larsjohannessen
commented
7 months ago Through a day or two of trial and error, I have found that I can almost reliably (75% of the time) dump stats in 10-20 register chunks if I break the list of registers I'm trying to dump up into the following groups (expressed in Python list[] format).
register_chunks = ["1,2,6,9,15,16,19,20,25,26",
"30,31,33,84,85,88-110,112,209,282",
"321-323,325,340-342,344,346,347,362",
"400,401,404,412,413,419,501,502,565,567",
"740,741,745-747,813,900,901,903,908",
"1103-1117,1119",
"1124-1126,1134-1136,1146-1157,1164,1165,3000,3001",
"3322,3323,3325-3327,3330-3332,3422-3425,3522-3524,3808,3903-3906",
"12_005,12_006,12_309,12_310,31_003,31_007,31_008,31_009,31_010,31_013,31_016,31_019,31_022,31_109,31_110,31_200"]Then, using a loop, I can iterate through the list and feed each chunk to aurora_fetch and usually get responses (about 50% of the time) without a timeout. If I add a 3-5 second sleep() at the end of each loop iteration to force a pause between successive calls to aurora_fetch, that gets me up to about a 75% success rate.
Admittedly, this is a bit of a hack, so any thoughts on how to make aurora_fetch perform more reliably/not timeout would be welcome!
First off, really handy tool - thanks for releasing it.
I've had some small successes querying a limited number of registers and have successfully gotten data back:
Unfortunately, when I try to query a larger number of registers, I get the following
ModBus::Errors::ModBusTimeouterrors:Is there something I can do to extend the timeout period? I believe I'm executing the
aurora_fetchcommand correctly, and am trying to follow the section of the README.md that begins with "This is the easiest way to make a "dump" of your system to send if you're having problems".Please let me know if there's any other information I can provide to assist with debugging. This looks like a really useful tool! To that end, here is some other generic information you might find useful: