Hi guys.
I have a healthy pve cluster with GlusterFS volumes in storage - GlusterFS servers (outside of pve cluster) are ver. 11.x and PVE is ver. 8.x (up-to-date)
When I enable/add quota on a GF volume very soon cluster dies, almost immediately, with:
and to confirm - quota removed/disabled and PVE cluster is back to normal.
Wanted to ask here - if anybody can see/reproduce this - for in my mind this is critical (bug?)
many thanks, L.
I have a healthy pve cluster with GlusterFS volumes in storage - GlusterFS servers (outside of pve cluster) are ver. 11.x and PVE is ver. 8.x (up-to-date)
When I enable/add quota on a GF volume very soon cluster dies, almost immediately, with:
Code:
[Wed Feb 7 12:08:16 2024] kvm[996]: memfd_create() called without MFD_EXEC or MFD_NOEXEC_SEAL set
[Wed Feb 7 12:12:02 2024] INFO: task pvestatd:1175 blocked for more than 120 seconds.
[Wed Feb 7 12:12:02 2024] Tainted: P O 6.5.11-8-pve #1
[Wed Feb 7 12:12:02 2024] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[Wed Feb 7 12:12:02 2024] task:pvestatd state:D stack:0 pid:1175 ppid:942 flags:0x00004006
[Wed Feb 7 12:12:02 2024] Call Trace:
[Wed Feb 7 12:12:02 2024] <TASK>
[Wed Feb 7 12:12:02 2024] __schedule+0x3fd/0x1450
[Wed Feb 7 12:12:02 2024] ? srso_return_thunk+0x5/0x10
[Wed Feb 7 12:12:02 2024] ? __wake_up_common_lock+0x8b/0xd0
[Wed Feb 7 12:12:02 2024] schedule+0x63/0x110
[Wed Feb 7 12:12:02 2024] request_wait_answer+0x1be/0x2a0
[Wed Feb 7 12:12:02 2024] ? __pfx_autoremove_wake_function+0x10/0x10
[Wed Feb 7 12:12:02 2024] fuse_simple_request+0x19d/0x2d0
[Wed Feb 7 12:12:02 2024] fuse_statfs+0xfc/0x170
[Wed Feb 7 12:12:02 2024] statfs_by_dentry+0x6d/0xb0
[Wed Feb 7 12:12:02 2024] user_statfs+0x6a/0xe0
[Wed Feb 7 12:12:02 2024] __do_sys_statfs+0x3d/0x80
[Wed Feb 7 12:12:02 2024] __x64_sys_statfs+0x16/0x20
[Wed Feb 7 12:12:02 2024] do_syscall_64+0x5b/0x90
[Wed Feb 7 12:12:02 2024] entry_SYSCALL_64_after_hwframe+0x6e/0xd8
[Wed Feb 7 12:12:02 2024] RIP: 0033:0x7fafa247baf7
[Wed Feb 7 12:12:02 2024] RSP: 002b:00007ffdb41860c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000089
[Wed Feb 7 12:12:02 2024] RAX: ffffffffffffffda RBX: 00007ffdb4186180 RCX: 00007fafa247baf7
[Wed Feb 7 12:12:02 2024] RDX: 00000000000012f8 RSI: 00007ffdb41860d0 RDI: 00005559e1481c20
[Wed Feb 7 12:12:02 2024] RBP: 00007ffdb41860d0 R08: 00005559e148bab0 R09: 0000000000000115
[Wed Feb 7 12:12:02 2024] R10: 0000000000000000 R11: 0000000000000246 R12: 00005559e1499878
[Wed Feb 7 12:12:02 2024] R13: 0000000000000000 R14: 00005559e1499848 R15: 00005559dc9465d0
[Wed Feb 7 12:12:02 2024] </TASK>and to confirm - quota removed/disabled and PVE cluster is back to normal.
Wanted to ask here - if anybody can see/reproduce this - for in my mind this is critical (bug?)
many thanks, L.