Hung task timeout error

[Deleted User]

Why am I seeing this error?

 

"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message

vertica       D 0000000000000002     0 60068      1 0x00000080

ffff88024d9e7c98 0000000000000082 ffffffff81ed09f0 ffff882012336ae0

ffff88024d9e7c68 ffffffff810aa570 ffff88024d9e7ca0 ffff882012336ae0

ffff882012337098 ffff88024d9e7fd8 000000000000fb88 ffff882012337098 Call Trace:

2401  [<ffffffff810aa570>] ? exit_robust_list+0x90/0x160

2402  [<ffffffff81072f95>] exit_mm+0x95/0x180

2403  [<ffffffff810733df>] do_exit+0x15f/0x870

2404  [<ffffffff81063340>] ? wake_up_state+0x10/0x20

2405  [<ffffffff81073b48>] do_group_exit+0x58/0xd0

2406  [<ffffffff81088e16>] get_signal_to_deliver+0x1f6/0x460

2407  [<ffffffff8100a265>] do_signal+0x75/0x800

2408  [<ffffffff81435b25>] ? sys_sendto+0x185/0x190

2409  [<ffffffff8100bbee>] ? invalidate_interrupt1+0xe/0x20

2410  [<ffffffff8100bc2e>] ? invalidate_interrupt3+0xe/0x20

2411  [<ffffffff8100bbce>] ? invalidate_interrupt0+0xe/0x20

2412  [<ffffffff810ace0b>] ? sys_futex+0x7b/0x170

2413  [<ffffffff8100aa80>] do_notify_resume+0x90/0xc0

2414  [<ffffffff8100b341>] int_signal+0x12/0x17

HPBigData_BP

This problem occurs when there is an imbalance in processing power. In Vertica, the memory and processing power (CPU) may be too high and the disk throughput may be comparatively too low to keep up. All Vertica nodes must have an appropriate balance of CPU, RAM, disk throughput and network bandwidth.

 

To avoid this problem, you need to tune the vm.dirty_ratio and vm.dirty_background_ratio parameters in Linux. In addition, you need to set hung_task_panic to 0.

 

For details, see Tuning Linux Dirty Data Parameters for Vertica in the Vertica Knowledge Base in the Developer Community.