That's really good to know. All machines are up and running. All of them use the same subject (OAS1_0001_MR1) from OASIS-Brains for analysis. I have a minimum archive of the current FreeSurfer 6.0.0 release, named "freesurfer-linux-x86_64-pipeline-6.0.0.tar.gz" (~ 400 MB) that is only capable to run "recon-all" with a subject.
My hardware is quite old, so it will take some time until I get the results. The current test setup is up and running. It looks as follows: OpenStack Ocata (4 VMs on 4 compute nodes): - neuroimaging01 (Ubuntu 16.04.2 LTS, freesurfer-linux-x86_64-pipeline-6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 - neuroimaging02 (Ubuntu 16.04.2 LTS, freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 - neuroimaging03 (Ubuntu 16.04.2 LTS, freesurfer-linux-x86_64-pipeline-6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all - neuroimaging04 (Ubuntu 16.04.2 LTS, freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0.tar.gz): screen -A -m -d -S ubuntu strace -y -v -f -s 100000 -e trace=file -o recon-all.trace recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 vSphere 6.0.0 (2 VMs): - neuroimaging01 (Ubuntu 16.04.2 LTS, freesurfer-linux-x86_64-pipeline-6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 - neuroimaging02 (Ubuntu 16.04.2 LTS, freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 Proxmox Virtual Environment (2 Containers): - msmc-qs-quick (Ubuntu 16.04.2 LTS, freesurfer-linux-x86_64-pipeline-6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 - msmc-build-st (Ubuntu 16.04.2 LTS, freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 Hardware (1 PC): - msmc-quickcheck (Ubuntu 16.04.2 LTS, freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0.tar.gz): recon-all -sd /home/ubuntu/freesurfer/subjects -s OAS1_0001_MR1 -all, DIAG=0xFFFFFFFF, DIAG_VERBOSE=1 I will check the results tomorrow... At the moment, everything looks good and doing its job. When I encounter a failure, I will upload the working directories of the corresponding runs and contact you via mailing list. Thank you very much! Regards, Jens Gesendet: Mittwoch, 31. Mai 2017 um 15:31 Uhr Von: "Bruce Fischl" <fis...@nmr.mgh.harvard.edu> An: "Freesurfer support list" <freesurfer@nmr.mgh.harvard.edu> Betreff: Re: [Freesurfer] DIAG influencing recon-all pipeline? Hi Jens we need more information if we are to help you. Can you send the recon-all.log? Or at least the bit that shows the command that fails onwards? The DIAG stuff should not change recon-all, but of course it might and if so we will track it down and fix it cheers Bruce p.s. The HIPS warning just means it couldn't recognize an input file. We don't support HIPS after we transitioned to open source ages ago as their license wasn't compatible On Wed, 31 May 2017, Jens Offenbach wrote: > Hallo, > first of all: Sorry to be a little bit clingy, but I have some problems in > understanding regarding the DIAG option in FreeSurfer. As suggested by Bruce, > I have set DIAG to 0xFFFFFFFF and DIAG_VERBOSE to 1. I am getting a lot of > output now and a very large recon-all.log. > > Unfortunately, I am not able to run a recon-all on the OASIS-Brains subject > "OAS1_0001_MR1" and "OAS1_0004_MR1" successfully after converting the *.img > to *.mgz via mri_convert. > > I order to get FreeSurfer in the cloud, I have created a minimum archive of > FreeSurfer 6.0.0 using strace to monitor all files that are being accessed in > the course of recon-all. The archive does not seem to work with DIAG set to > 0xFFFFFFFF. > > I am not sure, but my first guess is that the DIAG option also influences the > recon-all pipeline. Is this right? Is there a way to increase the log level > without influecing any analysis steps? The whole run takes much longer than a > "normal" recon-all without DIAG set to any level. My current run fails with: > > ... > HIPS unsupported > No such file or directory > Linux neuroimaging02 4.8.0-53-generic #56~16.04.1-Ubuntu SMP Tue May 16 > 01:18:56 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux > > recon-all -s OAS1_0001_MR1 exited with ERRORS at Wed May 31 02:54:32 UTC 2017 > > I will prepare a new run with strace and the original FreeSurfer-archive to > check process behavior when DIAG is set to 0xFFFFFFFF. Unfortunately, > everything takes so much time... > > I have a prepared a large test scenario with different hardware > constellations to check if the encountered SegFault is hardware on not > software related. > > Thank you very much! > > Regards, > Jens > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer[https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer] The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline[http://www.partners.org/complianceline] . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
