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

[stefanm]

@Ranjib I am building my set up. My DJ-8 doesn’t seem to be responding. What can I check to troubleshoot? On Facebook someone suggested to check my signal and power sufficient. What can be wrong?

Check that your grounds are tied, also check that your PSU to the relays is sufficient...

 

[Orbela]

Check that your grounds are tied, also check that your PSU to the relays is sufficient...

Thanks for response
So I checked and I am getting 12v on the connectors pins
the GPIO 27-22 an 12, 6,14 I am reading 0v

 

[NeonRabbit221B]

can you post your yaml file. i dont think these 9090 stuff are valid yaml

# my global config
global:
scrape_interval: 15s # Set the scrape interval to every 15 seconds. Defau$
evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is ev$
# scrape_timeout is set to the global default (10s).

# Alertmanager configuration
alerting:
alertmanagers:
- static_configs:
- targets:
# - alertmanager:9093

# Load rules once and periodically evaluate them according to the global 'evalu$
rule_files:
# - "first_rules.yml"
# - "second_rules.yml"

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
# The job name is added as a label `job=<job_name>` to any timeseries scraped$
- job_name: 'prometheus'

metrics_path: /x/metrics
# scheme defaults to 'http'.

static_configs:
- targets: ['192.168.xxx.xxx']

Here is my config and I am trying to access from a different raspberry pi so the IP is the IP of the reef-pi . Every guide I have found uses a port to access the webpage which is the reason for the 9090. Anyone have a solid guide that is fairly 1-1?

 

[Ranjib]

Thanks for response
So I checked and I am getting 12v on the connectors pins
the GPIO 27-22 an 12, 6,14 I am reading 0v

Those are sinking pins. When you are testing make sure you connect multimeter +ve to the 12v power pin, and multimeter gnd(black) to the uln28003 pins,

 

[Ranjib]

# my global config
global:
scrape_interval: 15s # Set the scrape interval to every 15 seconds. Defau$
evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is ev$
# scrape_timeout is set to the global default (10s).

# Alertmanager configuration
alerting:
alertmanagers:
- static_configs:
- targets:
# - alertmanager:9093

# Load rules once and periodically evaluate them according to the global 'evalu$
rule_files:
# - "first_rules.yml"
# - "second_rules.yml"

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
# The job name is added as a label `job=<job_name>` to any timeseries scraped$
- job_name: 'prometheus'

metrics_path: /x/metrics
# scheme defaults to 'http'.

static_configs:
- targets: ['192.168.xxx.xxx']

Here is my config and I am trying to access from a different raspberry pi so the IP is the IP of the reef-pi . Every guide I have found uses a port to access the webpage which is the reason for the 9090. Anyone have a solid guide that is fairly 1-1?

So , Prometheus is running now ? Do you have grafana running as well ?

 

[NeonRabbit221B]

Yes, I can run sudo systemctl status prometheus and I do not get an error but Its not connecting to the other reef-pi and I cannot load the page via the IP.

 

[Orbela]

Those are sinking pins. When you are testing make sure you connect multimeter +ve to the 12v power pin, and multimeter gnd(black) to the uln28003 pins,

I tested the sinking pins and the matching uln2803 pins the in test 12.3v and the out tested at 0v can the uln2803 be faulty? I placed the + in the sinking pins and the gnd - on the 'in' and 'out' which the 'in' has 12.3v and the 'out' 0v. Am I doing something wrong? Thanks for the help

 

[b4tn]

I know it was discussed somewhere but I can’t find it. Right now my temp(recent) shows readings for the past 12 hours. Is there a way to change that to 24 or 48 hours?

 

[Ranjib]

I know it was discussed somewhere but I can’t find it. Right now my temp(recent) shows readings for the past 12 hours. Is there a way to change that to 24 or 48 hours?

Under the telemetry configuration section you can define the number of data points that will be saved for individual metrics for recent data ( per check interval) and historical data (hourly average ). Those values and your actual check interval together determines how long look back time is. For example 60 data points for recent data at 5 min interval will only have 300 min or 5 hour worth of look back

 

[Orbela]

@Ranjib am I testing it correctly?

 

[dmolavi]

@Ranjib After a house move and quarantine, I'm finally ready to upgrade from 3.0 alpha...is there anything I need to know before upgrading? Do I need to wipe the database or is it a seamless upgrade?

 

[Orbela]

Just saw your message, sorry
This is my psu think it enough unless I need to get a 1A not 5A

 

[Ranjib]

@Ranjib After a house move and quarantine, I'm finally ready to upgrade from 3.0 alpha...is there anything I need to know before upgrading? Do I need to wipe the database or is it a seamless upgrade?

Should be seamless. Still keep the existing db copied in some place else (/home/pi)

 

[b4tn]

So we use the PCA9865 for additional outputs. I was looking at the connectors for inlets and noticed the PCA9865 was an option for a driver. Can the PCA9865 be used as an inlet also using the below for a float switch?

 

[Ranjib]

So we use the PCA9865 for additional outputs. I was looking at the connectors for inlets and noticed the PCA9865 was an option for a driver. Can the PCA9865 be used as an inlet also using the below for a float switch?

float switch are input (inlet), and can not be used with pca9685. You can use it for relay other than raw pwm, thats what it means. I have not tested it yet. its basically 100% duty cycle pwm.

 

[Orbela]

If anybody can help me I am sort of struggling to get my power relay to work.
I heard clicking and then channel 3 started to work then stopped then Channel 6 started to work. Now nothing works. I am lost. I check all connection were not lose and I double checked I didn’t have any cold solder. What I am thinking might be an issue is the connection between my raspberry pi and the perma hat. I am new was I suppose to solder them together or use the provide connection?

 

[b4tn]

float switch are input (inlet), and can not be used with pca9685. You can use it for relay other than raw pwm, thats what it means. I have not tested it yet. its basically 100% duty cycle pwm.

I actually forgot you can use the same float inlet for multiple rules. I just made a test macro that disables the ATO and enables/disables the rules then re-enables ATO. No need for another float switch :)

 

[Orbela]

Thanks for the help it was my Raspberry pi connection with my pi hat not fully inserted. Now just adjust my gpio channels. Thsnks again.

 

[Ranjib]

If anybody can help me I am sort of struggling to get my power relay to work.
I heard clicking and then channel 3 started to work then stopped then Channel 6 started to work. Now nothing works. I am lost. I check all connection were not lose and I double checked I didn’t have any cold solder. What I am thinking might be an issue is the connection between my raspberry pi and the perma hat. I am new was I suppose to solder them together or use the provide connection?

With reef-pi running, switch on/off the corresponding outlet and take voltage readings across pi gpio, perma proto /uln input pins and finally on the db9 connector pins. remember from uln onward multi meter black probe goes on to uln/db9 pins and red probe goes to 12 v. See where the voltage readings are not as expected . It will be easier if you have a dedicated thread, we are likely to loose context in this thread.

 

[Ranjib]

Thanksgiving is approaching, and i am thinking about 4.0 release. It has been very hectic and i have not been able to push a lot of code as expected, lately. I am thinking whether to do a 4.0 release or skip it. Technically we dont have to have a major version release, since there no breaking changes. But historically it has been a tradition to do yearly reef-pi release during thanksgiving, hence the dilemma.

I am thinking of wrapping up some minor bugs and polish documentation and cut a release (4.0 ?) , if we want stick to the tradition. We did have quiet a few changes across the board (new features, drivers, bug fixes etc) encompassing 3->4 development, but most of those have been already incrementally shipped.

Let me know what you all think