Opened 6 years ago

Closed 6 years ago

#1570 closed help (answered)

Near-identical job failing to converge

Reported by: kieranmrhunt Owned by: um_support
Component: UM Model Keywords:
Cc: Platform:
UM Version: <select version>


Having made a few minor differences between a working example (xlhba) and a new job (xlhbe, mostly switching off routines that required fields the start dump didn't have - aerosols, river routing, etc.), I'm getting a leave file with the standard

initial Absolute Norm : 293910934849135.44
GCR( 2 ) failed to converge in 100 iterations.
Final Absolute Norm : 1539622940.1247406

error. I've attached the difference file between this and the working job - could you possibly take a look and see why it might be failing?


Attachments (1)

diff.xlhba.xlhbe (4.4 KB) - added by kieranmrhunt 6 years ago.

Download all attachments as: .zip

Change History (2)

Changed 6 years ago by kieranmrhunt

comment:1 Changed 6 years ago by grenville

  • Resolution set to answered
  • Status changed from new to closed


I got a model running with a couple of extra tweaks to the reconfiguration - please look at xkskv (my copy of your job).

In the reconfiguration quality control I checked Reset w to zero, Reset sea ice, Reset convective cloud —- you may want to experiment more with these; they may or may not be relevant for your work?

I also switched off some ancillary file updating, but I don't think that was having an effect.


Note: See TracTickets for help on using tickets.