Commit b37f2475 authored by Laros's avatar Laros

Notes_2017_07_04.md

parent 3487a4a3
# Work discussion 20-06-2017
Present: Jonathan Vis (JV), Jeroen Laros (JL), Pieter Neerincx,
Roan Kanninga (RK), Gerben van der Vries GV, Leon Mei (LM),
Mark Santcroos (MS).
## GoNL re-analysis
- Test runs on different clusters using different frameworks.
- Discrepancies in the results, mainly related to read groups.
- Cause seem to be the sample sheet, conversion to Molgenis format failed.
- LM proposes to move on with the analysis, dropping the Molgenis framework.
- JL proposes to start with BioPet, giving Groningen the time to fix the
problems. Once everything is working, the Groningen pipeline can be used as
well.
- RK and GV will review the dry-run logs of BioPet. LM will also give a log of
a real run.
- No X11 on shark for QC output (LM looks in to this).
## eLAN
Resolved.
## CoManage
- MS: TLS is reported to be working.
- MS: LDAPS functionality will be offered.
- PN asked about the roadmap from Surf, Irene asked about extra funding. We
agree that apart from the POC, the funding should come from NFU / SURF.
## Ansible and Easybuild integration
- PN: Improvement of the playbook is still on the list. No updates because of
all the work on GoNL.
- Playbook is on GitHub: https://github.com/bbmri-nl/bbmri-nl-pipeline-deployment.
## Other
- On 23-06-2017 there will be a hands on session on Easybuild / Ansible.
\ No newline at end of file
# Work discussion 04-07-2017
Present: Jonathan Vis (JV), Jeroen Laros (JL), Pieter Neerincx (PN),
Roan Kanninga (RK), Gerben van der Vries GV, Leon Mei (LM),
Mark Santcroos (MS).
## GoNL re-analysis
- LM is comaring Molgenis with Biopet, concordance check on chromosome 1. About
0.1% difference, there seems to be a problem with the readgroup parameter in
the Molgenis pipeline. This is hard to change. Probably this is not the
reason for the difference.
- RK and GV had a look at the dry-run logs of BioPet, hard to interpret. LM
will ask for a dot graph to provide more insight.
- We should establish the stability of the pipeline runs (run the same sample
twice) to see whether the difference is in the same order of magnitude.
## eLAN
Resolved.
## CoManage
- PN: No updates, will have a meeting with two admins half July.
- MS: LDAPS functionality is being worked on at SURF.
- MS: First idea for a SIG meeting at SURF to present these ideas. CoManage has
to work in Groningen (cluster) and/or in the SURF cloud.
## Ansible and Easybuild integration
- PN: Missing X11 libraries not the problem, it appears to be a missing pango
library. There is an other library missing, can be fixed.
- EGI competence centre: robustly deploying software was difficult, there seems
to be some movement there. Today they get a couple of VMs to test a virtual
SLURM cluster, the BBMRI environment will be deployed there. LM wants to
deploy the BIOS pipeline on the same system.
- PN/LM make a 'lessons learned' report.
- EasyBuild is deemed heavy weight by the imaging group. They prefer a solution
that uses pre-compiled software.
## Other
None.
\ No newline at end of file
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment