The download will use a new gluster volume that will be created by infrastructure. (src-d/infrastructure#585). Also all processes will run in a namespace called pga.
[x] gluster volume ready src-d/infrastructure#585
[x] create batch job that uses pga-create container to create the list of repositories. The list and other files will reside in the gluster volume.
[x] check that the list is correct and compare with the previous one
[x] rabbitmq and postgres ready src-d/infrastructure#584
[x] start borges consumers pointing the pga gluster volume and rabbitmq + db servers
The download will use a new gluster volume that will be created by infrastructure. (src-d/infrastructure#585). Also all processes will run in a namespace called
pga
.pga-create
container to create the list of repositories. The list and other files will reside in the gluster volume.While the repositories are being downloaded check for possible problematic repositories.