possible bug when restarting from history file (missing val)

[jivica]

Well I have trouble with restarting from history file, think the problem could be when reading initial fields like zeta, temp, ubar ,....
then mask is not applied like it should be...
restarting his file that I am using have for example zeta (ncdump output:) like this:
zeta =
_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _,
_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, 0.06963402, 0.0698448,
0.06971419, 0.06916319, 0.06821018, 0.06693342, 0.06546789, 0.06429055,
0.06378041, 0.06314439, 0.06336054, 0.06413388, 0.06480777, 0.06512439,
land values are _,
When running model and reading initial fields from the file I got in the very first step values like 1E37,
think that mising value is leaking into wet points....
The only way I could do restart it to replace _ , and _; (those land flagged) in history file with 0

I am using rev 271..
Wondering if someone else bumped at the problem...
Thanks in advance,
cheers
Ivica

[arango]

There is not need to do so. This is taken care internally when data is read in routines nf_fread*. I made several corrections to these routines. I will recommned you to update. This works for me very nicely with any file and variable with the _FillValue attribute.