Michał Kowalczyk
d53729b201
This is a major refactor of the way manifests are loaded and handled, which will be followed by a complete rework of the loader code (which will include e.g. centralized config). Changes/fixes: - Huge part of manifest handling was refactored and untangled. - Starting without a manifest is now disallowed. This was actually accidentally broken for some time and no one complained. It also makes little sense in practice and in Graphene's overall design, e.g. it conflicts with protected argv. - Now we only allow starting by giving the executable, not manifest (the magic resolution logic was removed). - Now manifests are sent over pipes between parent and children, instead of children finding and loading them on their own. This is a preparation for the upcoming centralized manifests change. - Previously manifests were parsed 2 times on Linux and 3 times on Linux-SGX (by untrusted PAL, trusted PAL and LibOS). This is now fixed. - The common `pal_main()` now requires that the backend-specific PAL loader loads the manifest before calling it. SGX code already has to do it (for proper initialization), so let's unify this interface for all PALs. - Fix for a PAL crash when manifest size was divisible by page size (sic!). NULL termination was missing, but most of the time the padding to page size saved Graphene from crashing.
Apache
This directory contains the Makefile and the template manifest for the most recent version of Apache web server (as of this writing, version 2.4.41). This was tested on a machine with SGX v1 and Ubuntu 16.04.
The Makefile and the template manifest contain extensive comments. Please review them to understand the requirements for Apache running under Graphene-SGX.
We build Apache from the source code instead of using an existing installation. On Ubuntu 16.04, please make sure that the following packages are installed:
sudo apt-get install -y build-essential flex libapr1-dev libaprutil1-dev libpcre2-dev \
apache2-utils libssl-dev
Quick Start
# build Apache and the final manifest
make SGX=1
# run original Apache against HTTP and HTTPS benchmarks (benchmark-http.sh, uses ab)
make start-native-server &
./benchmark-http.sh 127.0.0.1:8001
./benchmark-http.sh https://127.0.0.1:8443
kill -SIGINT %%
# run Apache in non-SGX Graphene against HTTP and HTTPS benchmarks
make start-graphene-server &
./benchmark-http.sh 127.0.0.1:8001
./benchmark-http.sh https://127.0.0.1:8443
kill -SIGINT %%
# run Apache in Graphene-SGX against HTTP and HTTPS benchmarks
SGX=1 make start-graphene-server &
./benchmark-http.sh 127.0.0.1:8001
./benchmark-http.sh https://127.0.0.1:8443
kill -SIGINT %%
# you can also test the server using other utilities like wget
wget http://127.0.0.1:8001/random/10K.1.html
wget https://127.0.0.1:8443/random/10K.1.html
Running Apache with Different MPMs
The Apache server can run with several different multi-processing modules (MPMs). The two popular ones are Prefork and Worker MPMs. The Prefork MPM uses multiple child processes with one thread each, and each process handles one connection at a time. The Worker MPM uses multiple child processes with many threads each, and each thread handles one connection at a time.
The supplied Makefile allows to run Apache in both configurations:
make start-native-server # run with Prefork MPM
make start-graphene-server # run with Prefork MPM
make start-native-multithreaded-server # run with Worker MPM
make start-graphene-multithreaded-server # run with Worker MPM