reef-pi :: An opensource reef tank controller based on Raspberry Pi.

[Matevz Savarin]

Well not all feeds this 3 are missing entirely..so there must be a bug somewhere...

 

[Ranjib]

I want to access my Reef-Pi installation from the web using SSL. I'm used to use DDNS, portforwarding in my router to the servers in my LAN and an public certificate issued by Letsencrypt.

With Reef-Pi I face an issue:

pi@reefpi:~$ sudo letsencrypt

leads to:

Saving debug log to /var/log/letsencrypt/letsencrypt.log
Certbot doesn't know how to automatically configure the web server on this system. However, it can still get a certificate for you. Please run "certbot certonly" to do so. You'll need to manually configure your web server to use the resulting certificate.

I could run "certbot certonly", but I don't know how and where to install the generated the certificate.

Any hint from somebody more experienced than me? Thanks!

I have not tried this personally, but should be fairly easy. Try this: https://pimylifeup.com/raspberry-pi-ssl-lets-encrypt/ , its for apache web server (which we are not using) but the instructions provided here does show how to control the path of generated certificates.
Please do report back if this works. If not, i'll find out how to do this. Its an important topic (to run reef-pi with https and proper cert authority), which should be easy, and we should know if theres some obvious gotchas.

I run a configuration management system and install my own selfsigned root can and corresponding PKI infra ( certs, crl, csrs etc)

 

[Ranjib]

I'm having some trouble with my temp sensors. Running 3.3.1. Every couple days I will get an email alert with the following error for each sensor,

When I go to reef-pi and check the Temperature tab the sensors are gone from the drop down and I get the errors,

Usually a power cycle will fix it and it will be fine again for a couple more days. I thought it was the 3.5mm audio jack so I wired it direct a couple days ago but it just did it again tonight. Any thoughts? Thanks.

The error indicates something wrong on the physical/linux layer that not under reef-pi 's control. The first error means the value we are getting from sensor is failing on checksum (i.e. the data sent from the sensor is mangled on the wire by the time it reaches the linux i/o ports). The second one means the sensor is not being detected. Most likely its a sensor / wiring issue that got progressively bad. Its fascinating, as you are the first one to report these two different errors happening in order (first checksum mismatch, then lost sensor).

 

[Ranjib]

I'm having some trouble with my temp sensors. Running 3.3.1. Every couple days I will get an email alert with the following error for each sensor,

When I go to reef-pi and check the Temperature tab the sensors are gone from the drop down and I get the errors,

Usually a power cycle will fix it and it will be fine again for a couple more days. I thought it was the 3.5mm audio jack so I wired it direct a couple days ago but it just did it again tonight. Any thoughts? Thanks.

The jacks should only cause issue when you attach/detach sensors, if at all. Mine are all based on 3.5mm jacks and been working for more than two years now, in some builds.

 

[Ranjib]

Hi guys

Pro tip (or in my case stupid amateur mistake!) - do not fully immerse the PH probe like i did for the last week. I was getting consistently low PH readings (sulphuric acid level readings that my fish were living in of course ;Woot) but when i would put my probe in tap water it would read fine. I thought I had a ground leakage thinking the worst that it could kill my tank/me so i started turning off and on outlets with no result. I then thought to recalibrate which proved fruitless. In the end, it was me putting in more media in my sump thus forcing me to readjust the probe to above water which magically made my problem disappear.

This may not apply to all ph probes but the typical black looking ebay ones are not sealed at the top. I am going to order another one anyway incase it causes an issue in the future but its got me motivated to make a probe holder from spare acrylic i have to make sure it goes in right alongside my temperature probe.

Glad you sorted this out. and thank you for sharing your learning with us. we get better bit by bit with each other's learning, when shared.

 

[Ranjib]

Well not all feeds this 3 are missing entirely..so there must be a bug somewhere...

Do you see anything in reef-pi log? it could be something to do with rate limiting (too much data sent to adafruit.io ) or naming issue

 

[Des Westcott]

A little something I've noticed , but not sure it can be addressed or not. It relates to updating equipment status on the "Equipment" tab.

If I run a Macro, say my kalkstirrer one that just turns the stirrer on, runs for 5 minutes and then turns it off and then go to the "Equipment" tab, I see that the Stirrer status is "On". It runs and I hear the relay click off at the correct time, but it does not update the Equipment status ie the stirrer is off, but it shows as on. If I click onto another tab and go back to Equipment, it displays the correct status. I've noticed it with my ATO pump status as well.

I suspect it has something to do with the browser refreshing - I have tried it on Chrome as well as the new Edge and get the same behaviour. It's not a big deal, but I was wondering if there was a setting or something I had wrong.

Edit - Just to be clear, every time I open the "Equipment" tab it shows the status correctly. If I change it manually, it shows correctly. Only if an automated process causes a change in status, it doesn't update the display, but the equipment functions correctly.

 

[Michael Lane]

A little something I've noticed , but not sure it can be addressed or not. It relates to updating equipment status on the "Equipment" tab.

If I run a Macro, say my kalkstirrer one that just turns the stirrer on, runs for 5 minutes and then turns it off and then go to the "Equipment" tab, I see that the Stirrer status is "On". It runs and I hear the relay click off at the correct time, but it does not update the Equipment status ie the stirrer is off, but it shows as on. If I click onto another tab and go back to Equipment, it displays the correct status. I've noticed it with my ATO pump status as well.

I suspect it has something to do with the browser refreshing - I have tried it on Chrome as well as the new Edge and get the same behaviour. It's not a big deal, but I was wondering if there was a setting or something I had wrong.

Edit - Just to be clear, every time I open the "Equipment" tab it shows the status correctly. If I change it manually, it shows correctly. Only if an automated process causes a change in status, it doesn't update the display, but the equipment functions correctly.

This is a similar issue to outlets on timers not updating in the UI, and macro status not updating in real time. There was a quick discussion in the reef-pi UI thread about it. I agree that it would be nice to add this functionality, and the devs have talked about using websockets to publish updates back to the UI. I'll consider picking this up after I finish working on drivers for ADS1015 and ADS1115.

 

[Matevz Savarin]

Do you see anything in reef-pi log? it could be something to do with rate limiting (too much data sent to adafruit.io ) or naming issue

I am getting this errors, but only 2 feeds so there couldnt be too much of them..under log all is empty.
And today only system usage nothing else gets true..
I havent change nothing exept new key, but its same. Before I got all the datas, if there was too much for 60s I saw the masssage. But temps and ph and ato was working...

 

[Des Westcott]

I am getting this errors, but only 2 feeds so there couldnt be too much of them..under log all is empty.
And today only system usage nothing else gets true..
I havent change nothing exept new key, but its same. Before I got all the datas, if there was too much for 60s I saw the masssage. But temps and ph and ato was working...

I used to get those - it was around naming of the feeds. Change names to have no spaces or other non-allowed characters and all went away.

 

[Mike25]

Hi all.
I've been working on something but can't figure out if it's even doable.
Is it possible to wire a non-contact sensor like this: http://www.cqrobot.wiki/index.php/Non-contact_Water_/_Liquid_Level_Sensor but instead of the traditional method to trigger the ATO, turn on a dosing pump instead?
I have a small nano (10 gal) controlled by my reef pi, and only use 2 of my 3 dosing pumps. I was wondering if I could wire it up to detect water level drop, and use the third dosing pump as a quasi-ato? I use Michael Lane's dosing adapter (awesome device), a Jebao slave doser DP3S, and Michael's goby hat running on a pi4b.

 

[Michael Lane]

Hi all.
I've been working on something but can't figure out if it's even doable.
Is it possible to wire a non-contact sensor like this: http://www.cqrobot.wiki/index.php/Non-contact_Water_/_Liquid_Level_Sensor but instead of the traditional method to trigger the ATO, turn on a dosing pump instead?
I have a small nano (10 gal) controlled by my reef pi, and only use 2 of my 3 dosing pumps. I was wondering if I could wire it up to detect water level drop, and use the third dosing pump as a quasi-ato? I use Michael Lane's dosing adapter (awesome device), a Jebao slave doser DP3S, and Michael's goby hat running on a pi4b.

You should be able to do this by configuring the 3rd dosing pump pin as an outlet instead of a jack. Then you can treat it just like any kind of ATO pump.

 

[Mike25]

You should be able to do this by configuring the 3rd dosing pump pin as an outlet instead of a jack. Then you can treat it just like any kind of ATO pump.

Thanks Michael! I'll try that.

 

[Matevz Savarin]

I am using vnc for now, I just want to know 2 temps, if room one rises I turn on ac.. next step is to do this automated. Names used to work in adarfuit io before. I ll try rename them.. tnx

 

[Matevz Savarin]

I used to get those - it was around naming of the feeds. Change names to have no spaces or other non-allowed characters and all went away.

I am using vnc for now, I just want to know 2 temps, if room one rises I turn on ac.. next step is to do this automated. Names used to work in adarfuit io before. I ll try rename them.. tnx

 

[Ranjib]

A little something I've noticed , but not sure it can be addressed or not. It relates to updating equipment status on the "Equipment" tab.

If I run a Macro, say my kalkstirrer one that just turns the stirrer on, runs for 5 minutes and then turns it off and then go to the "Equipment" tab, I see that the Stirrer status is "On". It runs and I hear the relay click off at the correct time, but it does not update the Equipment status ie the stirrer is off, but it shows as on. If I click onto another tab and go back to Equipment, it displays the correct status. I've noticed it with my ATO pump status as well.

I suspect it has something to do with the browser refreshing - I have tried it on Chrome as well as the new Edge and get the same behaviour. It's not a big deal, but I was wondering if there was a setting or something I had wrong.

Edit - Just to be clear, every time I open the "Equipment" tab it shows the status correctly. If I change it manually, it shows correctly. Only if an automated process causes a change in status, it doesn't update the display, but the equipment functions correctly.

the state of the equipment in UI tab is loaded every time you go to that tab from other place, or refreshed at 10s interval. So there are chances that you may see stale state for upto 9 seconds at worst

 

[Des Westcott]

the state of the equipment in UI tab is loaded every time you go to that tab from other place, or refreshed at 10s interval. So there are chances that you may see stale state for upto 9 seconds at worst

Thanks Ranjib. I'll check that 10s interval now an confirm. Maybe it's just me being impatient

 

[Des Westcott]

@Ranjib There's definitely something going on with equipment status. I manually triggered my ATO pump so that it ran for a few seconds and then stopped as the ATO grabbed control of it. Status stayed showing "On" and I set a stopwatch. 5 minutes later and it was still showing "On" despite the pump being off. Jump to Dashboard and back to equipment and it has changed to "Off"

 

[brian2kgt]

The error indicates something wrong on the physical/linux layer that not under reef-pi 's control. The first error means the value we are getting from sensor is failing on checksum (i.e. the data sent from the sensor is mangled on the wire by the time it reaches the linux i/o ports). The second one means the sensor is not being detected. Most likely its a sensor / wiring issue that got progressively bad. Its fascinating, as you are the first one to report these two different errors happening in order (first checksum mismatch, then lost sensor).

Thanks, I had a hunch it might be a physical problem. I have one of the better sensors from Sparkfun arriving tomorrow to test. So would one bad sensor cause both sensors to throw the checksum error simultaneously? If not, I guess i would have to look for a wiring issue in a place they share in common? This would be the short section where they are connected in parallel to the Pi with the shared resistor.

It happened again last night. An email alert from both sensors at the same time that " Temperature sensor failure. Error: First line of device file does not end with YES", followed by errors and a loss of both sensors in reef-pi. A power cycle fixed it again both times.

 

[Auto-pilot]

I'm not really sure about the pico board ph since I haven't used it. Based on the driver code, I believe you should use pico-board driver, and 0x75 hex address might be the ph address, which should be entered as 117 (decimal) in reef-pi.

Could not get this working no matter what I've tried I can't get a reading. I've tryed contacting theatrus but it seems like hes busy and hasn't been able to get back to me. I was curious if anyone has recommendations for another board. I might try to make another system and give mine to a good friend