Data Spike is preventing CumulusMX from running

[lgeitner]

Hi,

I am receiving the following error when starting cumulusMX.

2022-01-24 19:35:20.802 GetArchiveData: Loaded archive record for Page=0 Record=4 Timestamp=1/24/2022 11:35:00 AM
2022-01-24 19:35:20.802 GetArchiveData: Processing archive record for 1/24/2022 11:35:00 AM
2022-01-24 19:35:20.851 Spike removal: Rain rate greater than specified; reading ignored
2022-01-24 19:35:20.851 Spike removal: Rate value = 1828.80 SpikeMaxRainRate = 999.00
2022-01-24 19:35:20.896 SendEmail: Sending email, to [emailaddressremoved], subject [Cumulus MX Alarm], body ["A Cumulus MX alarm has been triggered.\r\nA data spike from your weather station ha" +
"s been suppressed.\r\nLast error: Rain rate greater than spike value - value = 182" +
"8.80mm/hr"]...

How do I clear this error?

Running Cumulus MX 3.14.1 b3160

Thanks,
Lance

[Gyvate]

The answer lies in the log excerpt you provided:
look up your Jan22log.txt file in the ...\CumulusMX\data directory and find the minute time stamp there inside - in this line the spike will be and you can correct it there. Timestamp=1/24/2022 11:35:00 AM
find 1828.80 in the line and replace it with value which corresponds to that day and save the file - no rain: 0.0, if it rained you have to figure out the proper value from some other souce.

Make sure you complete the editing and save the file before the end of the minute in which you opened it (and open it 3 seconds after the full minute - you can see in your File Explorer when the time stamp of the file changes to the actual minute) - otherwise you are going to lose the new entry which comes in at the start of the minute which you would overwrite if you kept that file open longer and saved it only then - or wait until February, when a new month file Feb22log.txt is created and correct the entry only then. The error message will most likely no longer appear in February whether you corrected it or not as the Jan22log.txt file will no longer be considered at startup.

You can just open the file any time to locate the wrong entry - once you have it located, close the file without saving and open it again a few seconds after the full minute as decribed above and then make the change and save it before the minute completes.

[water01]

Rather than go through all the above just edit Jan22log.txt with CumulusMX stopped and once the reading has been changed on that line, save the file and restart CumulsMX.

I didn't think Spike removal stopped CumulusMX running, I thought it just ignored the erroneous reading, in which case you could correct the reading in the Data Log editor. If I am wrong you can just edit it with CumulusMX stopped as I said.

[mcrossley]

If the spike removal kicks in - as it has here - then the erroneous data value will not be logged in the data file, nor will it affect the records etc.

And as David said, it does not "prevent Cumulus MX from running", it just drops that one suspicious value.

[Gyvate]

Rather than go through all the above just edit Jan22log.txt with CumulusMX stopped and once the reading has been changed on that line, save the file and restart CumulsMX.

stopped or not stopped - if the procedure takes longer than one minute the current minute(s) reading(s) will be lost. Hence maybe not so smart.

I didn't think Spike removal stopped CumulusMX running, I thought it just ignored the erroneous reading, in which case you could correct the reading in the Data Log editor. If I am wrong you can just edit it with CumulusMX stopped as I said.

if the edit can be done in the data log editor, then this would be the approach not to lose any current readings

[lgeitner]

I did search through all the files in the data folder looking for value of 1828.80. There was not a single file in the data folder that had the value of 1828.80. The only two files I did not search where diary.db and cumuluxmx.db.

Thanks,
Lance

[water01]

I did search through all the files in the data folder looking for value of 1828.80. There was not a single file in the data folder that had the value of 1828.80. The only two files I did not search where diary.db and cumuluxmx.db.

Thanks,
Lance

That is because the spike removal would not allow that reading to be set.

You need to use the CumulusMX Data Log editor and look for the line that was written on 1/24/2022 at 11:35:00 AM and fix any values in that record that need fixing.