Open averagehat opened 9 years ago
That would be really cool to include the test in at least one of them, perhaps robot test first if that's easier. It might make a lot of server side testing easier if we later want to increase the converge of testing, or adding new features on solr side.
Currently it looks like the the download takes a long time, althought he problem may be in the script. We can get around Travis choking the build if it's longer than 10 minutes, although I'd rather have our builds be shorter. https://travis-ci.org/NCBI-Hackathons/seqr/builds/78597840
Right, the downloading of whole solr dist might be a waste of time and bandwidth for every build. Maybe we can skip the server test in travis for now? The "robot" seems enough to cover those tests.
I think the way to approach this is to include the solr distribution in our repo
Was the solr dist download the blocking issue?
-----Original Message----- From: Mike Panciera [notifications@github.commailto:notifications@github.com] Sent: Tuesday, September 22, 2015 06:40 PM Eastern Standard Time To: NCBI-Hackathons/seqr Cc: Han, Lianyi (NIH/NLM/NCBI) [C] Subject: Re: [seqr] Add server tests to travis (#32)
I think the way to approach this is to include the solr distribution in our repo
— Reply to this email directly or view it on GitHubhttps://github.com/NCBI-Hackathons/seqr/issues/32#issuecomment-142441486.
Yes. Maybe hosting on S3 would be faster somehow too but putting it on the repo is fine for now I think
On Tue, Sep 22, 2015 at 6:49 PM, Lianyi notifications@github.com wrote:
Was the solr dist download the blocking issue?
-----Original Message----- From: Mike Panciera [notifications@github.com<mailto: notifications@github.com>] Sent: Tuesday, September 22, 2015 06:40 PM Eastern Standard Time To: NCBI-Hackathons/seqr Cc: Han, Lianyi (NIH/NLM/NCBI) [C] Subject: Re: [seqr] Add server tests to travis (#32)
I think the way to approach this is to include the solr distribution in our repo
— Reply to this email directly or view it on GitHub< https://github.com/NCBI-Hackathons/seqr/issues/32#issuecomment-142441486>.
— Reply to this email directly or view it on GitHub https://github.com/NCBI-Hackathons/seqr/issues/32#issuecomment-142443003 .
The logic to start server (running
make
) could happen in.travis.yml
or within a robot test.This will add some time to the tests as it takes some time to download solr, I don't think that we will hit memory limits on travis.
Related to #17