What "TYPE" of data base do I need?

  1. what OS are you running on the NUC?
  2. have you determined if SQLITE is installed on that system?
VERSION="18.04 LTS (Bionic Beaver)"
PRETTY_NAME="Ubuntu 18.04 LTS"

Do I need to have it installed outside NR?

Looked at this site/link:


sudo apt-get install php5-sqlite
Reading package lists... Done
Building dependency tree       
Reading state information... Done
E: Unable to locate package php5-sqlite

Yes, it must be installed. It is not part of node-red - the node-red-node-sqlite interfaces only to SQLite. If SQLite is not installed, it will not work.

Well, ok, I need it installed.

But looking at the above, it doesn't seem to want to be installed. :frowning:

I have tried to install the squirrel server on Ubuntu and not getting anywhere.

(Given this is only a "test bed" before I do it on the RPI, it seems painful and slow on THIS machine. I shudder to think what it entails on the RPI.)

So is there a way I can make a "database" by simply writing data entry logs to a text file and parsing that?

Why on earth are you trying to do that?

Because somewhere here, I was told I need a squirrel thing installed outside NR to be able to run it.

So I blindly am trying. (Though I am kind of tied up with a couple of other things just now.)

Ok, it was SQLite. But I don't know the difference.

Trying to build opencv on a RPI is one. NOT fun!

I think you need a time series database, try InfluxDB

Well if you don't know the difference then you should do your research first.

Do you want opencv? If not why are you trying to build it?


You are getting things out of context.

I was asking about a database and SQL was mentioned (from memory).
Following that I saw reference to SQL and PHP - or what ever it is/was.
That added a whole new layer to what I had to install on the RPI - if ever it was going to be accepted/used as "the" database.

At this point I am doing all testing on a NUC which is running UBUNTU.

On another ENTIRELY DIFFERENT thing, I am tasked with getting OPENCV installed/working on a RPI with Jessie. NOT STRETCH.

This entails downloading a lot of stuff and compiling it. Something I am not good with doing - confidence wise.

It had taken me about 4 attempts to then and while posting it was the FIFTH attempt, and it was taking about 16 hours.

I was not in the mood for installing a heap of other stuff like PHP (or what ever it was shown with SQL.)


I am looking at influxDB, but not getting too far with it as I can't see a working example - from memory.

Since then I am writing lines of text to a file with a time stamp and the data. (Well I have all the nodes there and just missing the last part where they actually write to the file, as the hardware isn't ready yet anyway.)

So the OpenCV is a completely different horse to the database problem.

BTW, I still haven't got OpenCV working. It failed the compile. AARRGGH!

I think it would help us all if, when you open a thread, you keep to that topic and don't leap to something different in the middle. You opened this thread requesting help with a database and suddenly you are talking about opencv - which is not a database. @Colin was not getting things out of context, you are.

Please limit a thread to one topic. People are giving their time trying to help you so don't make it hard for them or they may choose not to help...

And it sounds like you have a solution to saving the data since you say you are writting it to a file, congratulations.


Yes, ok.

But I was asking about SQL and was told that I not only needed to install SQL in NR but on the machine.

That took me down yet another rabbit hole and PHP somehow came into being needed.

That confused the &*&^** out of me.

On top of trying to get OpenCV working because there is no "download and install" for Jessie.
Stretch seems to have that though.

But back to "Topic"...... Yeah, but when I am told that I have to install x, y, z, a, b, c and the rest of the family for simply one thing, it really throws a spanner in the works.

You were advised twice that you didn't need php for admin, but hey, if you then decide to use it then yes you will need the prerequisites and the learning curve that goes with it.

And that is when it falls on you to take things one at a time and to learn AND use the appropriate terminology. It also doesn't help others when you seem to jump from working on one machine and then another.

Lets take the suggestion by Nick about using sqlite.

First think you should have done is a google search to see if it was installed on the deice you are using. If it turned out not to be installed, then you needed to do a google search to find out how to install it. There are plenty of tutorials and how to's out there that you could follow. If you can't figure out how to follow them,you need to ask the author of the tutorial for help.

Next you will have to learn how to use SQLite - again you go to google and search and find some tutorials on using sqlite AND you try them out - and nonw ofthis involves Node Red yet. This is doing/learning the groundwork for what you want to do in NR.

Now once you have installed SQLite and created a database and created a table and inserted, modified and deleted records into it - now you are ready to install the node-red-node-sqlite into a test flow and try creating a table, inserting, updating and deleting rows.

After all that, now you can go use it in your flow.

Remember you have to learn to walk before you can run. You can't start using calculus without knowing basic math and it takes time and a lot of effort to learn things