Selective port Polling

I read over selective polling in the docs and tried

  librenms@librenms:~/scripts$ ./collect-port-polling.php -e
Full Polling: 6 14 15 19 17 20 7 12 11 22 8 23 2 3 4 5 21 1
Selective Polling: 6 14 15 19 17 20 7 12 11 22 8 23 2 3 4 5 21 1

| device_id | os          | port_count | inactive_ratio |  full_time | selective_time |     diff |  diff |   set |
|         6 | calix       |         58 |          0.759 |   213.012s |        44.880s | -168.132s |  -79% |  none |
|        14 | calix       |        117 |          0.991 |   522.563s |       109.918s | -412.645s |  -79% |  none |
|        15 | calix       |        122 |          0.992 |   335.101s |        59.467s | -275.634s |  -82% |  none |
|        19 | calix       |        116 |          0.991 |   265.848s |        25.178s | -240.669s |  -91% |  none |
|        17 | calix       |        117 |          0.974 |   225.223s |        63.146s | -162.076s |  -72% |  none |
|        20 | calix       |        119 |          0.739 |   397.173s |       105.489s | -291.684s |  -73% |  none |
|         7 | calix       |         58 |          0.569 |   185.101s |       109.170s | -75.931s |  -41% |  none |
|        12 | calix       |         58 |          0.655 |   224.043s |        62.294s | -161.749s |  -72% |  none |
|        11 | calix       |        116 |          0.948 |  1129.685s |       118.248s | -1011.438s |  -90% |  none |
|        22 | generic     |       1820 |          0.986 |   844.321s |         3.868s | -840.453s | -100% |  none |
|         8 | slms        |        142 |          0.937 |    46.215s |         0.683s | -45.532s |  -99% |  none |
|        23 | slms        |        752 |          0.912 |   753.273s |        50.338s | -702.935s |  -93% |  none |
|         2 | timos       |         66 |          0.288 |     0.781s |         1.060s |  +0.278s |  +36% |  none |
|         3 | timos       |         35 |          0.657 |     0.503s |         0.468s |  -0.035s |   -7% |  none |
|         4 | timos       |         20 |          0.400 |     0.473s |         0.403s |  -0.069s |  -15% |  none |
|         5 | timos       |         52 |          0.692 |     0.968s |         0.750s |  -0.218s |  -23% |  none |
|        21 | generic     |         11 |          1.000 |   144.623s |       132.559s | -12.063s |   -8% |  none |
|         1 | linux       |          3 |          0.333 |     0.113s |         0.067s |  -0.046s |  -41% |  none |
|   Totals: |             |       3782 |          0.768 |  5289.018s |       887.986s | -4401.032s |  -83% |     0 |

I’m confused on the percentage part. I thought the script would match the percentages wtih what it discovered to be the differences. How do I use this script?

I’m not sure what you mean by this.

However from those results, you can enable selected port polling globally and you will see a significant difference.

Running ./scripts/collect-port-polling.php will poll your devices with both full and selective polling, display a table with the difference and optionally enable or disable selected ports.

I was assuming that running the script would enable selective port polling based off the diff results if using the -e option.

Thank You for your help!