NetAlertX -Internet Speedtest

Ja, schwarze Schrift auf dunkelgrün ist nicht so der Hit. Wenn Du die Events History anklickst, bekommst Du Download und Upload Daten in willkürlicher Reihenfolge ohne farbigen Hintergrund.
Was hast Du jetzt als cron schedule eingestellt? */5 * * * * ? Könntest Du mal 0 */3 * * * für einen Tag laufen lassen?
 
1728836188312.png
und
1728836249857.png

- also selbes Ergebnis wie bei Dir :(. Vermutlich bekommst Du mit
Code:
[0-10] */3 * * *
wenigstens ein erfolgreiches Messergebnis denn mit einer 5 minuetlichen Messung kommen ja fast immer Ergebnisse - jedenfalls bei mir.

Keine Ahnung wie der Code misst. Offensichtlich muss er sich erst "warmlaufen" :oops:
 
Jetzt wollte ich den von mir vorgeschlagenen crontab Eintarg save und wieder die Loop :mad: Dann habe ich den Container removed und wieder mit docker compose erstellt und jetzt gibt es eine Start Loop :mad:

Ausserdem habe ich jetzt wieder zu meinem Desktop gewechselt und festgestellt dass die gestrigen Settings mit 5 Minuten dort aktiv sind obwohl ich die mit meinem Lappi auf 3 Stunden geaendert habe :mad:

Das Tool sieht fancy aus - das UI ist chaotisch - Settings werden offensichtlich lokal gespeichert (cache loeschen half nicht) ... das Tool zeigt ein merkwuerdiges Verhalten beim Speedcheck ...

Es lohnt sich meiner Meinung nach nicht Zeit in das Tool zu stecken. Es ist zu buggy. Ich stelle deshalb mein System jetzt nicht noch mal neu aus dem vorherigen Backup hin.

Aber wenn Du Lust und Zeit hast kannst Du ja mal meinen Vorschlag aufgreifen. Vielleicht hast Du ja Erfolg. Ich druecke Dir die Daumen 🫰
 
jetzt gibt es eine Start Loop
Die app.conf ist vermutlich beibehalten worden. Die muß man dann manuell bearbeiten.

Was das Internet speedtest plugin angeht
Solange nur der Minutenteil des cron ausgefüllt ist klappt es, sowie man anfägt */3 oder ähnliches bei den Stunden einzutragen, fängt es leider an merkwürdig zu werden. Ich gestehe, daß auch das Plugin nicht verstehe, was es genau da treibt, dafür reichen meine Python Kenntnisse nicht aus.

Wahrscheinlich lasse ich das Tool einfach mitlaufen, ohne den speedtest und schaue gelegentlich mal, ob inzwischen mal funktioniert.

Auf alle Fälle danke ich Dir, daß Du da auch Zeit hineingesteckt hast, hätte ja sein können, daß ich da etwas falsch mache, aber wenn Du in den gleichen Fehler hineinläufst, dann ist wohl eher das Plugin das Problem.
 
Auf alle Fälle danke ich Dir, daß Du da auch Zeit hineingesteckt hast, hätte ja sein können, daß ich da etwas falsch mache, aber wenn Du in den gleichen Fehler hineinläufst, dann ist wohl eher das Plugin das Problem.
Habe ich gerne gemacht denn es war auch Eigeninteresse dabei denn die Beschreibung klingt schon gut und vielleicht haette ich es bei mir auch eingesetzt :) Aber fuer mich ist das Tool einfach zu buggy.

Aber probiere doch mal meinen Vorschlag aus :)
 
Bei dem Versuch Deinen Vorschlag auszuprobieren ist mir das Ganze erst mal wieder im Loop gelandet :), mußte erst mal die app.conf wieder editieren, um das Ganze wieder ans Laufen zu bekommen.
 
Dear @jokob-sk thank you very much for joining the discussion on your tool and the interest in its continued development.
Basically, we have an issue with the internet speedtest plugin INTRSPD, which appears to function, when run “once”, started manually, or started with every fresh network scan (i.e. carrying out a speed test every minute or every five minutes), but it does not work consistently when run with a schedule, i.e. “0 */3 * * *”, “0 */6 * * *”, “0 4 * * *”, “59 23 * * *”, “0-1 */3 * * *” or similar. The result in these cases is mostly the error code -1 -1 as foreseen by the python script executed in case of failure.
I will look into filing an issue on the git later (I don’t have a git account).
 
Zuletzt bearbeitet:
Also, noticed the comment about the readability of the speedtest results. I'm not the best designer so if someone can help me out with the Css side of things that would help. 😅

Vielen dank wenn jemand helfen kann🙏
 
I have stopped the „production branch“-container and deployed a new container for the „development branch“. Currently I have set INTRSPD to 0 */2 * * *, I‘ll report back how this behaves.
 
Zuletzt bearbeitet:
the Css side of things
In the style sheet you might be using somewhere an entry which indicates the color hex as follows

div style=“background-color:#05483C”
IMG_1945.jpeg

IMG_1946.jpeg

This is an extremely dark green, where it is basically impossible (at least for me) to read a black number written into that coloured field. If you want to stay with that marine bluish green colour style, the colour tint would be better a bit lighter, such as #82a39d or #9bb5b1
Alternatively, one could consider writing the number in white letters.
 
Further feedback to @jokob-sk: thus far, the internet speedtest plugin of the development branch appears to work:
IMG_1951.jpeg
Great! Thanks a lot for investing the time to make it work!
The display page only shows a green bar for the Upload, and nothing for the Download (when the -1 -1 result was displayed, it gave two red bars per run, one for upload, one for download). But maybe this is the desired design.
Where I see a somewhat technical mistake is in the events history: first, it is not ordered by event time, second, it appears to work accumulative: after the first run, there one entry for run 1, after the second there are two entries for run 1 and one for run 2. After the third run, there are three entries for run 1, two entries for run 2 and one entry for run three. This keeps increasing with every run, to me it appears to follow a triangular number sequence [n*(n+1)/2]:
IMG_1952.jpeg
> Edit: the sixth run at 18:00 also sucessfully went to completion, and now the Events History for the speedtest has 21 entries. So the above formula seems confirmed.
I am not sure if this the intended design; if so, personally I would have preferred a simple ordered list.
A further odd behaviour which I observe sometimes is the following, this happens for example when I am on the internet speedtest Events History and click on the GUI refresh button. A full browser page reload usually fixes this:
IMG_1947.jpeg
My final observation I believe is already known to you, this is a Safari issue, occurring both on iOS/iPadOS devices and on MacOS devices. In the former case changing the browser does not help, because they are all built on Apple’s webkit, on MacOS devices changing to Firefox sorts it. It must have to do with how Safari handles the display of the “Network Overview”. All labels which usually are placed where they belong at the ends of the yellow lines are stacked one over the other in the upper left corner (here I only show the result of the current dev version, I have not yet sorted the things to their right places as I had done in the production branch):
IMG_1953.jpeg
 
Zuletzt bearbeitet:
Hi @Stationary,

Thanks for that! I adjusted the CSS and this should be available in the dev image in a few minutes or in the next release.

The download values should show up - it might be a caching issue - or you can try deleting existing entries:

1729288706810.png

The Safari one is a known bug (there are a few https://github.com/jokob-sk/NetAlertX/issues?q=is:issue+is:closed+safari), but I don't own any Mac devices so I rely on the community to fix this. I'm already struggling to support the app as it is 😅 (I wish this would be my job well, it's not bringing in money - maybe in the future - man kann auch hoffen).

The following falls also in the above category - too hard to fix now - don't have that much time:

> A further odd behaviour which I observe sometimes is the following, this happens for example when I am on the internet speedtest Events History and click on the GUI refresh button. A full browser page reload usually fixes this...

I really want to rewrite the whole section to use datatables...

> Where I see a somewhat technical mistake is in the events history: first, it is not ordered by event time, second, it appears to work accumulative: after the first run, there one entry for run 1, after the second there are two entries for run 1 and one for run 2. After the third run, there are three entries for run 1, two entries for run 2 and one entry for run three. This keeps increasing with every run, to me it appears to follow a triangular number sequence [n*(n+1)/2]:

So the events history tracks changes to an object, in this case the objects speed (1 entry in the Objects view = Object) is considered a change. So I think the [n*(n+1)/2] pattern is expected. I know it's a bit confusing, but hopefully I get to refactor the section in the future where some of these things get addressed.

If that's ok, I will consider the main issues in this thread as solved - the background color and the occasional `-1` results. In future, if you can, please open an issue on the GitHub project - I am checking google search results sometimes, but it would help me a lot if it would be centralized 🙏 Other NAX community members can help out too - or I'm also trying to start a Discord community (https://discord.com/invite/NczTUTWyRr) - if that's better.

Vielen Dank fur die Hilfe 👍 - Hope to see you on GitHub or Discord 😉

Also, thank you for switching to English, I hope I wasn't imposing in a German speaking Forum
 
Dear @jokob-sk that was already more than I could ever have hoped for in such short period of time! It is really amazing that you took the time to fix the issues nearly immediately.
IMG_1954.jpegSwitching to English for me was not a problem (I do 80% of my daily business in languages other than German), but yes, it might exclude some other contributors. I promise that I’ll get a github account and that I’ll log any future issue in the place indicated by you.
So thank you very much again!
What precisely do you wish a Fritzbox plugin to do? Access the AHA?
 
Zuletzt bearbeitet:

Zurzeit aktive Besucher

Letzte Anleitungen

Statistik des Forums

Themen
5.880
Beiträge
57.443
Mitglieder
5.813
Neuestes Mitglied
Blacksnofru
Zurück
Oben