Welcome to the Cumulus Support forum.

Latest Cumulus MX V3 release 3.28.6 (build 3283) - 21 March 2024

Cumulus MX V4 beta test release 4.0.0 (build 4019) - 03 April 2024

Legacy Cumulus 1 release 1.9.4 (build 1099) - 28 November 2014
(a patch is available for 1.9.4 build 1099 that extends the date range of drop-down menus to 2030)

Download the Software (Cumulus MX / Cumulus 1 and other related items) from the Wiki

CMX crash and realtime retain failure

From build 3044 the development baton passed to Mark Crossley. Mark has been responsible for all the Builds since. He has made the code available on GitHub. It is Mark's hope that others will join in this development, but at the very least he welcomes your ideas for future developments (see Cumulus MX Development suggestions).

Moderator: mcrossley

User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

Well, that's not right at all! :bash:

Something has obviously gone wrong at some point. I'm reworking the MySQL stuff at the moment, so I'll take a look at getting it fixed tomorrow...
User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

Hang on, the problem is purely in the help text. You should just try entering the text as "value unit". So for example "10 DAY"

See: https://dev.mysql.com/doc/refman/8.0/en ... -intervals
User avatar
HansR
Posts: 5964
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bookworm
Location: Wagenborgen (NL)
Contact:

Re: CMX crash and realtime retain failure

Post by HansR »

:lol: Indeed. Fooled by the help text. Room for improvement here. ;)
Thanks!

I suggest to give just the number (of days) and have days implicit. Nobody is going to retain more than a month of realtime unless you would want stress the limits of your database. And even then, you could give a 1000 days, 10.000 or whatever (would that be possible?). No formatting of (partial) SQL statements must be left to the user. Users are not programmers, they are users. CMX is being/becoming too much a programmers program iso of a meteo program.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

The problem is migration. People will already have configured that setting previously and may have 2 WEEK or 1 MONTH set for instance :groan:

I think you may be surprised how much real time data some people are retaining! :shock:
Users are not programmers, they are users. CMX is being/becoming too much a programmers program iso of a meteo program.
Yes, with hindsight it may have been better to to have a number of days field, but that isn't what was set up originally by Steve years ago.

I'm not sure I agree with that statement. I have been trying to simplify the settings screens by removing irrelevant settings, and removing the need for users to edit Cumulus.ini to alter settings, so I would argue the opposite case.
freddie
Posts: 2475
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2 + Ecowitt
Operating System: GNU/Linux Ubuntu 22.04 LXC
Location: Alcaston, Shropshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by freddie »

mcrossley wrote: Mon 12 Apr 2021 8:55 am The problem is migration. People will already have configured that setting previously and may have 2 WEEK or 1 MONTH set for instance :groan:

I think you may be surprised how much real time data some people are retaining! :shock:
Users are not programmers, they are users. CMX is being/becoming too much a programmers program iso of a meteo program.
Yes, with hindsight it may have been better to to have a number of days field, but that isn't what was set up originally by Steve years ago.

I'm not sure I agree with that statement. I have been trying to simplify the settings screens by removing irrelevant settings, and removing the need for users to edit Cumulus.ini to alter settings, so I would argue the opposite case.
I would argue the opposite case too - you have done a lot of work to make settings user-oriented.

Regarding the setting in question - how about a drop-down menu with some choices, with the translation to the SQL equivalent as the submitted values?
Freddie
Image
User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

freddie wrote: Mon 12 Apr 2021 9:11 am Regarding the setting in question - how about a drop-down menu with some choices, with the translation to the SQL equivalent as the submitted values?
How about...
realtime.jpg
You do not have the required permissions to view the files attached to this post.
User avatar
HansR
Posts: 5964
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bookworm
Location: Wagenborgen (NL)
Contact:

Re: CMX crash and realtime retain failure

Post by HansR »

@mcrossley, @freddie:
mcrossley wrote: Mon 12 Apr 2021 8:55 am The problem is migration. People will already have configured that setting previously and may have 2 WEEK or 1 MONTH set for instance :groan:

I think you may be surprised how much real time data some people are retaining! :shock:
Any numbers and a reason why?
mcrossley wrote: Mon 12 Apr 2021 8:55 am
Users are not programmers, they are users. CMX is being/becoming too much a programmers program iso of a meteo program.
Yes, with hindsight it may have been better to to have a number of days field, but that isn't what was set up originally by Steve years ago.
OK. Maybe freddie's suggestion may help here. I could live with that.
mcrossley wrote: Mon 12 Apr 2021 8:55 am I'm not sure I agree with that statement. I have been trying to simplify the settings screens by removing irrelevant settings, and removing the need for users to edit Cumulus.ini to alter settings, so I would argue the opposite case.
I see what you have done with the settings and I think that is a great action (like so many other actions). My remark comes not so much from this specific setting as well as from the fact that all other SQL is actually left to the user. The user is invited (at least not discouraged) to modify all kinds of code around - and within - CMX creating issues of its own (is there a repository of SQL queries in use?).

Anyway, further irrelevant probably, so just take it as a loose remark ;)

EDIT: Trying to post this reply I saw your solution. Seems great to me.
Btw: those fields are always below each other, is it possible to have them beside each other?
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
freddie
Posts: 2475
Joined: Wed 08 Jun 2011 11:19 am
Weather Station: Davis Vantage Pro 2 + Ecowitt
Operating System: GNU/Linux Ubuntu 22.04 LXC
Location: Alcaston, Shropshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by freddie »

mcrossley wrote: Mon 12 Apr 2021 9:56 am
freddie wrote: Mon 12 Apr 2021 9:11 am Regarding the setting in question - how about a drop-down menu with some choices, with the translation to the SQL equivalent as the submitted values?
How about...
realtime.jpg
Perfect :)
Freddie
Image
User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

HansR wrote: Mon 12 Apr 2021 10:00 am
Btw: those fields are always below each other, is it possible to have them beside each other?
Sadly no, there is a layout option for side-by-side, but it is an all or nothing option, it has to apply to the whole page. I have made a request that it be supported on a sub section basis, as have other people, but no response :(
User avatar
mcrossley
Posts: 12763
Joined: Thu 07 Jan 2010 9:44 pm
Weather Station: Davis VP2/WLL
Operating System: Bullseye Lite rPi
Location: Wilmslow, Cheshire, UK
Contact:

Re: CMX crash and realtime retain failure

Post by mcrossley »

HansR wrote: Mon 12 Apr 2021 10:00 am My remark comes not so much from this specific setting as well as from the fact that all other SQL is actually left to the user. The user is invited (at least not discouraged) to modify all kinds of code around - and within - CMX creating issues of its own (is there a repository of SQL queries in use?).
Originally because the MySQL is not something Cumulus MX is "interested" in, like MQTT etc is is something that it can provide data for, but has no interest in reading back. If MX used it as a storage resource then yes it should a fixed schema to ensure compatibility.

MX provides a suggested schema, which probably the vast majority of SQL users will use and that is fine.

But if people want to freewheel they can - I do, I store additional data that is of interest and useful to me. I have also added triggers, stored procedures and additional tables to drive my web site. I guess my view is if you are competent enough to stray away from the default you are knowledgeable enough to sort out any consequences, or live with them ;)

For me its a balance, you are proscriptive with settings and reduce flexibility, or allow have defaults but allow customisation and increase flexibility.
User avatar
HansR
Posts: 5964
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bookworm
Location: Wagenborgen (NL)
Contact:

Re: CMX crash and realtime retain failure

Post by HansR »

mcrossley wrote: Mon 12 Apr 2021 10:34 am But if people want to freewheel they can -
Yes. And that is great and good. My own coding started out there as well. No problem.

But let's be honest: there is e.g. a lot of code which apparently is based on SQL (PHP procedures with implicit SQL) and there is a lot of variability in there. I think those queries and that structure could/should become part of Core-CMX. The procedures around it as well. If it were only those things were assembled in some repository (or is it already? I could have missed it) where at least the basis is maintained / stabilised. The same holds for javascript.

It's the divergence which disturbs me.
mcrossley wrote: Mon 12 Apr 2021 10:34 am For me its a balance, you are proscriptive with settings and reduce flexibility, or allow have defaults but allow customisation and increase flexibility.
CMX is at the basis and has by definition a lot of divergence because of what you want to serve.
CUtils for a data presentation needs convergence. That is a big difference.

I don't think we differ very much though you bend more to one side I more to the other.
The difference looks big, but the feet are close.

If CMX had not been so limitless (which it isn't), CUtils had not become so limiting (which btw it isn't either).

But anyway, you never know, maybe sometime a nice discussion over a pint or two on applications, how to setup things, users and developers :groan:
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
Post Reply