XNAT Container Service (Freesurfer) - Creation Time Timestamp Issue.

[Richard Cole]

First, initial impressions of XNAT 1.8.  Awesome!   Lots of Cool additions.

So, I am running into an issue after Freesurfer finishes and it runs into an issue related to "Creation Time Stamp", which from the logs appears to not be recorded properly for some reason.

By the way, don't think this is a 1.8 issue, because I (just last week) had the same issue with 1.7.6, running Container Service 2.0... Most probably a setup issue on my part.

Has anyone else seen this and also wondering how to "sync" the time zone between your XNAT Server and the Docker Container running the Freesurfer Pipeline.

Server is CST

Container is UTC

Related Information: XNAT 1.8 - Container Services 3.0.0 - Freesurfer-Common-1.0.4 (DataType) - Container (xnat/freesurfer711-xnat) 

Results from "stderr"

rm: No match.

Unable to read generating program version from CVS version information

No creation time found in /output/20190709_001/stats/aseg.stats; using dummy timestamp.

Unable to read generating program version from CVS version information

Unable to read generating program version from CVS version information

Unable to read generating program version from CVS version information

No creation time found in /output/20190709_001/stats/wmparc.stats; using dummy timestamp.

Results from "stdout" (tail end...)

Started at Fri Mar 12 23:26:06 UTC 2021 

Ended   at Sat Mar 13 05:59:54 UTC 2021

#@#%# recon-all-run-time-hours 6.563

recon-all -s 20190709_001 finished without error at Sat Mar 13 05:59:54 UTC 2021

done

Finished recon-all

Sat Mar 13 05:59:54 UTC 2021

Generating assessor XML

Sat Mar 13 05:59:54 UTC 2021

/usr/local/bin/stats2xml_fs7.1.pl -o /output -p rtctest -x DevXNAT1_8_E00001 -t Freesurfer -f DevXNAT1_8_E00001_freesurfer_20210313055954 -m 3 /output/20190709_001/stats

all right then.

Processing /output/20190709_001/stats...done.

Finished generating assessor XML

Sat Mar 13 05:59:54 UTC 2021

Attached ContainerHistory.png, which shows some unusual timestamp information.

Attached BuildResults.png, seems to be complete..

Attached Command.txt

[Sarah K]

Hi, There's been some discussion relating to this container in another thread: https://groups.google.com/g/xnat_discussion/c/SDmO_eZtSHY/m/YDqobAfcBQAJ

Summary: I was the one that put together the xnat/freesurfer711-xnat container. The container uses a very old Perl script to generate the assessor XML that always gives timestamp warnings/errors when reading Freesurfer files, but the assessor XML still gets generated correctly. There was a data field missing in the freesurfercommon-1.0.4 schema that is needed for uploading Freesurfer 7 results. If you are having trouble getting Freesurfer 7 results to upload after running the container, try updating your freesurfercommon plugin jar with the 1.0.5 jar here: https://bitbucket.org/nrg_customizations/nrg_plugin_freesurfercommon/downloads/?tab=downloads

Unfortunately I can't help with timestamp syncing between the internal container bash scripts and XNAT though.

-Sarah

[Richard Cole]

Thanks , I have downloaded the .jar and am running a test now..  Appreciate your help.

Will give an update when completed..  

--
You received this message because you are subscribed to a topic in the Google Groups "xnat_discussion" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/xnat_discussion/SBHSUApuyZI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to xnat_discussi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/xnat_discussion/9598c982-c326-4d04-b464-3438eddf8158n%40googlegroups.com.

[Richard Cole]

Sarah, after downloading the "cnda-freesurfer-common-1.0.5.jar" to replace "freesurfer-common-1.0.4.jar" we were able to successfully run the "xnat/freesurfer711-xnat" container service pipeline and it successfully uploaded the data to the project..   I will continue to look into the "timestamp" errors, but they don't seem to impact the outcome.   We will continue to test, and let you know if we figure out that part... 

Thanks for your help!

[Sarah K]

Great, thanks for the update!