Iris chunking policy for netcdf files - new fix applied

[rswamina]

While writing CMOR-ization scripts for the FLUXCOM data set I encountered a problem - The program hung each time I tried to access the data (for writes and saves). After consulting with @valeriupredoi and @mattiarighi I posted this issue on the SciTools issue page. It looks like the problem was with chunk size in netcdf files. I am reporting here two issues (threads) from there which clarify the fix applied for the problem. It appears as though the fix was applied to v2.3.0 but the esmvaltool distribution links to iris v2.2.1?

I just wanted to bring this to the core developers' attention and also ask what I should do for this fix to propagate through to my environment for esmvaltool on JASMIN.

SciTools/iris#3361

SciTools/iris#3362

[bjlittle]

Note this comment in issue SciTools/iris#3362.

The fix for this issue is PR SciTools/iris#3361 which will be included in iris release 2.3.0, scheduled for release on 30 August, see https://github.com/SciTools/iris/milestones

[valeriupredoi]

awesome, cheers a bunch @bjlittle - we are breathlessly looking forward to that 2.3.0 😁
PS am back from holidays if you want me to test anything 👌

[bouweandela]

@rswamina Does the iris 2.3 release solve this issue?

[bjlittle]

We'll be cutting iris 2.4.0 soonish (the last to support both python2+3)

It would be good to know whether your issues are resolved...

[valeriupredoi]

cheers for the heads up @bjlittle - looking forward to it 🍺 @rswamina it'd be good to test with the current 2.3.0 🍺

[rswamina]

Will test and check with the pre-chunked data.

[mattiarighi]

@rswamina any news here?

[github-actions]

In order to maintain a backlog of relevant issues, we automatically label them as stale after 1080 days of inactivity.

If this issue is still important to you, please comment below to remove the stale label. Otherwise, this issue will be automatically closed in 60 days.