Wednesday, March 7, 2018

Arrggghh

Feel like I'm in a pool of quicksand at work.    Last night on my ground network where I think I grok how everything works one of the networks was generating nutso results on its coordinate step, moving stations by tens of meters --  I must check today to see if somehow two sites have the same name or one site has two names.   Its the only thing I can think of that might explain things working fine for all the prelim steps which are separate for each station or baseline and then blowing up like that in the network step.

I was made the lead for archiving data and received a notice of being overspent,  then a list of account keys w/o the archive names attached to them  and this morning an email that we are now right at limit but not over.   *I* did not remove nearly 40% of files so how did that happen?!

Having an issue with processing some older data where an error shows up only when a whole month of data is submitted for processing at once but not if days are done one at a time.  Those are almost impossible to track down.  Plus getting a mystery error in one processing step that does repeat but doesn't make sense, so it may be some goofy array overflow.   I don't know how to analyze software to find bugs of that nature.  Such a miserable pain in the butt. 

No comments: