Hi,

Short answer is it depends. 

Long answer. When WeeWX starts it will attempt to download any archive 
records since the timestamp of the last good record in the archive. Not all 
drivers support this feature; if it is not supported then the catchup is 
skipped and WeeWX continues and starts receiving loop packets/archive 
records from the driver as per normal. If there was a gap between the last 
good record and the current time then those records have been lost and the 
gap will remain. If the driver does support catchup WeeWX will attempt to 
download any archive records timestamped after the timestamp of the last 
good archive record in the archive. Once the catchup is complete WeeWX 
continues and starts receiving loop packets/archive records from the driver 
as per normal. In this case what happens to the archive records that would 
have been generated during the time the catchup was running depends on 
several factors. If the driver emits loop packets only then WeeWX will 
never see these loop packets nor will any archive records be generated from 
these loop packets. If the catchup takes less than an archive period and 
WeeWX subsequently receives at least one loop packet in that archive 
period, then the archive record will be generated but only based on the 
loop packet(s) actually received by WeeWX. If the driver/station supports 
hardware record generation (ie the station provides the archive records) 
then the behaviour will depend on the station and driver. If the station 
uses some form of logger to store historical archive records then you may 
be able to get the archive records produced during the catchup, you may 
need to restart WeeWX to perform another catchup to pickup these records, 
then again they may not be stored. If there is no logger then in all 
likelihood you will have lost the records, or at best you might get the 
most recent one.

How to minimise lost records. Start WeeWX as early as possible in the 
archive period (eg for a 5 minute archive period start as soon as possible 
after a 5 minute boundary) so as to give WeeWX the maximum amount of time 
to complete the catchup. Perform a restart in the archive period after the 
catchup is complete, it may help, it may not. Consider your architecture, 
if you have more than one source for data and preventing data loss is 
important maybe running multiple WeeWX instances may help; that way a slow 
performing source will not necessarily drag down other faster sources 
during a catchup.

Gary


On Saturday, 14 December 2019 04:20:09 UTC+10, Michi Kaa wrote:
>
> Is this a known limitation or intended or just my issue because of the 
> slow reading of the REST API?
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/weewx-user/48a0d726-a463-4028-9dae-c39e9a52ba0b%40googlegroups.com.

Reply via email to