Skip to Content.
Sympa Menu

perfsonar-user - Re: [perfsonar-user] perfsonar 4.0 MeshConfig problem

Subject: perfSONAR User Q&A and Other Discussion

List archive

Re: [perfsonar-user] perfsonar 4.0 MeshConfig problem


Chronological Thread 
  • From: Edward Toth <>
  • To: Andrew Lake <>
  • Cc: , Antoine Delvaux <>
  • Subject: Re: [perfsonar-user] perfsonar 4.0 MeshConfig problem
  • Date: Fri, 5 May 2017 10:47:23 +0200
  • Ironport-phdr: 9a23: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


Hi,

sorry for the late answer.


On 2017-04-28 15:46, Andrew Lake wrote:
Hi,

Looks like the problem is that 64.251.58.166 is in a traceroute mesh but
does not have a traceroute measurement_archive defined. You didn't see it
pre-4.0 because the meshconfig-guiagent did not generate traceroute
dashboards so it was ignoring those tests.

That problem looks separate from the central archive not getting results.
Looking at your pscheduler tasks on one of the hosts I could reach, it
looks like they are having some type of problem reaching the archive.
Running the command below shows an example of this if you are curious:

pscheduler result --archivings
https://192.35.241.164/pscheduler/tasks/323d255d-c7de-47e5-81c6-d03e04926790/runs/c3cec40b-1d42-41e1-902f-c2e32635b288

This command only gives me an error that the number of arguments supplied is wrong. I assume the reason is, that I cannot reach the server...
However I was able to run the command on one of the test nodes (I did not try it on the other ones) and get a result back.

>From one of the hosts in question, are you able to run the following?

curl -k "http://128.130.2.80/esmond/perfsonar/archive/?limit=1";

If it gives a 500 error or similar, could you send me the logs under
/var/log/esmond/ from the central archive host? It might be something is
wrong with the archive.
This command gives:

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
<html><head>
<title>302 Found</title>
</head><body>
<h1>Found</h1>
<p>The document has moved <a href="https://128.130.2.80/maddash-webui?limit=1";>here</a>.</p>
<hr>
<address>Apache/2.2.15 (CentOS) Server at 128.130.2.80 Port 80</address>
</body></html>

With https instead of http it gives the content of archive.txt (see attachment)
I also attached the central archive logs anyway, even though the command didn't give a 500.

Maybe I should mention that I changed the apache configuration to redirect any http request to an https request. I didn't think that this might have any impact, but the result of the curl command changed my mind on that.

Thanks again for your help,
Edward


Thanks,
Andy

On April 28, 2017 at 6:39:03 AM, Edward Toth
()
wrote:

Hi,

thanks for the quick response! I attached the .json file to this email.
The schema of the json file should be OK, because it is generated from
.conf file.

Might be, that I overlooked some changes in the configuration?

Thanks,
Edward

On 2017-04-28 12:27, Antoine Delvaux wrote:
Hi Edward,

Are you able to share with us the JSON file for your mesh? We'll then be
able to have closer look.
Thanks,

Antoine.

Le 28 avr. 2017 à 08:17, Edward Toth
<>
a écrit :

Hello!

I recently updated my nodes to perfsonar 4.0, which mostly worked fine
(except for some hiccups).
The setup contains a central management node and several test nodes in
full mesh configuration. A few days after the update I saw, that the
central archive is not getting any test results. The logs for pscheduler on
the test nodes and the management node do not show anything out of the
ordinary.
However I checked the meshconfig-guiagent log and saw that it is not
able to generate the configuration.
I copied the last entry (all entries look like this):

2017/04/28 09:50:44 (15151) ERROR> GUIAgent.pm:219
perfSONAR_PS::MeshConfig::GUIAgent::__configure_guis - Problem generating
maddash configuration: Problem generating maddash configuration: Couldn't
find ma for host: perfSONAR_PS::MeshConfig::Config::Address=HASH(0x4c26070)
at
/usr/lib/perfsonar/bin/../lib/perfSONAR_PS/MeshConfig/Generators/MaDDash.pm
line 352.

The mesh configuration has not changed in almost a year, and it worked
fine before the update and for some time after the update to perfsonar 4.0.
Searching the documentation I could not find anything either, except
that the perfsonar-meshconfig-guiagent replaces the
generate_gui_configuration + cronjob. Have I missed something?
I don't know either if this is related to the central archive not
getting results, or if this a different problem.
Thanks in advance for any help
Edward Toth




--

Technische Universität Wien http://www.tuwien.ac.at
ZID - Zentraler Informatikdienst/Kommunikation E020C
Wiedner Hauptstrasse 8-10/020, A-1040 Wien, Austria
Tel: +43 (1) 58801-420425


[{"url":"http://128.130.2.80/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/","metadata-key":"0397383adc4d44c58316ab04545d56fe","subject-type":"point-to-point","event-types":[{"summaries":[],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/time-error-estimates/base","event-type":"time-error-estimates"},{"summaries":[],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-duplicates/base","event-type":"packet-duplicates"},{"summaries":[{"summary-type":"statistics","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-ttl/statistics/0","time-updated":1462351295,"summary-window":"0"}],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-ttl/base","event-type":"histogram-ttl"},{"summaries":[],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-count-sent/base","event-type":"packet-count-sent"},{"summaries":[],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-count-lost/base","event-type":"packet-count-lost"},{"summaries":[{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/aggregations/300","time-updated":1462351295,"summary-window":"300"},{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/aggregations/3600","time-updated":1462351295,"summary-window":"3600"},{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/aggregations/86400","time-updated":1462351295,"summary-window":"86400"},{"summary-type":"statistics","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/statistics/0","time-updated":1462351295,"summary-window":"0"},{"summary-type":"statistics","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/statistics/300","time-updated":1462351295,"summary-window":"300"},{"summary-type":"statistics","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/statistics/3600","time-updated":1462351295,"summary-window":"3600"},{"summary-type":"statistics","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/statistics/86400","time-updated":1462351295,"summary-window":"86400"}],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/histogram-owdelay/base","event-type":"histogram-owdelay"},{"summaries":[],"time-updated":null,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/failures/base","event-type":"failures"},{"summaries":[{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-loss-rate/aggregations/300","time-updated":1462351295,"summary-window":"300"},{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-loss-rate/aggregations/3600","time-updated":1462351295,"summary-window":"3600"},{"summary-type":"aggregation","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-loss-rate/aggregations/86400","time-updated":1462351295,"summary-window":"86400"}],"time-updated":1462351295,"base-uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/packet-loss-rate/base","event-type":"packet-loss-rate"}],"source":"192.35.241.164","destination":"193.170.79.27","measurement-agent":"192.35.241.164","tool-name":"powstream","input-source":"192.35.241.164","input-destination":"sonar-t1-2.kom.tuwien.ac.at","ip-transport-protocol":"udp","sample-bucket-width":"0.0001","sample-size":"600","time-duration":"60","time-interval":"0","time-probe-interval":"0.1","uri":"/esmond/perfsonar/archive/0397383adc4d44c58316ab04545d56fe/","metadata-count-total":100,"metadata-previous-page":null,"metadata-next-page":"http://128.130.2.80/esmond/perfsonar/archive/?limit=1&offset=1"}]

Attachment: esmond_log.tar.gz
Description: application/gzip




Archive powered by MHonArc 2.6.19.

Top of Page