[LU-445] Send timestamps with LNet counters

• Type: Improvement
• Resolution: Fixed
• Priority: Blocker
• Fix Version/s: Lustre 2.3.0
• Affects Version/s: Lustre 2.3.0
• Bugzilla ID: 22,639
• Assignee: Doug Oucharek (Inactive)
• Reporter: Wally Wang (Inactive)

Description

We should send the timestamp data from the node along with the performance
counters. This will give us much more accurate performance data on a per-node
and per-group basis.

This is originally discussed in Oracle Bug 22639.

Comments

[Doug Oucharek (Inactive)]

The patch for making the original work backward compatible is: http://review.whamcloud.com/#change,3192

[Doug Oucharek (Inactive)]

With Liang's recommendation, I am reopening this ticket to add a patch to make the change previously done backwards compatible.

With the change as is, a 2.3 system running against a 2.2 or 2.1 system will have an invalid timestamp for doing bandwidth calculations.

I plan to add a new flag to "lst stat" which will trigger the use of the remote timestamps. If the flag is not given, then the previous behaviour, using the local timestamp, will be done.

This change will be set up to change the default from using local timestamps to using remote timestamp when the Lustre version hits 2.8.

[Build Master (Inactive)]

Integrated in lustre-dev » x86_64,client,el6,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/include/lnet/lnetst.h
• lnet/utils/lst.c
• lnet/selftest/framework.c
• lnet/selftest/selftest.h

[Build Master (Inactive)]

Integrated in lustre-dev » x86_64,server,el5,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/selftest/framework.c
• lnet/utils/lst.c
• lnet/selftest/selftest.h
• lnet/include/lnet/lnetst.h

[Build Master (Inactive)]

Integrated in lustre-dev » i686,client,el5,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/utils/lst.c
• lnet/selftest/selftest.h
• lnet/include/lnet/lnetst.h
• lnet/selftest/framework.c

[Build Master (Inactive)]

Integrated in lustre-dev » x86_64,server,el6,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/include/lnet/lnetst.h
• lnet/selftest/selftest.h
• lnet/utils/lst.c
• lnet/selftest/framework.c

[Build Master (Inactive)]

Integrated in lustre-dev » i686,server,el5,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/selftest/selftest.h
• lnet/selftest/framework.c
• lnet/include/lnet/lnetst.h
• lnet/utils/lst.c

[Build Master (Inactive)]

Integrated in lustre-dev » i686,client,el6,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/selftest/selftest.h
• lnet/selftest/framework.c
• lnet/utils/lst.c
• lnet/include/lnet/lnetst.h

[Build Master (Inactive)]

Integrated in lustre-dev » x86_64,client,el5,inkernel #340
LU-445 lnet: Send timestamps with LNet counters (Revision bc5b01bba4a6d934bb1092fab37adc9295a98487)

Result = SUCCESS
Oleg Drokin : bc5b01bba4a6d934bb1092fab37adc9295a98487
Files :

• lnet/utils/lst.c
• lnet/selftest/selftest.h
• lnet/selftest/framework.c
• lnet/include/lnet/lnetst.h

[Liang Zhen (Inactive)]

I just posted a patch at here: http://review.whamcloud.com/#change,2522

[Alexander Boyko]

Liang, can you tell LU number or gerrit request for b1_8 patch?