Update P3 document

p3/README.md

@@ -23,6 +23,7 @@ Before starting, please review the [general project directions](../projects.md).
 * Feb 24: feel free to use different tools to implement Part 2.
 * Feb 24: clarify that `bigdata.py` will be used in tests.
 * Feb 24: add link to lecture notes on parquet file operations.
+* Feb 24: remove port forwarding for `docker run` since we test server with `docker exec`
 
 ## Part 1: Communication (gRPC)
 
@@ -81,7 +82,7 @@ server like this:
 
 ```
 docker build . -t p3
-docker run -d -m 512m -p 127.0.0.1:5440:5440 p3
+docker run -d -m 512m p3
 ```
 
 The client programs should then be able to communicate with the
@@ -99,8 +100,8 @@ clients need to run.  When we test your code, we will run the clients
 in the same container as the server, like this:
 
 ```
-docker run --name=server -d -m 512m -p 127.0.0.1:5440:5440 -v ./inputs:/inputs p3   # server
-docker exec server python3 upload.py /inputs/test1.csv                              # client
+docker run --name=server -d -m 512m -v ./inputs:/inputs p3   # server
+docker exec server python3 upload.py /inputs/test1.csv       # client
 ```
 
 Note that you don't need to have an `inputs/test1.csv` file, as the
@@ -116,7 +117,7 @@ to re-run your container with newer server.py code without rebuilding
 first.  Here's an example:
 
 ```
-docker run --rm -m 512m -p 127.0.0.1:5440:5440 -v ./server.py:/server.py p3
+docker run --rm -m 512m -v ./server.py:/server.py p3
 ```
 
 ## Part 2: Upload
@@ -249,7 +250,7 @@ be able to run your client and server as follows:
 docker build . -t p3
 
 # run server in new container
-docker run --name=yournetid -d -m 512m -p 127.0.0.1:5440:5440 -v ./inputs:/inputs p3
+docker run --name=yournetid -d -m 512m -v ./inputs:/inputs p3
 
 # run clients in same container
 docker exec yournetid python3 upload.py /inputs/simple.csv