Bug 208907
Summary: | [fstests generic/074 on xfs]: 5.7.10 fails with a hung task on | ||
---|---|---|---|
Product: | File System | Reporter: | Luis Chamberlain (mcgrof) |
Component: | XFS | Assignee: | FileSystem/XFS Default Virtual Assignee (filesystem_xfs) |
Status: | NEW --- | ||
Severity: | normal | CC: | sandeen |
Priority: | P1 | ||
Hardware: | x86-64 | ||
OS: | Linux | ||
Kernel Version: | 5.7.10 | Subsystem: | |
Regression: | No | Bisected commit-id: |
Description
Luis Chamberlain
2020-08-14 18:43:18 UTC
This seems relevant, no? > Aug 14 18:28:16 kdevops-xfs-dev kernel: nvme nvme1: I/O 128 QID 2 timeout, > aborting > Aug 14 18:28:16 kdevops-xfs-dev kernel: nvme nvme1: Abort status: 0x4001 > Aug 14 18:28:47 kdevops-xfs-dev kernel: nvme nvme1: I/O 128 QID 2 timeout, > reset controller then 2.5 minutes later, > Aug 14 18:31:12 kdevops-xfs-dev kernel: INFO: task xfsaild/nvme1n1:289 > blocked for more than 120 seconds. On Fri, Aug 14, 2020 at 06:43:18PM +0000, bugzilla-daemon@bugzilla.kernel.org wrote: > Just ssh to kdevops-xfs and run: > > cd /var/lib/xfstests/ > ./gendisks.sh -m > ./check generic/074 > > Aug 14 18:27:34 kdevops-xfs-dev kernel: XFS (loop16): Mounting V5 Filesystem > Aug 14 18:27:34 kdevops-xfs-dev kernel: XFS (loop16): Ending clean mount > Aug 14 18:27:34 kdevops-xfs-dev kernel: xfs filesystem being mounted at > /media/test supports timestamps until 2038 (0x7fffffff) > Aug 14 18:28:16 kdevops-xfs-dev kernel: nvme nvme1: I/O 128 QID 2 timeout, > aborting > Aug 14 18:28:16 kdevops-xfs-dev kernel: nvme nvme1: Abort status: 0x4001 > Aug 14 18:28:47 kdevops-xfs-dev kernel: nvme nvme1: I/O 128 QID 2 timeout, > reset controller Hardware lost an IO. I'm guessing the error handling that reset the controller failed to error out the bio the lost IO belonged to, so XFS has hung waiting for it... Cheers, Dave. |