Experimental Physics and
Industrial Control System

Have you tried something like UNIXSECONDS FOR SECS OF VAL 3 + 1? There might be an issue of it using something like “<” rather than “<=”.

 

 

 

Hi all,

 

I have a question about using the different endpoints and the timestamp resolution of the archiver. Here is the situation that is confusing me:

 

1. I make a request to the getData endpoint for some <PV> that returns multiple values, each accompanied by a secs and nanos field indicating when the value was updated

Example response for pv ID:  [{ val: “2”, secs: “1627576833”, nanos: “342322864”}, {val: “3”, secs: “1627576922”, nanos: “563728637”} ]

2. When I make a request to getDataAtTime where the at parameter contains the full returned timestamp, down to the nanoseconds, I get the record value before:

Example request: <ARCHIVER DATA ENDPOINT>/getDataAtTime?at=<UNIXSECONDS FOR SECS OF VAL 3>.563728637 -d [“ID”]

Example response: { “ID”: “2”}

 

This occurs over shot intervals, so time zone issues are not the underlying cause. Is there some difference in the time resolution of the different endpoints? Is this a quirk of the archiver? We are doing a work around by rounding up to milliseconds, but it is confusing that the timestamp returned in nanoseconds can’t be used precisely in getdataattime

 

Thank you,  

 

Elena De La Paz