Robo-Tank v6.0 is Ready - Now v6.5

[gandalf]

I'm unable to reproduce, this night the dosing was ok (i'm using the "fixed" version but i didn't see any retry in the log)

No issue for the extensions. Shoild I need a PWM, i'll use the native one and not the extended.

[Rob F]

As this happened it might be good to add that reboot as I really don't think that fix will do anything if it happens again. If you want to leave the fix maybe add a 2nd fix after with the reboot so if it fails that happens. You can call the systemSettings.restart() function. Also have it log before the reset so you know it happened.

[gandalf]

sounds good.

[Craig]

Hey mate long time, Have just installed latest version on pi 4(all I could actually get my hands on)

Install was problem free, detected probe and pH circuit straight away. I did install Raspberry OS through Parallels and windows 11( to try and avoid the issues I had last time if you can remember)

Because its a pi 4 I would like to have a fan going, what voltage are the fan pins on the robotank board and do they run while powered up? or do they run of pi board temperature.

I was going to run dosing once installed but might just run a remote doser with a rule for powering it down if the pH goes wonky.


Anyway mainly Thanks for your efforts and this version seems to be working great.

IS there a

[Rob F]

Hey Craig, great, thanks for testing it out. I think the issues you had last time was caused by me, a few packages didn't install so well but it looks like it might be finally sorted out. If you come across any issues let me know.

The output on the fan plug is the same as the controller input, probably 12v. The port is controlled via a pin on the pca9685 so you need to setup a connector in reef-pi using PCA9685 driver and pin 5 assuming you have v3 controller. You can add it as a jack or connector as you'll likely need to run at 100% 24/7. If you want it to turn on when the controller is powered you can use the + positive pin on the fan plug and run the ground to a common ground point, maybe on a temp port, you can solder it internally.

Not sure, but looks like your remaining message was cut off.

[Craig]

Cool, Actually had problem activating dc ato pump, so went back to version 1 controller(swapped sd card with Robo Tank into version 1) still no joy so went back to reef pi and still no joy, my multimeter is at work and have covid in the house so can't go and pick it up. But assuming pump is kaput.

One other issue I do have though is that 2 of my ac outlets won't activate, I have the 1 outlet version, works fine on reef pi but not on robo tank any ideas? tried both ac bar ports on the controller but no joy

Other than that I have it on my tank and looks great, won't need another outlet for a while until it gets colder and need another heater as back up and also when I get round to getting my doser back up and running. but still working out dosages with the tank anyway

Thanks again mate

---

Oh and one other thing anyway to change temperature probe names? Too a bit of mucking round and had to change what probes do due to length of probe. Oh and 1 probe I can't set a rule for as it comes up as a float switch instead of a temp probe. Reads ok but can't get it to be used to control heaters

[gandalf]

As this happened it might be good to add that reboot as I really don't think that fix will do anything if it happens again. If you want to leave the fix maybe add a 2nd fix after with the reboot so if it fails that happens. You can call the systemSettings.restart() function. Also have it log before the reset so you know it happened.

but there is an issue. with the forced reboot, if you have an issue with the i2c bus not fixed by the reboot, you will Stick in a boot loop. every time the server boots, the controller starts, write fails and a reboot is made and so on, unlimited.

this makes very hard to connect via SSH because the server is rebooting cintinuosly

[Rob F]

Cool, Actually had problem activating dc ato pump, so went back to version 1 controller(swapped sd card with Robo Tank into version 1) still no joy so went back to reef pi and still no joy, my multimeter is at work and have covid in the house so can't go and pick it up. But assuming pump is kaput.

One other issue I do have though is that 2 of my ac outlets won't activate, I have the 1 outlet version, works fine on reef pi but not on robo tank any ideas? tried both ac bar ports on the controller but no joy

Other than that I have it on my tank and looks great, won't need another outlet for a while until it gets colder and need another heater as back up and also when I get round to getting my doser back up and running. but still working out dosages with the tank anyway

Thanks again mate

---

Oh and one other thing anyway to change temperature probe names? Too a bit of mucking round and had to change what probes do due to length of probe. Oh and 1 probe I can't set a rule for as it comes up as a float switch instead of a temp probe. Reads ok but can't get it to be used to control heaters

For the DC port, were you able to try it on a different DC port?

When you change controllers are you changing board type in the app on the settings page? As the power bar works with reef-pi it sounds like a GPIO configuration issue. 

At the moment to change the temp name you have to go into the database, the 2nd post on this thread explains how to log in. Once you get in you need to go to "ds18b20" table and you'll see all the sensors and can edit the name.

Sorry about the custom rules, currently there's an issue where you can't use pH probes and I suspect this will be related as it's probably an ID issue. I'll definitely get the rules sorted out for next update, I'll try and have it ready in about a week.

[Rob F]

but there is an issue. with the forced reboot, if you have an issue with the i2c bus not fixed by the reboot, you will Stick in a boot loop. every time the server boots, the controller starts, write fails and a reboot is made and so on, unlimited.

this makes very hard to connect via SSH because the server is rebooting cintinuosly

Your right, worse case scenario this could end up in a loop but if that happens something needs addressing. I hear you though and can see it would be difficult if the Pi keeps restarting with no chance to log into SSH. I'll add something so it knows if it restarts due to this and can't end up in a loop. Thanks for bring it up.

[Craig]

Cool, Actually had problem activating dc ato pump, so went back to version 1 controller(swapped sd card with Robo Tank into version 1) still no joy so went back to reef pi and still no joy, my multimeter is at work and have covid in the house so can't go and pick it up. But assuming pump is kaput.

One other issue I do have though is that 2 of my ac outlets won't activate, I have the 1 outlet version, works fine on reef pi but not on robo tank any ideas? tried both ac bar ports on the controller but no joy

Other than that I have it on my tank and looks great, won't need another outlet for a while until it gets colder and need another heater as back up and also when I get round to getting my doser back up and running. but still working out dosages with the tank anyway

Thanks again mate

---

Oh and one other thing anyway to change temperature probe names? Too a bit of mucking round and had to change what probes do due to length of probe. Oh and 1 probe I can't set a rule for as it comes up as a float switch instead of a temp probe. Reads ok but can't get it to be used to control heaters

For the DC port, were you able to try it on a different DC port?

When you change controllers are you changing board type in the app on the settings page? As the power bar works with reef-pi it sounds like a GPIO configuration issue. 

At the moment to change the temp name you have to go into the database, the 2nd post on this thread explains how to log in. Once you get in you need to go to "ds18b20" table and you'll see all the sensors and can edit the name.

Sorry about the custom rules, currently there's an issue where you can't use pH probes and I suspect this will be related as it's probably an ID issue. I'll definitely get the rules sorted out for next update, I'll try and have it ready in about a week.

Yep changed board type to match version, possibly, I think its the same plugs that don't work if I use either of the power board outputs on the controller. No stress though probably need another power board at some stage anyway

Yeah tried all DC ports, no go and swapping back to reef pi with old controller couldn't get the pump to work either, suspect pump is buggered, and cheap to get a replacement will let you know in a couple f days(will also have my multimeter by then as well so can confirm)
Yeah I remembered about database after I sent the last email and thought that would be how to do it.

[gandalf]

any eta for the next release?

[dlborum]

Have installed Ver 6.5 and it will install and run(with so assitance see below), but on a restart controller I get the following:(sudo journalctl -u robotank.service)

-- Logs begin at Thu 2019-02-14 05:11:59 EST, end at Wed 2022-05-18 06:45:35 EDT. --
May 18 06:28:16 RoboTank systemd[1393]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:16 RoboTank systemd[1393]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:16 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:16 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:16 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:19 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:19 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 1.
May 18 06:28:19 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:19 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:19 RoboTank systemd[1394]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:19 RoboTank systemd[1394]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:19 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:19 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:22 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:22 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 2.
May 18 06:28:22 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:22 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:22 RoboTank systemd[1398]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:22 RoboTank systemd[1398]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:22 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:22 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:26 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:26 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 3.
May 18 06:28:26 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:26 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:26 RoboTank systemd[1411]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:26 RoboTank systemd[1411]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:26 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:26 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:29 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.

May 18 06:28:29 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 4.
May 18 06:28:29 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:29 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:29 RoboTank systemd[1412]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:29 RoboTank systemd[1412]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:29 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:29 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:32 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:32 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 5.
May 18 06:28:32 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:32 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:32 RoboTank systemd[1413]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:32 RoboTank systemd[1413]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:32 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:32 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:35 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:35 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 6.
May 18 06:28:35 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:35 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:35 RoboTank systemd[1414]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:35 RoboTank systemd[1414]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:35 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:35 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:39 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:39 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 7.
May 18 06:28:39 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:39 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:39 RoboTank systemd[1427]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:39 RoboTank systemd[1427]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:39 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:39 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:42 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:42 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 8.
May 18 06:28:42 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:42 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:42 RoboTank systemd[1428]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:42 RoboTank systemd[1428]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:42 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:42 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:45 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:45 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 9.
May 18 06:28:45 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:45 RoboTank systemd[1]: Started Robo-Tank.
May 18 06:28:45 RoboTank systemd[1429]: robotank.service: Failed to execute command: Permission denied
May 18 06:28:45 RoboTank systemd[1429]: robotank.service: Failed at step EXEC spawning /var/www/html/cpp/startup: Permission denied
May 18 06:28:45 RoboTank systemd[1]: robotank.service: Main process exited, code=exited, status=203/EXEC
May 18 06:28:45 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:48 RoboTank systemd[1]: robotank.service: Service RestartSec=3s expired, scheduling restart.
May 18 06:28:48 RoboTank systemd[1]: robotank.service: Scheduled restart job, restart counter is at 10.
May 18 06:28:48 RoboTank systemd[1]: Stopped Robo-Tank.
May 18 06:28:48 RoboTank systemd[1]: robotank.service: Start request repeated too quickly.
May 18 06:28:48 RoboTank systemd[1]: robotank.service: Failed with result 'exit-code'.
May 18 06:28:48 RoboTank systemd[1]: Failed to start Robo-Tank.


In order for to even get anything to work with a fresh clean install on the SD card I have to do the following:

sudo wget -c https://www.robo-tank.ca/app/releases/Ro...k_v6_5.zip &> /dev/null && sudo unzip -o Robo-Tank_v6_5.zip -d /boot/ &> /dev/null && sudo bash /boot/robotank/install_robotank.sh

sudo find /var/www/html -type d -exec chmod 755 {} \;

sudo find /var/www/html -type f -exec chmod 644 {} \;

sudo systemctl restart apache2.service


After a reboot if I run these commands it will work untill I reboot the next time, issue is I can not get /boot/config.txt pin changes(dtoverlay=w1-gpio,gpiopin=18) to stick for a DS18B20 Sensor configuration in Port 4 of the sensor module.

Also as a suggestion add output ports to rules along with the sensors for rules, example since ATO is not working yet we can crate a rule if ATO full level sensor on and ATO Fill pump (or Valve) is on then turn off ATO Fill pump.

[Rob F]

Hi gandalf, I'm hoping to have a good weekend and get it ready for the end but might be mid next week.

Hi dlborum, thanks for posting, sorry it didn't go so well. This is yet another issue I've never come across, it's almost seems like the service file isn't setup properly. Can you try running sudo ./startup to see if it starts or run sudo ./robotank to see if that works. If you see any errors let me know. Maybe before running those commands stop the robotank service file using sudo systemctl stop robotank.service

The main install command you have is correct but it will only work on Buster variant of Pi OS, maybe verify that's what you installed. You shouldn't have to manually edit the /boot/config.txt file, when you change a sensor port to a DS18B20 via the configure ports screen it should update that file and prompt you for a reboot.

[dlborum]

sudo ./startup and sudo ./robotank both show this: (With or without running sudo systemctl stop robotank.service)

xxxxxxx@RoboTank:/ $ sudo ./robotank
sudo: ./robotank: command not found

Once I run sudo systemctl stop robotank.service I get the same errors I originally sent after I run the command sudo systemctl start robotank.service when checking the log file. I have Imaged 3 different SD cards(32gb) with Buster variant of Pi OS, have even tried the lite version vs the desktop version same issues.
I am going to try and image the SD card from my work laptop in the morning to see if that changes anything. If that works how do you move the database from one SD card to another? aka so I don't have to recreate all the rules, define sensors, set schedules... all the time consuming process's but functionality of the system.

If RPI 3B's where not stupidly priced at $200 I would try another micro controller.

[Rob F]

I'm sorry, before running those commands the directory needs to change. Run this command then the other two. Something is stopping the program from starting that's why those errors in the log. We will attempt to start it manually so we can see what's tripping it up, probably something in the database. This should tell us something.

cd /var/www/html/cpp

sudo ./robotank

sudo ./startup

Do try a fresh install, if your haven't try using this method.

https://docs.google.com/document/d/1fED4...N7COY/edit

I really don't think it's a problem with the Pi, if anything hardware it would be the SD card but that's not likely either. I don't have a process to copy the database yet, you would have to go into phpmyadmin and do it manually. There's an export tab to get it and import to add it.

[dlborum]

Ok Quick update....
I imaged the SD card on a different computer and Robo-Tank installed with no issues and booted right up. Restarted a few times to check and all looks good, now to attempt the database export ---> import. Fingers crossed

[Rob F]

Good to hear, hopefully the problem doesn't come with the database. If it does feel free to email it to me so I can see what caused the issue, first go in the systemSettings table and delete your email info.

[Rob F]

Sorry I don't have an update yet, I ended up touching lots this update so have a bunch of lose ends I need to tie up, I'll try to not do this again. Hoping to get a bunch done this weekend and have it ready next week.

One thing I've done is setup a VPS for the app. Right now I'm using my hosting for the website but a while ago I was in the control panel and they added a new web stat system so I checked it out and was surprised how many visitors the website was getting as my normal stats certainly didn't show all that, then it dawned on me it's from the auto update check as that pings the website every couple hours for each user. That's probably not good for my website so it'll start checking the VPS instead. I'm also very interested to know how many people are using the app so I'm adding a one time registration as it's easy now with the VPS. After registration a configuration file on the Pi will have a registration code that it'll look for to see if it's registered, this will be based off the Pi serial number. I'm doing it like this so it doesn't need to connect to the server every time it starts up, only for registration. I'll also add a button on the settings page "I'm not using app anymore" so I might know if someone gave up. This will help tell me how stable or unstable the program is, right now I don't know how many have tried it or who has abandoned it and why. Based on the hits to the website I think quite a few are using it but it's impossible to know so this will be very valuable info.

What got me down the VPS path though was I've always wanted to have a method for users to explore the app before buying. I think if I can do a good job (still lots to go first) people will log in and say ship me one, at least some might lol. I was thinking of setting up a Pi for people to access but it sounds like that's not a good idea using my personal internet connection so it would require a 2nd internet but obviously that's not feasible. Then I started thinking if the app runs on Linux why wouldn't it run on any Linux system other than a Pi so as the VPS only costs $1 a month which is amazing I order one to see what would happen. I installed Debian 11 ran the install script for the app and low and behold it works. Now anyone can go to a subdomain of robo-tank.ca and have secure access. :)

This also means the app should run on any Raspberry Pi clone such as the Banana Pi which would be awesome, I spent some time looking up alternative boards as the Pi shortage isn't going away any time soon and there's a lot of them, I think I came up with 4 other boards that have the same pinout and form factor so I expect they will plug into the controller and work just as well, once the app is closer to ready I'll get some to test out.

So yeah I'm quiet but things are slowly happening, I know a lot of above has nothing to do with feature set but it's all important on my side. Now that's it's possible outsiders can have access and if I find people are actually running it these will push me even more to get it done.

[gandalf]

can you tell us which feature we should expect in the next release?

[Rob F]

Not a lot of new features, mainly improvements. The scheduling interface was completely redone although it looks the same, now it uses the new table system that's much more efficient and fits better on small screens and I think everything is fully validated now. There was previously no validation when you edit a schedule and if a certain option wasn't checked it would crash the program. Fixed the "repeat every 1 day" option as that would skip the first day. It was also possible to create an "off" dosing schedule which wouldn't do anything and can confuse. Also when you added or edited a schedule there was a 250ms delay after you hit save before the table was loaded, it was enough to know it's happening. It was like that because the schedule data is sent to backend and saved from there but I had to make the UI delay until that happened otherwise it would read the database before the entry was made. Since I first added that I've figured out how to send confirmation from the backend to the frontend so it knows when actions are completed so no more delays are used. The same was for custom rules but that's been fixed as well. Speaking of custom rules I fixed them for pH, previously those custom rules wouldn't show the probe attached or work. I made a change on the new logging table, previously the "source" field just said schedule or custom rule, now it says that along with the schedule or custom rule name so you know which was logged better. I also updated almost all the variable array's in the backend to be fully dynamic, previously array's had a fixed length which put limits on things, for example schedule array's would only allow a maximum of 100 schedules, now the array's are sized to the actual number of schedules that exist so there's no hard cap, that took some time but with the help of valgrind memory checker it should be good. I also changed how the RTC works, now it's setup so when the Pi boots it reads the RTC first which updates the Pi clock text file in case NTP isn't available. The Pi will still attempt to read from NTP but if no internet is available it won't matter. In the app I'm no longer reading from the RTC, from actual time aspect it doesn't know the RTC exists. My main reason for this change is because the RTC doesn't track milliseconds, only seconds. As I'm picking and want milliseconds I had to do some wonky stuff to stay in that format with the RTC, now as it always reads the Pi time it's not an issue and the wonkiness is gone.

The biggest feature change is with dosing pumps, now I wish I hadn't started that as it's holding back the update because I got dragged into all the above but now I've made too many changes to put on pause. The UI is almost done but haven't made any backend changes yet. I added the ability to define dose amount for each schedule and added a copy of the dosing schedules to the dosing settings page so you can see and do everything related to dosing from there. The dosing schedules can still be altered from the main schedules screen, now it's just 2 options of doing it. You can also create daily batch schedules using total mL, number of doses per day and time to run first schedule, you can also select days of week these would apply to or repeat X number of days. Down the road I'll add other options for batches. Due to these changes I also have to update custom rules to allow user to define how many mL are dosed as the number of mL defined has been removed from dosing settings. I forgot this needs to be done so more time needed. I think I will leave the stirrer options and calibration for another day.

"mL" column has been added.


The "Milliliters to Dose" field only shows if creating a dosing schedule.




Just noticed in picture below it has a schedule "Repeat every 1 days", now that's not possible, if someone selects this as an option it'll change it to checkmarks for each day of the week.


Add group schedule