[LU-12746] max_read_ahead_async_active can't set 0 from non-zero Created: 11/Sep/19  Updated: 13/Dec/19  Resolved: 13/Dec/19

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Minor
Reporter: Shuichi Ihara Assignee: Wang Shilong (Inactive)
Resolution: Won't Fix Votes: 0
Labels: None
Environment:

master


Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

max_read_ahead_async_active can't be back to 0 (default value) from non-zero value.

# mount -t lustre 10.0.11.220@o2ib10:/scratch0 /scratch0
# lctl get_param  llite.*.max_read_ahead_async_active
llite.scratch0-ffff8f1e453f3800.max_read_ahead_async_active=0
# lctl set_param  llite.*.max_read_ahead_async_active=256
llite.scratch0-ffff8f1e453f3800.max_read_ahead_async_active=256

# lctl set_param  llite.*.max_read_ahead_async_active=0
error: set_param: setting /sys/fs/lustre/llite/scratch0-ffff8f1e453f3800/max_read_ahead_async_active=0: Numerical result out of range

# lctl get_param  llite.*.max_read_ahead_async_active
llite.scratch0-ffff8f1e453f3800.max_read_ahead_async_active=256


 Comments   
Comment by Wang Shilong (Inactive) [ 11/Sep/19 ]

Yup, this is a known problem, the limitation comes from kernel workqueue did not support set to be 0 back.

Comment by Wang Shilong (Inactive) [ 13/Dec/19 ]

This is limitation comes from kernel workqueue

Generated at Sat Feb 10 02:55:17 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.