First version of architecture overview doc #28
Conversation
| @@ -0,0 +1,66 @@ | |||
| # Overview # | |||
|
|
|||
| Here it is described a brief overview of xdist's internal architecture. | |||
There was a problem hiding this comment.
strike that sentence as it doesn't add news value :)
|
apart from a few minor comments a great start! thanks! |
My pleasure. You got the idea right, but I think is worth mentioning explicitly to others that this is really just a first version to get the ball rolling, it needs refinement and some xdist aspects are not even covered (for example node recovery). |
|
it's already helpful as it is, though. |
|
👍 |
| Here it is described a brief overview of xdist's internal architecture. | ||
|
|
||
|
|
||
| `xdist` works by spawning one or more **worker nodes**, which are controlled |
There was a problem hiding this comment.
Should we drop "node" from the terminology? A "node" implies a separate host, which is not always the case. How about just a "master" and "workers"? The workers can just be separate processes on the same host, or they can be remote hosts, as you explain later.
(I think @hpk42 was advocating for this change in another issue.)
There was a problem hiding this comment.
So using just the terminology you propose:
xdistworks by spawning one or more workers, which are controlled by the master. Each worker is responsible for performing a full test collection and afterwards running tests as dictated by the master.
Does that sound good? If all agree I will apply the change to the rest of the document.
There was a problem hiding this comment.
Yep, that sounds better to me at least.
* Fixed typo * Removed superfluous introduction
|
I think I applied all change requests. 😄 |
First version of architecture overview doc
As commented in #18 in this comment. Comments, oversights and corrections are welcome!
cc @RonnyPfannschmidt @nchammas @hpk42