Opened 8 years ago
Closed 8 years ago
#1560 closed Bug (fixed)
not using key_iomput makes the model crash
| Reported by: | clem | Owned by: | nemo |
|---|---|---|---|
| Priority: | low | Milestone: | |
| Component: | OCE | Version: | v3.6 |
| Severity: | Keywords: | ||
| Cc: |
Description
while testing memory leaks for a regional configuration (NEMO-LIM3 with BDY), I noticed that the model cannot run without key_iomput. It crashes with the enigmatic message concerning IOIPSL:
FATAL ERROR FROM ROUTINE mathop_r21
--> Error while executing an indexing function
--> only
To be able to run the model, the call to diawri in step.F90 must be commented. Any idea how to correct that (though I do not think this is a priority)?
Commit History (2)
| Changeset | Author | Time | ChangeLog |
|---|---|---|---|
| 5566 | clem | 2015-07-08T15:15:40+02:00 | no use of landmask if no key_iomput (avoids crashing). see ticket #1560. |
| 5565 | clem | 2015-07-08T15:15:04+02:00 | no use of landmask if no key_iomput (avoids crashing). see ticket #1560. |
Change History (2)
comment:1 Changed 8 years ago by clem
comment:2 Changed 8 years ago by clem
- Resolution set to fixed
- Status changed from new to closed
Note: See
TracTickets for help on using
tickets.
Rachid found the problem (as usual). The logical ln_mskland has to be put to FALSE.
In the case key_iomput is not activated, we should define clop as "x" in diawri.F90. Since it is not possible to mask the land with walues of 1.e30, I propose to change these lines (440:442):
By this one:
clop = "x" ! no use of the mask value
Any objection?