Writing of record length 1 & failure to write header correctly

The cf card directories are in a .tar file on the cloud at https://cloud.passcal.nmt.edu/index.php/apps/files/?dir=/Public%20Uploads/202359_ISLA_NAVARINO/202359_troubleshoot&fileid=2621892.

PW01 (9DF6) and PW02 (9E5D): one channel of one day had record length 1. The affected mseed files are PW01.X9..EH1.2024.018 and PW02.X9..EH2.2024.071.

msi -S PW01.X9..EH1.2024.018
ERROR: Record length is out of range: 1 (allowed: 256 to 1048576)
DCC|2024,232
X9|PW01||EH1|2024,018,00:00:00.000000|2024,018,17:30:55.990000||100|6305600|||||||2024,232
X9|PW01||EH1|2024,018,17:31:17.700000|2024,018,23:59:59.990000||100|2332230|||||||2024,232

msi -S PW02.X9..EH2.2024.071
ERROR: Record length is out of range: 1 (allowed: 256 to 1048576)
DCC|2024,232
X9|PW02||EH2|2024,071,00:00:00.005000|2024,071,20:11:58.595000||100|7271860|||||||2024,232
X9|PW02||EH2|2024,071,20:12:11.675000|2024,071,23:42:12.275000||100|1260061|||||||2024,232
X9|PW02||EH2|2024,071,23:42:41.625000|2024,071,23:59:59.995000||100|103838|||||||2024,232

PW01 also has a day file with 0 sps as well as 100 sps:
msi -S PW01.X9..EHZ.2024.013
DCC|2024,232
X9|PW01||EHZ|2024,013,00:00:00.000000|2024,013,13:41:19.290000||100|4927930|||||||2024,232
X9|PW01||EHZ|2024,013,13:41:19.300000|2024,013,13:41:19.300000||0|0|||||||2024,232
X9|PW01||EHZ|2024,013,13:41:40.260000|2024,013,23:59:59.990000||100|3709974|||||||2024,232

Station PW07 has a day file with the same issue as PW01 EHZ 2024 013, but I did not request that raw data as it seemed more efficient to examine PW01.

YA03 (9BB8) has a day file that contains a segment lacking correct headers.
msi -S YA03.X9..EHZ.2024.016
DCC|2024,232
X9|YA03||EHZ|2024,016,00:00:00.000000|2024,016,18:15:49.320000||100|6574933|||||||2024,232
X9|YA03||EHZ|2024,016,18:16:23.030000|2024,016,20:16:53.610000||100|723059|||||||2024,232
X9|YA03||EHZ|2024,016,20:17:15.050000|2024,016,23:59:59.990000||100|1336495|||||||2024,232
|YA03||| -1,365,00:00:00.000000| -1,365,00:00:00.000000||0|0|||||||2024,232

rt2ms version: I can inquire if it is relevant. The experiment is 202359 with a ship date (including the laptop) of 2023-09-13. All of the RT130s at this experiment are our 3-channel units. I've looked at all 3 stations in logpeek and there is nothing obviously wrong (to me) at the times of the day files I've listed, though all 3 stations later did not keep a healthy voltage.