Open cwaazywabbit opened 4 years ago
I used to have all commands require a -w if you remember back a few years ago. I thought that would still work but maybe I removed the -w on some commands to clean up the code paths. I'll look into it, but also, here are some thoughts/suggestions:
I suggest each user works off their own vm/cloud instance, each with celerystalk installed. Unless you have a really large machine and bump up the celery concurrent tasks to something very high [vi config.ini], multiple users might have to wait a considerable amount of time until their jobs get processed by celery.
If you do need to use a shared machine, one suggestion is to copy the vanilla celeryststalk repo to a new folder for each tester. /opt/celerystalk-user1
, /opt/celerystalk-user2
, etc. All of the dependancies should still work as everything is called via relative path.
Workspaces were designed for one tester to switch between a few jobs they are running at roughly at the same time. Celery has the ability to track tasks in multiple unique queues, but I just never went that deep into it.
At the moment we are able to create multiple Workspaces and have different scopes, however, it is not possible to work on more than one workspace at the same time.
Would it be possible to keep the current behaviour but allow selecting the Workspace as a commandline argument in order work on multiple scopes simultaneously?
This would allow us to put celerystalk on a shared host, where multiple pentesters could be working on the tool, for instance.
If this is already possible, please ignore my ignorance, but I haven't found a solid proof in the documentation.
PS: I am still thinking about
workspace switch -w <ws>
as a possible solution -- but not sure if technically we would run into issues.