Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[roseway]

I suspect that this is something to do with file/directory permissions, because the webserver works fine for me on both the RPi model B and the RPi2. I haven't been able to get to the bottom of it yet, and I've had to set it aside while I attend to other issues. In the meantime, jid has posted a workaround here: http://forum.kitz.co.uk/index.php/topic,14844.0.html

[ktz392837]

I suspect that this is something to do with file/directory permissions, because the webserver works fine for me on both the RPi model B and the RPi2. I haven't been able to get to the bottom of it yet, and I've had to set it aside while I attend to other issues. In the meantime, jid has posted a workaround here: http://forum.kitz.co.uk/index.php/topic,14844.0.html

Do any other RPi2 users have any tips what could be causing this. 

It appears dslstats does not work with the webserver and it does not allow connection to MyDSLWebStats also (at least on my RPi2).  I click the Check Now button and it says Unable to check at present regardless of what is in the username and key fields.  The path to curl is also correct.  Could it be specific to a Curl version?  What port does MyDSLWebStats use to upload the data? Thanks.

[loonylion]

I click the Check Now button and it says Unable to check at present regardless of what is in the username and key fields.  The path to curl is also correct.  Could it be specific to a Curl version?  What port does MyDSLWebStats use to upload the data? Thanks.

I had that and it turned out to be something silly but I dont quite remember what. It may have been related to needing to start/stop recording.

[roseway]

The reason I suspect that it's a permissions problem is that both the webserver and MDWS do work, on both the original RPi model B and the RPi2. In the latter case, this is with a vanilla Raspbian installation, so I know that the version of curl is OK.

If you extracted the files from the downloaded zip file using Windows, there may be an issue with permissions. To avoid this I suggest that you copy the downloaded zip file to the Pi user's home directory (/home/pi) and extract the files using the command line utility unzip. Run the program from there as an ordinary user.

[ktz392837]

Thanks I will do some more testing.  I wget and unzipped on the pi itself so it should have kept permissions correct.  Using the standard raspbian image feb 2015 and dslstats 5.4.  Very strange.

[ktz392837]

Is it possible to disable all local graph/text file creations and just use MDWS? Or does dslstats do calculations on the data it uploads to MDWS? Thanks

[roseway]

No, it's not possible. I could probably make it an option, but it would involve quite a lot of work, and it would be on the end of rather a long todo list.

Most of the data uploaded to MDWS is 'delta' data, i.e. it's the difference between successive samples.

[NewtronStar]

I have noticed since the 5.4.3 release the MAX/Attainable down stream rate in the telnet data connection stats becomes stuck this is happening daily only on the PC version the RPi is fine.

The only way to get the correct attainable is to turn off DSLstats and restart DSLstats on the PC.

[roseway]

I've responded to your other report in the v5.4.3 thread.

[boost]

Howdy!
Hopefully you've recovered from your birthday?

1. Dunno if it's been mentioned but DSLstats appears to be sending all power values as positive to MDWS. My aggregate upstream is -13 odd I think.
 
2. Some time ago HLOG changed slightly. I think you mentioned elsewhere you only plot the tones in use as opposed to the full HLOG values. Might it be possible to have the option for both or just revert to full?

I may have misunderstood tbh but the graphs definitely changed to gaps instead of dips. Dips are much more useful for noobs like me

Keep up the excellent work. Please.

[roseway]

Thank you, I have recovered

1. This was raised elsewhere recently. I'd never seen negative aggregate values personally, so I was wrongly assuming that these values would always be positive. It's now corrected for the next release.

2. The gaps in HLog (and also in QLN) are not directly to do with unused tones, but are sections of the graph where the reported values are not true. These values are -160 dBm/Hz for QLN and -96 dB for HLog. Somewhere in a specification it says that these values are, in effect, dummy values which the modem uses when it doesn't have a proper value to report. Plotting these values wouldn't add any more useful information, but perhaps I could plot them in dotted lines. I'll have a look at this.

[boost]

Gotcha, thanks!

The JDSU docs suggest any irregularities in the HLOG are copper faults. 'Fault' is relative I guess so whether that's worth highlighting or not I dunno

[roseway]

The sections we're talking about are not irregularities in that sense.

[boost]

Are you 100% sure?

[roseway]

Yes. Although the judgement DSLstats makes is based on the baseline values I mentioned, the tones in question are also the tones which are not included in the band plan. Although I don't know the details of how these figures are determined by the modem, I've no doubt that it's quite normal for these tones to show baseline values in the QLN and HLog graphs.