Closed philcanman closed 1 week ago
hi @philcanman ,
would the elbencho --livecsv /path/to/some/output.csv
option work for you in combination with csv file import option of influx as described here?
https://docs.influxdata.com/influxdb/cloud/write-data/developer-tools/csv/#example-write-command
The live CSV file in its current form will not work with the CSV tool built into Influx as I understand it. Is there a way to output the live CSV data to the terminal with each update on a new line with fixed columns? Then I can pipe that data off to a script for real time metrics.
Would also be good with a timestamps in the livecsv file. Then I could write something to import the livecsv data into influx.
Ok, no problem to add an option to have the CSV lines as console output. (The side effect will be that the human-readable end result summary will no longer get printed to the console, because the csv reader into which you pipe the output wouldn't like that non-CSV tail - but I assume that's not a problem here.)
When you say "fixed columns", are you referring to fixed column width or am I getting this wrong?
Adding timestamps will also be no problem.
Awesome glad to hear that. When I say fixed columns I mean the header or column value never changes. For instance if the third column is MB/s it is always MB/s kinda deal. That and a time stamp should make writing a script to import the data a breeze. Thanks!
Hi @philcanman ,
the new elbencho version v3.0.19 (released a few hours ago) supports a new special value "stdout" (for elbencho --livecsv stdout
) to print the stats in csv format to the console. The MiB/s column in integer format is sometimes not fine-grained enough and thus the values of this column might get decimal places at some point in the future, but the column unit MiB/s will stay the same, so I assume this still counts as "fixed columns".
Also, the live csv stats now have a new column for "ISO Date" timestamps (as you requested), so the columns did actually change in this release, but hopefully for the last time :-)
If you create any script or guide of which you think it would be good to get hosted in the elbencho repo (under the contrib/
subdir), so that others can also benefit from it, then of course please just let me know.
Thanks a ton for this. I have it working now using graphite see repo: https://github.com/philcanman/elbencho_graphite
I am wanting to capture the data in a time series database so that I can visualize the data in Grafana.