System crash in nvme_mpath_revalidate_paths() due to NULL gendisk
Issue
- System crash in
nvme_mpath_revalidate_paths()due to NULLgendisk:
nvme nvme1: rescanning namespaces.
nvme nvme2: rescanning namespaces.
nvme nvme0: rescanning namespaces.
nvme nvme3: rescanning namespaces.
nvme0n3: detected capacity change from 0 to 107374182400
nvme0n3: detected capacity change from 0 to 107374182400
BUG: unable to handle kernel NULL pointer dereference at 0000000000000050
nvme0n3: detected capacity change from 0 to 107374182400
PGD 0 P4D 0
Oops: 0000 [#1] SMP NOPTI
CPU: 8 PID: 674 Comm: kworker/u33:8 Kdump: loaded Tainted: G X --------- - - 4.18.0-425.10.1.el8_7.x86_64 #1
Hardware name: Lenovo ThinkSystem SR850 V2/7D32CTO1WW, BIOS M5E120N-1.61 10/28/2022
Workqueue: nvme-wq nvme_scan_work [nvme_core]
RIP: 0010:nvme_mpath_revalidate_paths+0x26/0xb0 [nvme_core]
....
Call Trace:
nvme_update_ns_info+0x13b/0x260 [nvme_core]
nvme_validate_or_alloc_ns+0x2ff/0xa00 [nvme_core]
? __blk_mq_free_request+0x6b/0x90
nvme_scan_work+0x151/0x240 [nvme_core]
process_one_work+0x1a7/0x360
worker_thread+0x30/0x390
? create_worker+0x1a0/0x1a0
kthread+0x10b/0x130
? set_kthread_struct+0x50/0x50
ret_from_fork+0x1f/0x40
Environment
- Red Hat Enterprise Linux 8
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.