crush while running in batch mode

[Luca]

Hello,

i'm using a home made script to feed madevent a set of instruction to perform a scan. For other processes the script works with no problem; for this peculiar process it reports an error after correctly computing a bunch of points (the error happens for different run numbers every time).

Here is the error code I get:

Error when reading /Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat
Command "import command ./j_4nus.mg5" interrupted in sub-command:
"launch" with error:
IOError : [Errno 2] No such file or directory: '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat'
Please report this bug on https://bugs.launchpad.net/mg5amcnlo
More information is found in '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/run_02_point_2_debug.log'.

I attached this file below, as well as the instruction file I used generated by my script. As I said, I don't believe the problem is caused by the latter because I had no problems in using it for other processes.

Thanks for your help and kind regards,
L.

[Luca (luca-marzola)]

As further checks I lunched three tests using every time the same command file (.mg5) included in my previous message.

1) didn't decay the p0, i.e. generated p p > p0 p0 j. Crushes again at the second point.

2) generated p p > p0 p0. No problem

3) generated p p > p0 p0, p0 > vl vl~. No problem

So it seems it doesn't like the jet in the final state. Any clue?

[Olivier Mattelaer (olivier-mattelaer)]

Hi Luca,

Can you send me the log file corresponding to your failing job:
it should be in /Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/

It might be great if I can reproduce your bug but for that I would need to have your UFO model.
Is it possible for your to send it to me? or to attach it here (or put it somewhere online)

Cheers,

Olivier

> On 8 Feb 2018, at 09:50, Luca <email address hidden> wrote:
>
> As further checks I lunched three tests using every time the same
> command file (.mg5) included in my previous message.
>
> 1) didn't decay the p0, i.e. generated p p > p0 p0 j. Crushes again at
> the second point.
>
> 2) generated p p > p0 p0. No problem
>
> 3) generated p p > p0 p0, p0 > vl vl~. No problem
>
> So it seems it doesn't like the jet in the final state. Any clue?
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1748128
>
> Title:
> crush while running in batch mode
>
> Status in MadGraph5_aMC@NLO:
> New
>
> Bug description:
> Hello,
>
> i'm using a home made script to feed madevent a set of instruction to
> perform a scan. For other processes the script works with no problem;
> for this peculiar process it reports an error after correctly
> computing a bunch of points (the error happens for different run
> numbers every time).
>
> Here is the error code I get:
>
> Error when reading /Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat
> Command "import command ./j_4nus.mg5" interrupted in sub-command:
> "launch" with error:
> IOError : [Errno 2] No such file or directory: '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat'
> Please report this bug on https://bugs.launchpad.net/mg5amcnlo
> More information is found in '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/run_02_point_2_debug.log'.
>
> I attached this file below, as well as the instruction file I used
> generated by my script. As I said, I don't believe the problem is
> caused by the latter because I had no problems in using it for other
> processes.
>
> Thanks for your help and kind regards,
> L.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1748128/+subscriptions

[Luca (luca-marzola)]

Hi Oliver,

thanks. I am attaching the G1 subfolder and the UFO model file.

Best,
L.

[Luca (luca-marzola)]

Hi Oliver,

did you manage to reproduce the bug? Can I help you in some other way?

Best,
L.

[Olivier Mattelaer (olivier-mattelaer)]

Sorry was in holliday last week.
Tommorow I'm following some formation, so I will try to look on Thursday.

Cheers,

Olivier

> On 13 Feb 2018, at 08:05, Luca <email address hidden> wrote:
>
> Hi Oliver,
>
> did you manage to reproduce the bug? Can I help you in some other way?
>
> Best,
> L.
>
> --
> You received this bug notification because you are subscribed to
> MadGraph5_aMC@NLO.
> https://bugs.launchpad.net/bugs/1748128
>
> Title:
> crush while running in batch mode
>
> Status in MadGraph5_aMC@NLO:
> Incomplete
>
> Bug description:
> Hello,
>
> i'm using a home made script to feed madevent a set of instruction to
> perform a scan. For other processes the script works with no problem;
> for this peculiar process it reports an error after correctly
> computing a bunch of points (the error happens for different run
> numbers every time).
>
> Here is the error code I get:
>
> Error when reading /Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat
> Command "import command ./j_4nus.mg5" interrupted in sub-command:
> "launch" with error:
> IOError : [Errno 2] No such file or directory: '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/SubProcesses/P1_gq_p0p0q_p0_vlvl_p0_vlvl/G1/results.dat'
> Please report this bug on https://bugs.launchpad.net/mg5amcnlo
> More information is found in '/Users/luca/Programming/Physics/B_physics/Neutral_Scalar/j_4nus/run_02_point_2_debug.log'.
>
> I attached this file below, as well as the instruction file I used
> generated by my script. As I said, I don't believe the problem is
> caused by the latter because I had no problems in using it for other
> processes.
>
> Thanks for your help and kind regards,
> L.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mg5amcnlo/+bug/1748128/+subscriptions

[Olivier Mattelaer (olivier-mattelaer)]

Hi,

Ok this was a funny bug related to some variable that were not correctly reset between run.
In most cases such variable is actually the same between all the benchmark and this is therefore not a problem.

Cheers and thanks,

Olivier