[LTP] "cpu_hotplug/functional/cpuhotplug05.sh" failed

[LTP] "cpu_hotplug/functional/cpuhotplug05.sh" failed

[Liao Guoqi-B25230]

[-- Attachment #1.1: Type: text/plain, Size: 677 bytes --]

Hi All:

When I run cpuhotplug05 cases on our platform(powerpc), it failed:
Name:   cpuhotplug05
Date:   Wed Oct 16 10:49:31 UTC 2013
Desc:   Does sar behave properly during CPU hotplug events?

CPU is 1
cpuhotplug05    1  TFAIL  :  CPU1 Not Found on SAR!

After some investigation:

the following code:

        while ! grep -iq nan $TMP/log_$$; do

                tst_resm TFAIL "CPU${CPU_TO_TEST} Not Found on SAR!"

                exit_clean 1

        done

root cause:
output of command "sar" has been changed, the log from "sar" output did NOT include "nan".

Enrionment:
Sysstat version: 10.1.4
Ltp version: 20130109-r0

B.R.
Guoqi



[-- Attachment #1.2: Type: text/html, Size: 4438 bytes --]

[-- Attachment #2: Type: text/plain, Size: 416 bytes --]

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk

[-- Attachment #3: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] "cpu_hotplug/functional/cpuhotplug05.sh" failed

[chrubis]

Hi!
> When I run cpuhotplug05 cases on our platform(powerpc), it failed:
> Name:   cpuhotplug05
> Date:   Wed Oct 16 10:49:31 UTC 2013
> Desc:   Does sar behave properly during CPU hotplug events?
> 
> CPU is 1
> cpuhotplug05    1  TFAIL  :  CPU1 Not Found on SAR!
> 
> After some investigation:
> 
> the following code:
> 
>         while ! grep -iq nan $TMP/log_$$; do
> 
>                 tst_resm TFAIL "CPU${CPU_TO_TEST} Not Found on SAR!"
> 
>                 exit_clean 1
> 
>         done
> 
> root cause:
> output of command "sar" has been changed, the log from "sar" output did NOT include "nan".
> 
> Enrionment:
> Sysstat version: 10.1.4
> Ltp version: 20130109-r0

It looks like the sar parameters and output changes from time to time.
On my system (systat 10.0.5) sar does not accept zero as a inifinte
count and needs to be executed as 'sar -P ALL 1' instead of
'sar -P ALL 1 0'.

I guess that the only solution is to switch on systat version and change
expectations accordingly. Feel free to send a patch.

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] "cpu_hotplug/functional/cpuhotplug05.sh" failed

[chrubis]

Hi!
> ---
>  .../hotplug/cpu_hotplug/functional/cpuhotplug05.sh |    8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/testcases/kernel/hotplug/cpu_hotplug/functional/cpuhotplug05.sh b/testcases/kernel/hotplug/cpu_hotplug/functional/cpuhotplug05.sh
> index e64b91c..e893f04 100755
> --- a/testcases/kernel/hotplug/cpu_hotplug/functional/cpuhotplug05.sh
> +++ b/testcases/kernel/hotplug/cpu_hotplug/functional/cpuhotplug05.sh
> @@ -57,8 +57,8 @@ until [ $TST_COUNT -gt $TST_TOTAL ]; do
>         sleep 2
>         SAR_PID=$!
> 
> -       # Verify that SAR has correctly listed the missing CPU as 'nan'
> -       while ! grep -iq nan $TMP/log_$$; do
> +       # Verify that SAR has correctly listed the missing CPU
> +       while ! awk '{print $8}' $TMP/log_$$|grep -i "^0.00"; do
>                 tst_resm TFAIL "CPU${CPU_TO_TEST} Not Found on SAR!"
>                 exit_clean 1
>         done
> @@ -66,7 +66,7 @@ until [ $TST_COUNT -gt $TST_TOTAL ]; do
>         sleep .5
> 
>         # Verify that at least some of the CPUs are offline
> -       NUMBER_CPU_OFF=$(grep "$time" $TMP/log_$$ | grep -i nan | wc -l)
> +        NUMBER_CPU_OFF=$(grep "$time" $TMP/log_$$ | awk '{print $8}' |grep -i "^0.00" | wc -l)
>         if [ ${NUMBER_CPU_OFF} -eq 0 ]; then
>                 tst_resm TBROK "no CPUs found offline"
>                 exit_clean 1
> @@ -83,7 +83,7 @@ until [ $TST_COUNT -gt $TST_TOTAL ]; do
>         sleep .5
> 
>         # Check that SAR registered the change in CPU online/offline states
> -       NEW_NUMBER_CPU_OFF=$(grep "$time" $TMP/log_$$ | grep -i nan | wc -l)
> +        NEW_NUMBER_CPU_OFF=$(grep "$time" $TMP/log_$$|awk '{print $8}'|grep -i "^0.00"| wc -l)
>         : $(( NUMBER_CPU_OFF -= 1 ))
>         if [ "$NUMBER_CPU_OFF" = "$NEW_NUMBER_CPU_OFF" ]; then
>                 tst_resm TPASS "CPU was found after turned on."

Ok, that fixes it for sar that does not print nan. Do you have an idea
for how long the sar behaves this way?

Because I do not wan't to break the test for anybody with slightly older
sasstat package.

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list

Re: [LTP] "cpu_hotplug/functional/cpuhotplug05.sh" failed

[chrubis]

Hi!
> I have no idea that when it was changed. But I think we could update the matching program to compatibility old and new.

Hmm, what about modifying the test to accept either nan or 0.00? That
way we don't have to find you when it was changed.

-- 
Cyril Hrubis
chrubis@suse.cz

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list