perfSONAR User Q&A and Other Discussion

[Jarett DeAngelis <>]

Actually, quick question: the docs say, for example:

Archiving Globally

pScheduler can be configured to apply an archive specification to every run it performs on a host by placing each one in a file in /etc/pscheduler/default-archives. Files must be readable by the pscheduler user.

For example, this file will use the HTTP archiver to post the results of all throughput tests to https://host.example.com/place/to/post:

{
    "archiver": "http",
    "data": {
        "_url": "https://host.example.com/place/to/post",
        "op": "post"
    },
    "transform": {
        "script": "if (.test.type == \"throughput\") then . else null end"
    },
    "ttl": "PT5M"
}

Okay, so on which grid host would I do this? I thought since our central management server (which I understand is also going away as we understand it currently in 5.0) currently collects all this data it might be on there, but nope.

Does it matter? Does the config just get replicated across all nodes?

Thanks,

J

On Dec 15, 2022, at 1:38 PM, Mark Feit <> wrote:

Jarett DeAngelis writes:

We're interested in getting data from our grid into Grafana for somewhat more advanced (/prettier ;) ) visualization. Has anyone else gone down this road either with a combination of the Postgres/Cassandra DBs directly or via the API?

 

Since 4.0, pScheduler has a raft of archiver plugins that allow sending the results of measurements directly to systems other than Esmond (and Esmond, too) using various methods.  Documentation for that is at https://docs.perfsonar.net/pscheduler_ref_archivers.html ; the one you’ll probably be interested in is the HTTP archiver since most systems can take HTTP POSTs.

 

Don’t invest a lot of time in Esmond (which is PostgreSQL and Cassandra underneath); it’s going away in the 5.0 release.  There will be backward-compatibility software to query its replacement as if it were Esmond, but that won’t remain forever, either.

 

--Mark