simplify AWS setup

.gitignore

@@ -19,3 +19,8 @@ node_modules
 /kiwix-tools
 
 bin/zimdump
+
+*.tfstate
+*.tfstate.*
+*.terraform
+*.terraform.*

Dockerfile

@@ -1,34 +1,34 @@
-# This Dockerfile creates a self-contained image in which mirrorzim.sh can be executed
+# This Dockerfile creates a self-contained image in which mirrorzim.sh can be executed.
+#  It also runs ipfs daemon.
 #
 # You can build the image as follows (remember to use this repo as context for the build):
-#  docker build . -f Dockerfile -t distributed-wikipedia-mirror
+#  docker build . --platform=linux/amd64 -f Dockerfile -t distributed-wikipedia-mirror
 #
-# You can then run the container anywhere as follows
-#  docker run --rm -v $(pwd)/snapshots:/github/workspace/snapshots -v $(pwd)/tmp:/github/workspace/tmp distributed-wikipedia-mirror <mirrorzim.sh arguments>
-#  NOTE(s):
-#   - volume attached at /github/workspace/snapshots will contain downloaded zim files after the run
-#   - volume attached at /github/workspace/tmp will contain created website directories after the run
+# You can then run the container anywhere as follows:
+#  docker run --ulimit nofile=65536:65536 -p 4001:4001/tcp -p 4001:4001/udp distributed-wikipedia-mirror <mirrorzim_arguments>
 
-FROM openzim/zim-tools:3.1.0 AS openzim
+FROM stedolan/jq:latest AS jq
+FROM openzim/zim-tools:3.1.0 AS zimdump
+FROM ipfs/go-ipfs:v0.12.0 AS ipfs
+FROM node:16
 
-FROM node:16.14.0-buster-slim
+RUN apt-get update && apt-get install --no-install-recommends --assume-yes rsync moreutils
 
-RUN apt update && apt upgrade && apt install -y curl wget rsync
+COPY --from=jq /usr/local/bin/jq /usr/local/bin/
+COPY --from=zimdump /usr/local/bin/zimdump /usr/local/bin/
+COPY --from=ipfs /usr/local/bin/ipfs /usr/local/bin/
 
-COPY --from=openzim /usr/local/bin/zimdump /usr/local/bin
+COPY assets /root/assets
+COPY bin /root/bin
+COPY src /root/src
+COPY tools /root/tools
+COPY mirrorzim.sh package.json tsconfig.json /root/
 
-COPY tools/docker_entrypoint.sh /usr/local/bin
+RUN mkdir /root/snapshots /root/tmp
+RUN cd /root && yarn
 
-RUN mkdir -p /github/distributed-wikipedia-mirror
-RUN mkdir -p /github/distributed-wikipedia-mirror/snapshots
-RUN mkdir -p /github/distributed-wikipedia-mirror/tmp
-RUN mkdir -p /github/workspace
+EXPOSE 4001/tcp
+EXPOSE 4001/udp
 
-COPY . /github/distributed-wikipedia-mirror
-
-RUN cd /github/distributed-wikipedia-mirror && yarn
-
-VOLUME [ "/github/workspace" ]
-
-WORKDIR /github/distributed-wikipedia-mirror
-ENTRYPOINT [ "docker_entrypoint.sh" ]
+WORKDIR /root
+ENTRYPOINT [ "tools/entrypoint.sh" ]

README.md

@@ -254,21 +254,25 @@ $ ./mirrorzim.sh --languagecode=cu --wikitype=wikipedia --hostingdnsdomain=cu.wi
 ## Docker build
 
 A `Dockerfile` with all the software requirements is provided.
-For now it is only a handy container for running the process on non-Linux
-systems or if you don't want to pollute your system with all the dependencies.
-In the future it will be end-to-end blackbox that takes ZIM and spits out CID
-and repo.
+It is an end-to-end blackbox that takes mirrorzim.sh arguments, spits out CID
+and runs IPFS daemon.
 
-To build the docker image:
+To run the publicly available docker image:
 
 ```sh
-docker build . -t distributed-wikipedia-mirror-build
+docker run --ulimit nofile=65536:65536 -p 4001:4001/tcp -p 4001:4001/udp public.ecr.aws/c4h1q7d1/distributed-wikipedia-mirror:latest <mirrorzim_arguments>
 ```
 
-To use it as a development environment:
+Alternatively, to build the docker image:
 
 ```sh
-docker run -it -v $(pwd):/root/distributed-wikipedia-mirror --net=host --entrypoint bash distributed-wikipedia-mirror-build
+docker build . --platform=linux/amd64 -f Dockerfile -t distributed-wikipedia-mirror
+```
+
+And then, to run it:
+
+```sh
+docker run --ulimit nofile=65536:65536 -p 4001:4001/tcp -p 4001:4001/udp distributed-wikipedia-mirror <mirrorzim_arguments>
 ```
 
 # How to Help
@@ -340,7 +344,3 @@ We are working on improving deduplication between snapshots, but for now YMMV.
 ## Code
 
 If you would like to contribute more to this effort, look at the [issues](https://github.com/ipfs/distributed-wikipedia-mirror/issues) in this github repo. Especially check for [issues marked with the "wishlist" label](https://github.com/ipfs/distributed-wikipedia-mirror/labels/wishlist) and issues marked ["help wanted"](https://github.com/ipfs/distributed-wikipedia-mirror/labels/help%20wanted).
-
-## GitHub Actions Workflow
-
-The GitHub Actions workflow that is available in this repository takes information about the wiki website that you want to mirror, downloads its' zim, unpacks it, converts it to a website and uploads it to S3 as a tar.gz package which is publicly accessible.

mirrorzim.sh

@@ -16,7 +16,6 @@ usage() {
 	echo "    [--hostingdnsdomain=<HOSTING_DNS_DOMAIN>]"
 	echo "    [--hostingipnshash=<HOSTING_IPNS_HASH>]"
 	echo "    [--mainpageversion=<MAIN_PAGE_VERSION>]"
-    echo "    [--push=<true|false>]"
 	echo ""
 	echo "OPTIONS"
 	echo ""
@@ -28,7 +27,6 @@ usage() {
     echo "  -d, --hostingdnsdomain   string     - the DNS domain name the mirror will be hosted at e.g. tr.wikipedia-on-ipfs.org"
 	echo "  -i, --hostingipnshash    string     - the IPNS hash the mirror will be hosted at e.g. QmVH1VzGBydSfmNG7rmdDjAeBZ71UVeEahVbNpFQtwZK8W"
 	echo "  -v, --mainpageversion    string     - an override hack used on Turkish Wikipedia, it sets the main page version as there are issues with the Kiwix version id"
-	echo "  -p, --push               boolean    - push to local ipfs instance (defaults to true)"
     exit 2
 }
 
@@ -68,10 +66,6 @@ case $i in
     MAIN_PAGE_VERSION="${i#*=}"
     shift
     ;;
-    -p=*|--push=*)
-    PUSH="${i#*=}"
-    shift
-    ;;
     --default)
     DEFAULT=YES
     shift
@@ -116,10 +110,6 @@ if [ -z ${MAIN_PAGE_VERSION+x} ]; then
 	MAIN_PAGE_VERSION=""
 fi
 
-if [ -z ${PUSH+x} ]; then
-    PUSH="true"
-fi
-
 printf "\nEnsure zimdump is present...\n"
 PATH=$PATH:$(realpath ./bin)
 which zimdump &> /dev/null || (curl --progress-bar -L https://download.openzim.org/release/zim-tools/zim-tools_linux-x86_64-3.0.0.tar.gz | tar -xvz --strip-components=1 -C ./bin zim-tools_linux-x86_64-3.0.0/zimdump && chmod +x ./bin/zimdump)
@@ -154,11 +144,14 @@ node ./bin/run $TMP_DIRECTORY \
   ${HOSTING_IPNS_HASH:+--hostingipnshash=$HOSTING_IPNS_HASH} \
   ${MAIN_PAGE_VERSION:+--mainpageversion=$MAIN_PAGE_VERSION}
 
-if [[ "$PUSH" == "true" ]]; then
-    ./tools/add_website_to_ipfs.sh "$ZIM_FILE" "$TMP_DIRECTORY" "-p"
-else
-    printf "\n\n-------------------------\nD O N E !\n-------------------------\n"
-    printf "ZIM: $ZIM_FILE\n"
-    printf "TMP: $TMP_DIRECTORY"
-    printf "\n-------------------------\n"
-fi
+printf "\nAdding the processed tmp directory to IPFS\n(this part may take long time on a slow disk):\n"
+CID=$(ipfs add -r --cid-version 1 --pin=false --offline -Q -p $TMP_DIRECTORY)
+MFS_DIR="/${ZIM_FILE}__$(date +%F_%T)"
+
+# pin by adding to MFS under a meaningful name
+ipfs files cp /ipfs/$CID "$MFS_DIR"
+
+printf "\n\n-------------------------\nD O N E !\n-------------------------\n"
+printf "MFS: $MFS_DIR\n"
+printf "CID: $CID"
+printf "\n-------------------------\n"