For the sub-zero fix, did you confirm if this is now working as expected? Also, rather than documenting how to fix the code, it might be better to upload the latest version of the code once we'll have fixed all the issues (although I'll try to document what we did in the code).
For the crash, the output from grep is interesting: - first, it confirms that when weewx is generating an archive record, it is pulling data from the station (hence the duplicate every 2 minutes). - since the loop packet and the archive record are generated within a very short time, it is likely that this is what's causing the issue (as I mentioned, that's something I ran into when I initially tested the driver). I wonder if it makes sense to modify the driver to return the latest 'cached' packet when several attempts are made to fetch the data within a short period of time (less than 10 seconds maybe?). But let's first try to fix this with the configuration: can you try setting record_generation = software to see if it helps to avoid this issue (after the change, leave weewx running for some time then redo the grep)? Using hardware doesn't really make sense anyway for this station: as far as I can tell the WS-3000 has no concept of loop vs archive packets, all you can do it fetch the current values from the station. So if this works as I expect when using 'software', you could generate loop packets every 30 or 60 seconds, and let weewx generate the archive record every 2 to 5 minutes. weewx will aggregate the data from all the previous loop packets to produce the min/max during this 2 to 5 minutes interval but won't fetch the data again, thus avoid the issue. -- 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]. For more options, visit https://groups.google.com/d/optout.
