One issue with mcsControlSystems

[Mario from Spain]

Hi.

1: Each hour (after the "hourly security off" sent to the Rain8) I get in the log <BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR> 29/06/2004 15:00:27~!~mcsControlSystems~!~ Log_Sprinkler SQL=INSERT INTO tblSprinklers(SampleDate,RiegoFrenteMeasured,RiegoFrenteCalc ulated,RiegoFrenteFuture,RiegoFrenteMinutesCum,RiegoAtrasMea sured,RiegoAtrasCalculated,RiegoAtrasFuture,RiegoAtrasMinute sCum,S32,S33,S35) SELECT #2004-06-29 15:00:27# AS QT,4,724079E+08 AS Q0,-100 AS Q1,-100 AS Q2,336850100 AS Q3,7500 AS Q4,-100 AS Q5,-100 AS Q6,15600 AS Q7,0 AS Q8,0 AS Q9,0 AS Q10 El número de valores de consulta y el número de campos de destino son diferentes.
<HR></BLOCKQUOTE>
"El número de valores de consulta y el número de campos de destino son diferentes." is like "The number of values for the query and the destiny fields are different.

2: When the sprinklers are working I get in the log <BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR> 30/06/2004 0:10:38~!~mcsControlSystems~!~ Log_Discrete: INSERT INTO tblRiegoRiegoActivo (SampleDate,Discrete) SELECT #2004-06-30 0:10:38# AS QT,1 AS Q1 No se pudo encontrar la tabla de resultados 'tblRiegoRiegoActivo'.
<HR></BLOCKQUOTE>
It happens two times each minute (I have two zones).
Also, the "Wk Mins" in the ControlSystems.asp is wrong (I already reported to you this issue last year). Rigth now it shows -3856716 for one zone and 48 for the other.


Will the plugin create a new database file with the rigth tables and values if I just delete the file, go into the setup page and press "Update"?

[Michael McSharry]

Problem #1 is due to the the 4,724079E+08 which looks to the database as two numbers. A 4 and a 724079E+08. This number seems to represent a soil moisture measurement. Are you measuring soil moisture or have a device defined for this purpose? If the value in this device's string can be made smaller so the E+08 notation will not be invoked then you have a workaround until I limit it directly in the plugin.

Problem #2 seems to indicate that the tblRiegoRiegoActivo table does not exist in the datbase. I suspect this is a problem between how mcsTemperature and mcsControlSystems manages the device to table relationships. I'll need to change mcsControlSystems to be consistent.

mcsControlSystems assumes the database table name is the same as the Homeseer Device location & name with a tbl prefix and spaces removed. It will create this table for you if the mcsTemperature.ini file, [Database] group entry for the device code has a null value or the device code does not exist in this group. I think you can also achieve the desired result by renaming the database table for this device in mcsTemperature setup to tblRiegoRiegoActivo.

Now that you remind me about the Wk mins I see that there are several open PRs. Looks like it is time to address these since it is watering season.

[Mario from Spain]

I just noticed something wrong about the versions.

I'm runing mcsTemperature 4.23.3 (file hspi_mcsTemperature.exe 1.953.792 bytes).

Also I have in the homeseer folder the file hspi_mcsControlSystems.ocx with 888.832 bytes. The file says it's version 1.12.0.3 but the Setup dialog and the web page says its v 1.7.0 and can't see nothing about 8 zones or multiple exclude times. Somehow homeseer is loading a old version but can't find from where. I did a Search for the file hspi_mcsControlSystems.ocx in the whole hard disk and I just can't find this one.

Any idea?

[Michael McSharry]

I suggest that you unregister the current ocx and register a new download. I attached an update here that tries to address your database problems in a manner that does not require any editing on your part. I have no way to check it so it would be nice if you could try.

The registration utility is contained in the zip file. Double-click it once and then registration will be via right click on Windows Explorer

[Mario from Spain]

Michael, that is fast!

I just upgraded the OCX (the new file is smaller, 792 vs 868 KB), unregistered and registered and run HomeSeer and a message from HomeSeer appeared in the initialization saying "This action cannot be completed because the other application is busy. Choose "Swith To" to activate the busy application and correct the problem.". After pressing "Swith To" the HS initialitation continued.
I reverted to version 1.12.3 and the message appeared at least six times instead of one.
I again upgraded to 1.12.4 the message still appears few times.

If I remove the plugin from the HS menu, the problem still exists. I think it's stoping initializing the WebCam plugin. If I removed the WebCam plugin HS starts without problem.


REM: I already saw this kind of message in mcsTemperature when I had the connection to the internet via dialup modem, the connection down and mcdTemperature tried to update the forecast. Probably there is a loop in mcsTemperature when the connection is down.

[Michael McSharry]

There is quite a bit of consistency checking at the startup of the plugin and this can capture the CPU for awhile and generate the switch to messages. I sprinkled some escapes back to the OS to free up the CPU at various points during initialization. See if this makes any difference. Also, what about the database problems?

[Mario from Spain]

Looks like the 1.12.5 solved the problem with the switch to messages.

The database problems seems to be fixed also.

I'm accesing this computer on remote (it's at 100 km from my home) but this weekend will do some more testing.

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Problem #1 is due to the the 4,724079E+08 which looks to the database as two numbers. A 4 and a 724079E+08. This number seems to represent a soil moisture measurement. Are you measuring soil moisture or have a device defined for this purpose? If the value in this device's string can be made smaller so the E+08 notation will not be invoked then you have a workaround until I limit it directly in the plugin. <HR></BLOCKQUOTE>

I'm not very sure if I understand this. I have not soil moisture sensors and have no idea where this value is comming from. Where must I look, in mcsTemperature or mcsControlSystems? In a first check I can´t find where it is in the configuration (it's a little hard to check many things in remote).

[Michael McSharry]

You have a database field RiegoFrenteMeasured which should correspond to the Measured Soil Moisture level for your first area.

What I suspect is happening is the mcsControlSystem.ini file has some bad historical data which affects your wkMins and this measurement. This data is stored in the [Sprinkler Status] group of this file. You should be able to delete the contents of this group and it should reset the history back to zero for everything and you will run into fewer problems. As a precaution backup the file before you edit it. Do the editing while Homeseer is shutdown so the plugin is able to startup with the new values during init.

[Mario from Spain]

Hello Michael.

It looks like everything is messed!

I already deleted the ini section but everything still the same.

Also. I have to start times (hopping to rain two times a day) at 8 and 21 but in the status web page the next wathering time always shows like tomorrow at 12:00 (I tried without "Don't water between hours" and with it 0 to 6 and 1 to 6). It looks like I can update the GUI but the logic is ignoring the changes.

The manual control is not working also.

Attached is an export of my tblSprinklers table in Excel format.

And bellow is my mcsControlSystems.ini file

As I noticed I had duplicate virtual devices in different codes (like having "SontrolSystems|Sprinklers Active" two times" I deselected the plugin in the HS config, deleted all the devices in house code "_" and resellect the plugin in the HS config. At next start, the plugin recreated all the devices but also a location "Sprinklers" I never seen before with one device "Manual Control" and a Device Code "1 0"

Maybe I sould delete everything, including the tblSprinklers table, and start over again?

[Control Systems]
HouseCode=94
[Sprinklers]
ForecastInches="43"
SolarDeviceCode=""
DontDayofWeek="0,0,0,0,0,0,0"
DayofWeek="1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0 ,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 ,0"
DontEvenDays="0"
DontOddDays="0"
VirtualDeviceEnableOption="0"
DontWindSpeed="25"
DontWindMinutes="30"
DontWaterHoursStart="1"
DontWaterHoursStart2=""
DontWaterHoursStart3=""
DontWaterHoursEnd="6"
DontWaterHoursEnd2=""
DontWaterHoursEnd3=""
DontVirtualDevice="v12"
NewDayHours="8,21"
SolarIntensity="1"
SaturationInches="1"
MinDays=""
MaxCycles=""
MaxDays="2"
ManualHouseCode=""
OverrideDeviceCode="v10"
OverrideOption="0"
ForecastWithin="0"
ForecastWithinLast="0"
ForecastFor="0"
ForecastWithinDays=""
ForecastWithinLastDays=""
ForecastForInches=""
EvenDays="0"
EvenDays2="0"
EvenDays3="0"
EvenDays4="0"
EvenDays5="0"
EvenDays6="0"
EvenDays7="0"
EvenDays8="0"
ControlStrategy="0"
ControlStrategy2="0"
ControlStrategy3="0"
ControlStrategy4="0"
ControlStrategy5="0"
ControlStrategy6="0"
ControlStrategy7="0"
ControlStrategy8="0"
PumpOption="1"
PumpOption2="0"
PumpOption3="0"
PumpOption4="0"
PumpOption5="0"
PumpOption6="0"
PumpOption7="0"
PumpOption8="0"
IntervalDays="2"
PumpDeviceCode="v11"
MeasuredDevice=""
CalcMoistureLevel=""
MeasuredThreshold=""
IntervalDays2="2"
PumpDeviceCode2="v11"
MeasuredDevice2=""
CalcMoistureLevel2=""
MeasuredThreshold2=""
IntervalDays3="2"
PumpDeviceCode3="v11"
MeasuredDevice3=""
CalcMoistureLevel3=""
MeasuredThreshold3=""
IntervalDays4="2"
PumpDeviceCode4="v11"
MeasuredDevice4=""
CalcMoistureLevel4=""
MeasuredThreshold4=""
IntervalDays5="2"
PumpDeviceCode5="v11"
MeasuredDevice5=""
CalcMoistureLevel5=""
MeasuredThreshold5=""
IntervalDays6="2"
PumpDeviceCode6="v11"
MeasuredDevice6=""
CalcMoistureLevel6=""
MeasuredThreshold6=""
IntervalDays7="2"
PumpDeviceCode7="v11"
MeasuredDevice7=""
CalcMoistureLevel7=""
MeasuredThreshold7=""
IntervalDays8="2"
PumpDeviceCode8="v11"
MeasuredDevice8=""
CalcMoistureLevel8=""
MeasuredThreshold8=""
DebugLogSprinkler="0"
Safety="1"
MeasuredSense="0"
OneZone="1"
CompleteAtSunrise="0"
ValveProtect=""
NotifyMinutes=""
PriorHour="12"
PriorDay="10"
IntervalCount0="0"
IntervalCount1="0"
IntervalCount2="0"
IntervalCount3="0"
IntervalCount4="0"
IntervalCount5="0"
IntervalCount6="0"
IntervalCount7="0"
[Sprinkler Zones]
SI1="15,1,0,0"
SI2="15,1,1,1"
[Control Systems Status]
S_1="0"
S_2="0"
S_3="0"
S_4="0"
S_5="0"
S_6="0"
S_7="0"
S_8="0"
S_9="0"
S_10="0"
S_11="0"
S_12="0"
S_13="0"
S_14="0"
S_15="0"
S_16="0"
S_17="0"
S_18="0"
S_19="0"
S_20="0"
S_21="0"
S_22="0"
S_23="0"
S_24="0"
S_25="0"
S_26="0"
S_27="0"
S_28="0"
S_29="0"
S_30="0"
S_31="0"
S_32="1"
S_33="1"
S_34="0"
S_35="0"
S_36="0"
S_37="0"
S_38="0"
S_39="0"
S_40="1"
S_41="0"
S_42="0"
S_43="0"
S_44="0"
S_45="0"
[Sprinkler Status]
STI1="True,15,11/07/2004,False,False,False"
STI2="True,15,11/07/2004 0:15:00,False,False,False"
SSI1="0,8087987,1"
SSI2="0,8087987,1"

[Michael McSharry]

It looks as if it must read the *** minutes from the database during init so simply deleting the section does not erase the memory.

The prediction algorithm on the version you are running does not reflect the second and subsequent start times within a day. The control algorithm, however, will obey it. The user interface to the start time specification and the prediction algorithm to support it have changed in the version 1.13.x that DaveCB is evaluating.

Average winds of 25 for 30 minutes are pretty strong. Is this a windy environment?

There are several forms of manual control. Some are with palmpad-type and others are web-based. What in particular are you having a problem with in this area?

There are other problems that your previously reported that were addressed in the v1.13.x. I would not recommend using it yet until DaveCB has completed his checkout, but the solutions are near. You can delete/rename the database to get a clean start with what your are running. While not much is happening yet for my sprinling season, the version posted at the top of the forum is the one that I have running.