Skip to Content.
Sympa Menu

perfsonar-user - AW: [perfsonar-user] Debian ARM testpoint and pscheduler

Subject: perfSONAR User Q&A and Other Discussion

List archive

AW: [perfsonar-user] Debian ARM testpoint and pscheduler


Chronological Thread 
  • From: "Garnizov, Ivan" <>
  • To: Alan Whinery <>, "" <>
  • Subject: AW: [perfsonar-user] Debian ARM testpoint and pscheduler
  • Date: Wed, 3 Apr 2019 09:54:27 +0000

Hi Alan,

 

With these errors about memory limits reached, please share the hardware specification.

Also please consider the list of suggested actions in my previous email.

 

 

Regards,

Ivan Garnizov

 

GEANT WP6T3: pS development team

GEANT WP6T1: pS deployments GN Operations

GEANT WP9T2: Software governance in GEANT

 

Von: [mailto:] Im Auftrag von Alan Whinery
Gesendet: Mittwoch, 3. April 2019 03:42
An:
Betreff: Re: [perfsonar-user] Debian ARM testpoint and pscheduler

 

 So, interleaved with other duties, I got a fresh Deb 9.5 image and ran the testpoint install on it today to look at Ivan's question about install errors. 

 

I did get an install error when the install was almost at completion. I don't think that I had this before, I think I would have noticed.  There were no other errors or warnings than what is shown below. 

 

Processing triggers for pscheduler-server (1.1.6-1) ...

WARNING:  Test "idleex" failed to enumerate: Process took too long to run.

WARNING:  Test "idlebgm" failed to enumerate: Process took too long to run.

WARNING:  Test "clock" failed to enumerate: Process took too long to run.

WARNING:  Test "trace" failed to enumerate: OSError: [Errno 12] Cannot allocate memory

Traceback (most recent call last):

  File "/usr/lib/python2.7/dist-packages/pscheduler/program.py", line 262, in run_program

    process = __get_process(argv, new_env, attempts)

  File "/usr/lib/python2.7/dist-packages/pscheduler/program.py", line 254, in __get_process

    raise ex

OSError: [Errno 12] Cannot allocate memory

 

 

 

 

 

 

On Tue, Apr 2, 2019 at 6:27 AM Antoine Delvaux <> wrote:

Hello Alan,

 

Just a small thing to add to what Ivan said, that would both answer your question and be useful for our debugging, running the command

 

apt-cache show pscheduler-server

 

(or any other package) will show you which pscheduler version you have installed.  And

 

pscheduler plugins (tests|tools|archivers|contexts)

 

will show you which version of plugins you have.

 

Antoine.



Le 2 avr. 2019 à 15:39, Garnizov, Ivan <> a écrit :

 

Hello Alan,

 

From your message it does not become clear if there were any issues during the installation. I would assume there might have been warning and/or errors for you to end up with a non-operational perfSONAR system after a clean install on Debian 9.

 

I would suggest you to:

-           start pscheduler in a debuging mode: pscheduler debug on

-          Run a troubleshoot command: pscheduler troubleshoot

-          Stop pscheduler debug mode: pscheduler debug off

-          Review the pscheduler logs: /var/log/pscheduler/pscheduler.log

 

 

Regards,

Ivan Garnizov

 

GEANT WP6T3: pS development team

GEANT WP6T1: pS deployments GN Operations

GEANT WP9T2:Software governance in GEANT

 

Von:  [] Im Auftrag von Alan Whinery
Gesendet: Montag, 1. April 2019 23:57
An: 
Betreff: Re: [perfsonar-user] Debian ARM testpoint and pscheduler

 

and -- currently limits.conf is a copy of the "identifier-always" example, just to simplify the limits question,

 

 

On Mon, Apr 1, 2019 at 10:06 AM Alan Whinery <> wrote:

After installing per 

 

 

on BeagleBone Black, pscheduler did not, in general, function. Typical response when asking for apparently-any task is:

 

>localhost: Process took too long to run.

 

After copying pscheduler auth rules from a working toolkit to /etc/postgresql/9.6/main/pg_hba.conf and also copying the limits file from a CentOS toolkit, I am able to run tests to the Beaglebone from a remote pscheduler host, but still unable to intitate tests from the Beaglebone itself. 

 

[sidebar: how does one check the installed version of pscheduler?]

 

'troubleshoot' gets:

 

>root@beaglebone:/etc/pscheduler/tool/iperf3# pscheduler troubleshoot

>Performing basic troubleshooting of localhost.

>Checking for pScheduler on localhost... OK.

>Idle test on localhost... Failed.

>  Unable to post task: Unable to determine participants: Process took too long to run.

 

Does anyone recognize a symptom, or know of a shortcoming that prevents the pscheduler command line from working here?

 

Beaglebone Black Debian 9:

Linux beaglebone 4.9.78-ti-r94 #1 SMP PREEMPT Fri Jan 26 21:26:24 UTC 2018 armv7l GNU/Linux

 

root@beaglebone:/etc/pscheduler/tool/iperf3# cat /proc/cpuinfo

processor       : 0

model name      : ARMv7 Processor rev 2 (v7l)

BogoMIPS        : 995.32

Features        : half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpd32

CPU implementer : 0x41

CPU architecture: 7

CPU variant     : 0x3

CPU part        : 0xc08

CPU revision    : 2

 

Hardware        : Generic AM33XX (Flattened Device Tree)

Revision        : 0000

Serial          : 0000000000000000

 

--
To unsubscribe from this list: https://lists.internet2.edu/sympa/signoff/perfsonar-user

 




Archive powered by MHonArc 2.6.19.

Top of Page