Can someone see the related bug?

Issues related to configuring your network
Post Reply
Whoever
Posts: 1357
Joined: 2013/09/06 03:12:10

Can someone see the related bug?

Post by Whoever » 2022/01/23 02:20:03

This issue suggests that the root cause of the issue is a bug and references a bug number. Can anyone tell me the status of the bug?
"Solution":
https://access.redhat.com/solutions/5738871
Refers to bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1920715

I am asking because I am seeing NFS hanging when attempting to mount an export.

Note that this only happens with NFS v4. While this problem exists, I can mount the same export with all the same options (except that I change "nfsvers=4" to "nfsvers=3"). The system isn't particularly busy: all the user processes are waiting on the mount, but uptime shows a system load average of around 2.

This is happening on virtual machines. The hosts are typically busy.

The log shows:
Jan 22 03:25:35 myvm kernel: INFO: task mount.nfs:26167 blocked for more than 120 seconds.
Jan 22 03:25:35 myvm kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jan 22 03:25:35 myvm kernel: mount.nfs D ffffffffc058ce08 0 26167 26163 0x00000080
Jan 22 03:25:35 myvm kernel: ffff880135a6b950 0000000000000082 ffff88003670dee0 ffff880135a6bfd8
Jan 22 03:25:35 myvm kernel: ffff880135a6bfd8 ffff880135a6bfd8 ffff88003670dee0 ffffffffc058ce00
Jan 22 03:25:35 myvm kernel: ffffffffc058ce04 ffff88003670dee0 00000000ffffffff ffffffffc058ce08
Jan 22 03:25:35 myvm kernel: Call Trace:
Jan 22 03:25:35 myvm kernel: [<ffffffff816aa3e9>] schedule_preempt_disabled+0x29/0x70
Jan 22 03:25:35 myvm kernel: [<ffffffff816a8317>] __mutex_lock_slowpath+0xc7/0x1d0
Jan 22 03:25:35 myvm kernel: [<ffffffff816a772f>] mutex_lock+0x1f/0x2f
Jan 22 03:25:35 myvm kernel: [<ffffffffc053ed08>] nfs4_discover_server_trunking+0x48/0x2e0 [nfsv4]
Jan 22 03:25:35 myvm kernel: [<ffffffffc0547896>] nfs4_init_client+0x126/0x300 [nfsv4]
Jan 22 03:25:35 myvm kernel: [<ffffffff811df545>] ? kmem_cache_alloc+0x35/0x1e0
Jan 22 03:25:35 myvm kernel: [<ffffffffc049d526>] ? __fscache_acquire_cookie+0x66/0x180 [fscache]
Jan 22 03:25:35 myvm kernel: [<ffffffffc049d526>] ? __fscache_acquire_cookie+0x66/0x180 [fscache]
Jan 22 03:25:36 myvm kernel: [<ffffffffc02deb51>] ? __rpc_init_priority_wait_queue+0x81/0xc0 [sunrpc]
Jan 22 03:25:36 myvm kernel: [<ffffffffc04b76d6>] nfs_get_client+0x2c6/0x3e0 [nfs]
Jan 22 03:25:36 myvm kernel: [<ffffffffc0546d78>] nfs4_set_client+0x98/0x130 [nfsv4]
Jan 22 03:25:36 myvm kernel: [<ffffffffc054856e>] nfs4_create_server+0x13e/0x3b0 [nfsv4]
Jan 22 03:25:36 myvm kernel: [<ffffffffc053f77e>] nfs4_remote_mount+0x2e/0x60 [nfsv4]
Jan 22 03:25:36 myvm kernel: [<ffffffff81205589>] mount_fs+0x39/0x1b0
Jan 22 03:25:36 myvm kernel: [<ffffffff811a5f95>] ? __alloc_percpu+0x15/0x20
Jan 22 03:25:36 myvm kernel: [<ffffffff81221ec7>] vfs_kern_mount+0x67/0x110
Jan 22 03:25:36 myvm kernel: [<ffffffffc053f6a6>] nfs_do_root_mount+0x86/0xc0 [nfsv4]
Jan 22 03:25:36 myvm kernel: [<ffffffffc053faa4>] nfs4_try_mount+0x44/0xc0 [nfsv4]
Jan 22 03:25:36 myvm kernel: [<ffffffffc04b8627>] ? get_nfs_version+0x27/0x90 [nfs]
Jan 22 03:25:36 myvm kernel: [<ffffffffc04c4a85>] nfs_fs_mount+0x4c5/0xd90 [nfs]
Jan 22 03:25:36 myvm kernel: [<ffffffffc04c59b0>] ? nfs_clone_super+0x140/0x140 [nfs]
Jan 22 03:25:36 myvm kernel: [<ffffffffc04c38b0>] ? param_set_portnr+0x70/0x70 [nfs]
Jan 22 03:25:36 myvm kernel: [<ffffffff81205589>] mount_fs+0x39/0x1b0
Jan 22 03:25:36 myvm kernel: [<ffffffff811a5f95>] ? __alloc_percpu+0x15/0x20
Jan 22 03:25:36 myvm kernel: [<ffffffff81221ec7>] vfs_kern_mount+0x67/0x110
Jan 22 03:25:36 myvm kernel: [<ffffffff812243d3>] do_mount+0x233/0xaf0
Last edited by Whoever on 2022/01/23 03:53:02, edited 1 time in total.

User avatar
TrevorH
Site Admin
Posts: 33191
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: Can someone see the related bug?

Post by TrevorH » 2022/01/23 03:51:05

No, it's a kernel bug so it's automatically marked private. The reporter can see it and they can add people to the cc list on it and then those people can see it. Otherwise it's Red Hat people only.
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke

Whoever
Posts: 1357
Joined: 2013/09/06 03:12:10

Re: Can someone see the related bug?

Post by Whoever » 2022/01/23 07:10:25

What I am seeing here is that NFS v4 is unusable on machines that are busy. Has anyone else experienced this?

Post Reply