Add cx , varobs namelists for EPSMWS obsgroup - #246
Conversation
|
Tests in sith I have built the code on the EX matchines and have run a single cylc of my sith branch for EPSMWS. Change performs as expected. see output here |
|
This is now ready for review. |
Chris Thomas (ctgh)
left a comment
Chris Thomas (ctgh)
left a comment
There was a problem hiding this comment.
Thanks for adding this. Would you be able to add a CX writer test too, please?
|
The additional ctests are ready now and work - do you require anything else? |
Chris Thomas (ctgh)
left a comment
Chris Thomas (ctgh)
left a comment
There was a problem hiding this comment.
Thanks for adding the CX writer test. Can you please add a 'ready to merge' label and also retrigger the CI? It looks like there was a transient problem which should disappear if it is rerun.
|
Any idea why the CI tests are failing Matt Shin (@matthewrmshin) ? i have retriggered several times over several days |
I do wonder if we should just use our own CI rather than trying to use there image for this testing. opsinputs is only ever going to be used in our environment. |
|
Michael Cooke (@mikecooke77) Shall we go ahead and merge this, given it passes the tests on our local system? |
yes or advise me what to do to get CI working please |
|
bump - any progress here ? |
|
I'll just bypass the rules. It shouldn't affect our internal CI, but I will keep an eye on the nightlies. |
|
grazias |
|
Most likely because CI image is old. Just checked. The image jcsda/docker-clang-mpich-dev:latest has eckit 1.24.5 and the recent change to ioda requires a more up-to-date version of eckit. |
|
Matt Shin (@matthewrmshin) Thanks for looking into that. Is there anything that can be done from our side to get that up to date? |
|
Chris Thomas (@ctgh) Yes, (and bearing in mind that this is a public repo like orca-jedi), we can do what orca-jedi is doing and build our own container image for CI purpose and storing the image under GitHub container registry. |
5 participants
This change adds namelist files specifying cx and varobs fields for obsgroup EPSMWS (mw sounder on MetopSG). These are identical to SternaMWS in terms of content. I also noticed that the specification for default varobs fields for both Sterna and EPS were incorrect - the bias predictors are specified twice.