perfSONAR User Q&A and Other Discussion

[Phil Reese <>]

Hi,

As I've moved to PS4.2, a few gotcha's jumped out but resolved themselves.

One last issue remains.  That is, difficulty with the throughput tests 
reporting to esmond and then being posted to the MaDDash webui.

I'm getting these errors in my agent's psconfig-pscheduler-agent.log:
> 2019/09/14 13:17:42 WARN pid=32433 
> prog=perfSONAR_PS::PSConfig::PScheduler::Agent::_run_end line=227 
> guid=A6320BD2-D72C-11E9-9A0B-99DF719431D2 msg=Problem adding test 
> throughput(192.168.1.45->192.168.1.66), continuing with rest of 
> config: Inactivity timeout
or
> 2019/09/13 12:38:58 WARN pid=32396 
> prog=perfSONAR_PS::PSConfig::PScheduler::Agent::_run_end line=227 
> guid=CA195462-D65D-11E9-857D-71D33FEC092A msg=Problem adding test 
> throughput(sups-dc1.stanford.edu->sups-dc3.stanford.edu), continuing 
> with rest of config: INTERNAL SERVER ERROR
The Loss, RTT, and traceroute grids all work fine.

I've several grids going, some at home, using the 192.168.1.x range 
while others on campus using FQDNs, both appear to have similar problems.

I've one grid that has older agents, agents updated from 4.16 to 4.2 
using the auto update process.  They continue to pass throughput info to 
the grid.  I've started to update those agents with new hardware and 
every new agent instance has been unable to post data to the throughput 
grid.

I'm sure its something I've got wrong but I'm not sure where to look.  
Thanks for any pointers.

Phil