Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Multiple GenomicsDBImport commands can't be run concurrently in the same directory? #4106

Closed
ldgauthier opened this issue Jan 9, 2018 · 4 comments

Comments

@ldgauthier
Copy link
Contributor

There was an unverified bug report from a user on the forum:
"I did also run in to a problem with running all of these imports from the same directory. It creates a .tmp_loader.json file in the directory you kick the imports off from, which is problematic when you kick off 24 imports, one for each chromosome. This json file seems to specify the location of the GenomicsDB which that job is to write to. With that file being overwritten each time I kicked off a new import, I got core dumps and jobs writing to the wrong db. I had specified a unique db for each, but with jobs reading from the same json file, what I had specified on the command line and what the jobs actually did didn't line up. I ended up creating a separate dir for each chromosome, and kicked each job off from those dirs, which solved the problem."

I haven't tried to reproduce it yet. I've created multiple GDBs by running commands in the same directory, though only one at a time. I can't find any .tmp_loader.json files, but they may be deleted on completion.

@lbergelson
Copy link
Member

This is very believable. I've certainly never tried running multiple in the same directory...

@kgururaj
Copy link
Collaborator

That sounds like a bad implementation from our side - apologies.

We are working on fully fixing the Protobuf implementation. As part of that task, the temp loader/query JSON files will no longer be created (the vid and callset JSONs will still be needed).

@droazen
Copy link
Contributor

droazen commented Jan 11, 2018

Assigning to @kgururaj for a fix.

@droazen droazen added this to the Engine-4.1 milestone Jan 11, 2018
kgururaj added a commit to Intel-HLS/GenomicsDB that referenced this issue Jan 12, 2018
temp file function for creating loader JSON
@droazen droazen modified the milestones: Engine-4.1, Engine-4.0.1.0 Jan 29, 2018
@droazen
Copy link
Contributor

droazen commented Jan 29, 2018

Resolved by #4261

@droazen droazen closed this as completed Jan 29, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants