6 Aug
2024
6 Aug
'24
6:55 p.m.
Try r13609!
adam.
Corwin Ziegler Hunts via observium wrote on 2024-08-05 20:13:
These sensors show up as power sensors rather than as dBm (or SNR).
Could the "autoscale" button simply toggle whether the "--alt-autoscale" option is set? With the default state of the toggle coming from the current logic?
On Wed, Jul 31, 2024 at 8:09 PM Adam Armstrong <adama@observium.org mailto:adama@observium.org> wrote:
I can't think of any really good ways to handle this. The way the graphs are generated doesn't make having a per-type autoscale setting easy. The "autoscale" button in the UI forces it on for those graphs where it's normally off. You could try removing the 'snr' from the case statement setting the scale min/max from html/includes/graphs/sensor/graph.inc.php and see if it improves rendering for these graphs (without breaking the rest of the snr graphs). If that doesn't improve it, perhaps we need an explicit toggle to turn the autoscale off, though the logic for the autoscale decision code is already pretty messy and confusing. adam. Corwin Ziegler Hunts via observium wrote on 2024-07-27 00:54:Is there any guidance towards resolving these issues? Manually logging in to the Observium server to generate graphs every time someone needs to investigate these SFP light levels is disruptive and time consuming. Just being able to prevent the broken "--alt-autoscale" option from being passed to rrd from within the web UI (issue #2) would help significantly. We have a Professional subscription. Corwin On Tue, Jun 25, 2024 at 2:03 PM Corwin Ziegler Hunts <corwin.ziegler.hunts@nextlevel.net <mailto:corwin.ziegler.hunts@nextlevel.net>> wrote: Pictures attached. On Tue, Jun 25, 2024 at 11:47 AM Adam Armstrong via observium <observium@lists.observium.org <mailto:observium@lists.observium.org>> wrote: so many words, but no pictures. Corwin Ziegler Hunts via observium wrote on 2024-06-25 19:13:Hi Observium team, We're having issues with some of the generated graphs. Some of our devices report the light level received by SFPs using mW (rather than dBm). We are on the latest stable revision (13582). Issue 1: The graphs generated by Observium for these sensors usually show an excessive scale of at least -100 mW to 100 mW when the data in the specified time interval are entirely within the range 100 nW to 1 mW and usually only a small chunk of that. The calculated minimum and maximum values shown below the graph are correct. These graphs worked properly in the past with a prior version of Observium. Issue 2: Selecting and deselecting the Force Autoscale option appears to have no effect on the generated graph or displayed rrd graph command. Manually running the generated rrd graph command after removing the "--alt-autoscale" option results in a good graph. Here is the raw SNMP output for one of the these sensors showing a current value of 0.5560mW: ENTITY-SENSOR-MIB::entPhySensorType.1000004631 = INTEGER: watts(6) ENTITY-SENSOR-MIB::entPhySensorScale.1000004631 = INTEGER: milli(8) ENTITY-SENSOR-MIB::entPhySensorPrecision.1000004631 = INTEGER: 4 ENTITY-SENSOR-MIB::entPhySensorValue.1000004631 = INTEGER: 5560 ENTITY-SENSOR-MIB::entPhySensorOperStatus.1000004631 = INTEGER: ok(1) Corwin _______________________________________________ observium mailing list --observium@lists.observium.org <mailto:observium@lists.observium.org> To unsubscribe send an email toobservium-leave@lists.observium.org <mailto:observium-leave@lists.observium.org>_______________________________________________ observium mailing list -- observium@lists.observium.org <mailto:observium@lists.observium.org> To unsubscribe send an email to observium-leave@lists.observium.org <mailto:observium-leave@lists.observium.org> _______________________________________________ observium mailing list --observium@lists.observium.org <mailto:observium@lists.observium.org> To unsubscribe send an email toobservium-leave@lists.observium.org <mailto:observium-leave@lists.observium.org>
observium mailing list -- observium@lists.observium.org To unsubscribe send an email to observium-leave@lists.observium.org