I understand that you have three (3) separate RINEX files from three different sites that fail when submitted to OPUS. The error message indicates that there is more than one MARKER record in each individual file indicating the possible existence of multiple points being occupied during the observations session.
As a plain text file, opening the RINEX file is possible in a text editor. The ??search? function in the editor should be used to find instances of ??MARKER.? An option would be to manually delete the errant lines. I would not.
I would note the start and end times of the records between MARKER recordsI for use with the UNAVCO utility ??teqc?.
Working from memory, I would use the following command line to create a new ??windowed? file.
An made-up example follows:
For site name: alfa, day-of-year: 300, session a, my observation file would be: alfa300a.19o
If I have an extraneous MARKER record at 2019 08 08 05 12 33 15 and an ending date/time of 2019 08 05 19 33 15, I can create a new file with the following syntax: teqc -st 201908051235 +dn 7.0 alfa300a.19o > alfa3000.19o This command uses the original file, specifies the new starting time and duration of the file in hours and redirects the output to a new file: alfa3000.19o
Reference:https://www.unavco.org/software/data-processing/teqc/doc/UNAVCO_Teqc_Tutorial.pdf
Another option, depending on your translator to RINEX is to set a start and end time as parameters for the translation,
See other replies for corrections to mine as well as other potentially better suggestions.
HTH,
DMM