98deb9d9af70 ti: ffff98deb9de4000 task.ti: ffff98deb9de4000
[ 810.158944] RIP: 0010:[<ffffffff86aa2f9c>] [<ffffffff86aa2f9c>] scsi_init_io+0x1cc/0x200
[ 810.158950] RSP: 0018:ffff98deb9de7af0 EFLAGS: 00010046
[ 810.158952] RAX: ffff98e1406e9800 RBX: ffff98e12a3f6f00 RCX: 0000000000000000
[ 810.158955] RDX: ffff98e12a3f6f00 RSI: 0000000000000020 RDI: ffff98e137a748c0
[ 810.158958] RBP: ffff98deb9de7b28 R08: ffff98e1406e9838 R09: ffffffff86a9a81b
[ 810.158960] R10: 0000000048ab7b01 R11: ffffe53c1022ad00 R12: ffff98e137a748c0
[ 810.158963] R13: 0000000000000000 R14: ffff98e137a748c0 R15: 00000000ffd16b74
[ 810.158966] FS: 0000000000000000(0000) GS:ffff98e15dd00000(0000) knlGS:0000000000000000
[ 810.158969] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 810.158972] CR2: 00005596237e2cd0 CR3: 00000002fd20e000 CR4: 00000000003607e0
[ 810.158975] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 810.158977] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 810.158980] Call Trace:
[ 810.158984] [<ffffffffc080b3fd>] sd_setup_read_write_cmnd+0x3d/0x950 [sd_mod]
[ 810.158989] [<ffffffff86a9a8ba>] ? __scsi_get_command+0x1a/0xf0
[ 810.158992] [<ffffffffc080bd3f>] sd_init_command+0x2f/0xc0 [sd_mod]
[ 810.158996] [<ffffffff86aa30e1>] scsi_setup_cmnd+0x111/0x1c0
[ 810.158999] [<ffffffff86aa326b>] scsi_prep_fn+0xdb/0x180
[ 810.159003] [<ffffffff8691c95a>] blk_peek_request+0x17a/0x2a0
[ 810.159006] [<ffffffff86aa4c48>] scsi_request_fn+0x48/0x680
[ 810.159010] [<ffffffff86919519>] __blk_run_queue+0x39/0x50
[ 810.159013] [<ffffffff86919596>] blk_run_queue+0x26/0x40
[ 810.159016] [<ffffffff86aa3668>] scsi_run_queue+0x258/0x2f0
[ 810.159019] [<ffffffff86914fbd>] ? elv_requeue_request+0x5d/0xb0
[ 810.159022] [<ffffffff86aa598f>] scsi_io_completion+0x46f/0x6a0
[ 810.159025] [<ffffffff86a9ab6c>] scsi_finish_command+0xdc/0x140
[ 810.159029] [<ffffffff86aa4bd2>] scsi_softirq_done+0x132/0x160
[ 810.159032] [<ffffffff869231d6>] blk_done_softirq+0x96/0xc0
[ 810.159035] [<ffffffff8669b085>] __do_softirq+0xf5/0x280
[ 810.159038] [<ffffffff8669b248>] run_ksoftirqd+0x38/0x50
[ 810.159042] [<ffffffff866c3dd4>] smpboot_thread_fn+0x144/0x1a0
[ 810.159045] [<ffffffff866c3c90>] ? lg_double_unlock+0x40/0x40
[ 810.159048] [<ffffffff866bb621>] kthread+0xd1/0xe0
[ 810.159051] [<ffffffff866bb550>] ? insert_kthread_work+0x40/0x40
[ 810.159055] [<ffffffff86d205dd>] ret_from_fork_nospec_begin+0x7/0x21
[ 810.159058] [<ffffffff866bb550>] ? insert_kthread_work+0x40/0x40
[ 810.159060] Code: 8b 17 e8 d8 5b ea ff 39 45 cc 7c 43 48 8b 53 38 0f b7 92 18 06 00 00 39 d0 7f 32 4d 89 be f0 00 00 00 41 89 47 08 e9 24 ff ff ff <0f> 0b be 3a 04 00 00 48 c7 c7 f3 aa 0a 87 41 b7 01 e8 8e ef be
[ 810.159086] RIP [<ffffffff86aa2f9c>] scsi_init_io+0x1cc/0x200
[ 810.159090] RSP <ffff98deb9de7af0>
TagsNo tags attached.
abrt_hash
URL

Activities

Dennis Keefe

Dennis Keefe

2018-08-16 14:37

reporter   ~0032504

None of those functions in the call trace are VDO functions.

Checking on the line cited by the kernel bug, upstream Linux kernel commit 635d98b1d0cfc2ba3426a701725d31a6102c059a is the origin of the BUG_ON() call that halted the system, with the following comment: "scsi_init_io should only be called for requests that transfer data..."

BUG_ON(!rq->nr_phys_segments);
Dennis Keefe

Dennis Keefe

2018-08-16 14:41

reporter   ~0032505

This issue Category should be changed to Kernel from General

Issue History

Date Modified Username Field Change
2018-08-04 20:56 trekkie New Issue
2018-08-16 14:37 Dennis Keefe Note Added: 0032504
2018-08-16 14:41 Dennis Keefe Note Added: 0032505
2018-08-16 17:48 toracat Category general => kernel