Cumulus MX FAQ: Difference between revisions

m
# MX needs to read from [[Cumulus.ini]] the earliest date to process as it ignores any content for earlier dates. (Cumulus 1 just read all the files it could find in the data folder.
# When you first start running MX, it does not seem to be able to populate charts and recent history web tags for earlier times.
 
= Dealing with rogue or missing input or derivatives =
 
INPUT ISSUES: Sometimes a weather station will supply incorrect data bits that mean wrong numbers are input to Cumulus MX or that MX looking at byte level does not recognise what it is being sent.
 
DERIVATIVES ISSUES: Cumulus MX ignores any maximum and minimum values supplied by a weather station; it calculates these derived values for itself. For rainfall, MX does not use any rainfall readings supplied by the weather station, instead it works with a long term counter, and gets daily readings to output from subtracting one counter value (at end of required period) from another at start of day. It calculates further derived values like apparent temperature and feels like (and many others) by looking at various simultaneous readings (temperature, wind speed, and humidity for those two examples).
 
EXTREMES ISSUES: When MX is calculating those maximum and minimum derivatives, these are within an individual day, but they are compared against others in the same month (to update both the this month web tags and the monthly extremes web tags), against others in the same year (to update the this year web tags) and against all previous derivatives (to update the all-time web tags). I say update web tags, because that is how you view the extremes, but actually there is a log file for maximum and minimum in a day (today.ini), for this month (month.ini), for monthly-all-time (monthlyalltime.ini), for this year (year.ini), and for all time (alltime.ini). These log files are where MX stores the extremes and where you need to do corrections. For total rain this month, this year, maximum rain for this month in all years and highest rainfall (hourly, daily, monthly), the same log files are used. The highest daily wind run per month, per year, in the same month for all years and all-time is similarly stored in these log files and made available as web tags.
 
#Dealing with input problems depends on your weather station type
#*If there is a problem with input from a Davis weather station, stop MX, then it may be if you reset the Davis console, and the logger you use with your David, it might stop the bad data from then on, this won't correct the past data, but hopefully will stop the problem
#*If there is a problem with an old style Fine Offset, often with rain or pressure reported, try stopping MX, then resetting the Fine Offset (some have a small switch below a small hole on the back of the console so use a strong long pointed instrument with a small diameter, maybe a small nail or a sewing tool to initiate a reset). Disconnecting the USB connection and removing a console battery may also reset a Fine Offset, or there may be some combination of touches or switch pushes that does a reset.
#* For other weather station types, look in the forum for advice or read the station instructions.
# If the output values are frequently incorrect, MX provides some calibration settings in the admin interface. These vary dependent on the type of weather reading you want to correct. In general they use a formula like <tt> Output = Multiplier_2 X Input<sup>2</sup> + Multiplier_1 X input + Offset</tt> so in the settings you can specify the multipliers and offset to use in that equation.
#If only a small numbers are incorrect, MX provides editors for the logs that store extremes (although for today only the rainfall can be edited). See the pages linked from [[:Category:Log Files|Category:Log Files]] to discover more about each log and how to deal with rogue values. View and edit these on'''Edit''' tab of admin interface.
#You can correct any individual rogue value in a monthly log as MX (see viewer editor on '''Data logs''' tab of admin interface.
#As rainfall has a lot of different derivatives stored for it (hourly totals, daily totals, rain rate) it is quite complicated to correct, so there are a series of instruction in the [[FAQ#My_station_invented_some_rain_that_didn.27t_really_occur.2C_and_I_want_to_set_it_to_zero_.28or_some_other_figure.29|C1 FAQ]].
 
= Problems with Cumulus MX use of MySQL database =
#If there are extra columns in the MX schema compared to the schema you were using in Cumulus 1, follow instruction above as far as the necessary settings, but before you enable the updates either use any script provided in release post for when the extra columns were added, or use whatever administrative tool (it might be phpMyAdmin) is provide to add the extra columns. The enable the updates.
#If you want to use different columns to those MX uses (you don't have solar, or you want snow or extra sensor data to be in your database) then you need to use the Custom upload facilities that MX provides. To keep this answer short, I won't explain how here.
 
=== Using Custom SQL ===
 
Obviously to use Custom SQL you need some knowledge of SQL and this answer cannot teach you SQL. But the general format of an update instruction for a whole row is given below, ensure the command you supply to MX has all the parts shown. The instructions in {} are optional, they just make the SQL read better, parts in lower case need to be replaced by whatever is appropriate for your table, values are web tags enclosed in single quotes e.g. <tt>'<#windrun>'</tt>, if there are any columns for which you don't have a value, use NULL without any quotes as this is SQL.
<pre> INSERT [INTO] database_table_name (primary_key_column_name, column_name_1, column_name_2, .... column_name_last) VALUES ('primary_key_value', 'web_tag_1', 'web_tag_2', ... 'web_tag_last');
 
= Problems with web pages or NOAA reports =
5,838

edits