Details
-
Question/Request
-
Resolution: Fixed
-
Minor
-
Lustre 2.7.0
-
None
-
9223372036854775807
Description
From the documentation it is not clear whether wild-card are suppose work in building tbf jobid rules. My test show it doesn't work.
For example
lctl set_param mds.MDS.mdt.nrs_tbf_rule="start rule1 { *.pbssrv1 } 100" lctl set_param mds.MDS.mdt.nrs_tbf_rule="start rule2 { *.pbssrv2 } 500"
Attachments
Issue Links
- is related to
-
LU-3558 NRS TBF policy for QoS purposes
-
- Resolved
-
One user reported that:
oss#lctl set_param ost.OSS.ost_io.nrs_policies="tbf jobid"
oss#lctl set_param ost.OSS.ost_io.nrs_tbf_rule="start runas jobid={*.500} rate=10"
oss# lctl get_param ost.OSS.ost_io.nrs_tbf_rule
ost.OSS.ost_io.nrs_tbf_rule=
regular_requests:
CPT 0:
runas {*.500} 10, ref 0
default {*} 10000, ref 2
high_priority_requests:
CPT 0:
runas {*.500} 10, ref 0
default {*} 10000, ref 1
------------------------------------------------------------------------------------------------------------
cli#lfs setstripe -c 1 -i 0 /mnt/lustre_2_10/qos/a1
cli#dd if=/dev/zero of=/mnt/lustre_2_10/qos/a1 count=1000 bs=1M
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 12.2554 s, 85.6 MB/s
The rate is not under control.
The patch above fixes the problem.