Compare commits

...

47 Commits

Author SHA1 Message Date
4f9b662223 Pin the landing page to a version known to work 2024-08-29 15:04:28 +02:00
e323c2765f Added ECDC Beam Broker site to list of Firewall exceptions 2024-07-26 09:51:46 +02:00
8c1c0de927 Made scripts executable 2024-07-18 12:52:48 +02:00
56cdd0dcc4 Got automatic Bridgehead restart to work
Cherry picked from ehds2_develop branch
2024-07-18 12:47:47 +02:00
1b9af45991 Replaced tabs with spaces 2024-07-17 09:35:50 +02:00
353fc010cb Cherry picked the stop/start files from ehds2_develop branch 2024-07-17 09:28:40 +02:00
c15c3962a7 Switched to bridgehead user for git to avoid auth problems 2024-07-17 09:24:37 +02:00
e0fe39ab83 Perform git actions with the bridgehead user, to prevent authentication issues 2024-07-16 16:58:13 +02:00
cce76bab84 Updated README for the new run/stop scripts 2024-07-16 16:55:08 +02:00
013d6d219b Provided more detailed instructions on uploading data to a Bridgehead 2024-07-15 10:28:21 +02:00
cc228e20d9 Reordered the Bridgehead restart instructions
Ordered them so that first a stop is performed, then a start. This
reflects what administrators will be doing in real life. Also included a
git pull for the Bridgehead, since this will normally also be part of
the regular administrative tasks.

Also removed a bunch of instructions for the installation, since these
steps are now performed automatically by the bridgehead script.
2024-07-15 09:20:18 +02:00
153f654397 Deal with multiple data files, update repos
The check for the presence of data in the TransFAIR updating function
only worked for single CSV files. Tis commit corrects that.

Now also do a git pull on the various repos, to keep things up to date.
2024-07-12 11:07:06 +02:00
70aca8e99c Wait longer for Blaze to start 2024-07-10 14:34:14 +02:00
f925c27791 Suggested more robust Bridgehead start 2024-07-10 14:26:46 +02:00
350a4c84af Added extra detail relating to EHDS2 2024-07-10 14:20:40 +02:00
b52931e46c Pinned Blaze version 2024-07-02 14:11:34 +02:00
eb0f392e3c Croatia is having problems with a slow starting Blaze, increase Focus retry count 2024-06-13 09:25:57 +02:00
a738f1468f Fixed bug in CVS test logic
The -f test with a wildcard works OK if there is 0 or 1 CSV files but
breaks if there are 2 or more. The replacement uses ls to check if
there are any CSV files.
2024-06-07 16:08:16 +02:00
aa122d0bff Shortened national node names in config
For ECDC, the Beam broker URL is very long, so the resulting Beam ID
can be longer than 64 characters, which is too much for the CSR
generation algorithm.
2024-05-17 11:44:37 +02:00
ec4e91fade Corrected data target directory 2024-05-17 09:59:33 +02:00
c5f82bf93c Removed outdated steps relating to transFAIR 2024-05-17 09:57:31 +02:00
eb563113e0 Merge branch 'ehds2' of https://github.com/samply/bridgehead into ehds2 2024-05-17 09:50:59 +02:00
a10d0f9605 Added instructions for starting/stopping ECDC Bridgeheads 2024-05-17 09:49:39 +02:00
994b0a4124 remove time 2024-04-25 14:03:39 +02:00
593f93fffc Turned off obfuscation, because ECDC does not want this feature 2024-03-28 11:06:10 +01:00
b2b9865e8c Improved documentation for EHDS2/ECDC 2024-03-28 09:22:41 +01:00
4588960e19 Removed DKFZ URLs and certificates from EHDS2 branch 2024-03-27 16:08:32 +01:00
1b82207934 Migrated TransFAIR start logic from launch file to docker-compose
The start-up logic for TransFAIR is specific to the ECDC/EHDS2 Bridgehead,
and did not belong in the TransFAIR repo. This commit fixes that.

It also honours the change of the TransFAIR branch name for this
project, "ehds2".
2024-03-26 17:00:42 +01:00
4880c5cc9b Remove dependancy on local blazectl image 2024-03-20 11:43:43 +01:00
b9b44d2530 Using standardized conditions for starting transfair 2024-03-19 09:07:56 +01:00
0d1f425df0 Auto-build focus and transfair
This auto-build functionality is a workaround for a bug, described in README.

The README now has additional documentation for EHDS2.

The root certificate files have been expanded to contain certificates for
both the ECDC and the DKFZ setup.
2024-03-18 11:03:50 +01:00
7a33b54416 Merge branch 'ehds2' of https://github.com/samply/bridgehead into ehds2 2024-03-15 09:48:46 +01:00
40e15f4d84 Added extra stuff for EHDS2
This includes docker-compose files that include transfair plus
configuration variables specific to EHDS2.
2024-03-15 09:35:49 +01:00
99956f3477 Added instructions to get latest TransFAIR 2024-02-14 13:10:07 +01:00
d73e6ee7a3 Merge branch 'ehds2' of https://github.com/samply/bridgehead into ehds2 2024-02-14 11:18:37 +01:00
b6bfaba855 Removed unnecessary chown 2024-02-14 11:17:28 +01:00
5bd9baaff7 Corrected things that led to errors during install 2024-02-14 11:16:39 +01:00
a629d87a5f Inserted right urlInserted right urlInserted right urlInserted right urlInserted right urlInserted right urlInserted right urlInserted right urlInserted right url 2024-02-14 10:19:19 +01:00
4abe193c58 Made the file type more explicit. 2024-02-13 14:08:53 +01:00
5e8db39e2a Data needs to be readable 2024-02-13 14:05:48 +01:00
b28a48da0a Needed sudo 2024-02-13 14:02:19 +01:00
7c2e9af947 Set up sensible volumes for TransFAIR
Also gave instructions to copy over data file.
2024-02-13 14:00:21 +01:00
81e5a6ea3f Explained the file edits for EHDS2 2024-02-13 13:42:55 +01:00
a364a4b3f8 Added instructions for getting EHDS2 branch 2024-02-13 13:35:52 +01:00
9459e1a979 Inserted ECDC-specific URLs and certificates
Also updated README to include instructions for manual Beam
registration.
2024-02-13 13:19:41 +01:00
d5760ed3d0 Modified instructions for ECDC
Also included a template configuration file.
2024-02-12 16:40:32 +01:00
8af5cf3f01 Changes made for EHDS2
The EHDS2 project has initiated a use case for antibiotic resistance
under the control of the ECDC. This branch caters for the ECDC
requirements.

Specifically, it contains the ETL to transform the ECDC CSV format to
FHIR, plus additional configuration options that have been packaged into
the bbmri project.
2024-02-12 16:07:57 +01:00
17 changed files with 611 additions and 13 deletions

101
README.md
View File

@ -9,6 +9,7 @@ This repository is the starting point for any information and tools you will nee
- [Software](#software)
- [Network](#network)
2. [Deployment](#deployment)
- [EHDS2/ECDC](#ehds2-ecdc)
- [Site name](#site-name)
- [Projects](#projects)
- [GitLab repository](#gitlab-repository)
@ -87,6 +88,8 @@ The following URLs need to be accessible (prefix with `https://`):
* gitlab.bbmri-eric.eu
* only for German Biobank Node
* broker.bbmri.de
* only for EHDS2/ECDC
* ecdc-vm-ehds-test1.swedencentral.cloudapp.azure.com
> 📝 This URL list is subject to change. Instead of the individual names, we highly recommend whitelisting wildcard domains: *.dkfz.de, github.com, *.docker.com, *.docker.io, *.samply.de, *.bbmri.de.
@ -94,6 +97,34 @@ The following URLs need to be accessible (prefix with `https://`):
## Deployment
### EHDS2/ECDC
The ECDC Bridgehead allows you to connect your site/node to the [AMR Explorer](http://ehds2-lens.swedencentral.cloudapp.azure.com/), a non-public central web site that allow certified researchers to search for information relating to antiobiotic resistance, Europe-wide. You can supply the Bridgehead with data from your site in the form of CSV files, which will then be made available to the Explorer for searching purposes.
You will need to set up some configuration before you can start a Bridgehead. This can be done as follows:
```shell
sudo mkdir -p /etc/bridgehead
sudo cp /srv/docker/bridgehead/bbmri/modules/bbmri.conf /etc/bridgehead
```
Now edit ```/etc/bridgehead/bbmri.conf``` and customize the following variables for your site:
- SITE_NAME
- SITE_ID
- OPERATOR_FIRST_NAME
- OPERATOR_LAST_NAME
- OPERATOR_EMAIL
If you run a proxy at your site, you will also need to give values to the ```HTTP*_PROXY*``` variables.
When you first start the Bridgehead, it will clone two extra repositories into /srv/docker, namely, ```focus``` and ```transfair```. It will automatically build local images of these repositories for you. These components have the following functionality that has been customized for ECDC:
- *focus.* This component is responsible for completing the CQL that is used for running queries against the Blaze FHIR store. It uses a set of templates for doing this. Extra templates have been written for the ECDC use case. They can be found in /srv/docker/focus/resources/cql/EHDS2*.
- *transfair.* This is an ETL component. It takes the CSV data that you provide, converts it to FHIR, and loads it to Blaze. This will be run once, if there is data in /srv/docker/ecdc/data. A lock file in the data directory ensures that it does not get run again. Remove this lock file and restart the Bridgehead if you want to load new data.
These images will normally be rebuilt every time you restart the Bridgehead. This is a workaround to fix a bug: if you don't rebuild these images for every start, then legacy versions will be used and you will lose the new ECDC functionality. The reason for this is still under investigation.
### Site name
You will need to choose a short name for your site. This is not a URL, just a simple identifying string. For the examples below, we will use "your-site-name", but you should obviously choose something that is meaningful to you and which is unique.
@ -108,6 +139,8 @@ Site names should adhere to the following conventions:
### GitLab repository
You can skip this section if you are doing an ECDC/EHDS2 installation.
In order to be able to install, you will need to have your own repository in GitLab for your site's configuration settings. This allows automated updates of the Bridgehead software.
To request a new repository, please contact your research network administration or send an email to one of the project specific addresses:
@ -130,7 +163,24 @@ During the installation, your Bridgehead will download your site's configuration
### Base Installation
First, download your site specific configuration repository:
Clone the bridgehead repository:
```shell
sudo mkdir -p /srv/docker/
sudo git clone https://github.com/samply/bridgehead.git /srv/docker/bridgehead
```
If this is an ECDC/EHDS2 installation, switch to the ```ehds2``` branch and copy the configuration file to the required location:
```shell
cd /srv/docker/bridgehead
sudo git checkout ehds2
sudo mkdir -p /etc/bridgehead/
sudo cp bbmri/modules/bbmri.conf /etc/bridgehead/
sudo vi /etc/bridgehead/bbmri.conf # Modify to include national node name and admin contact details
```
For an ECDC/EHDS2 installation, you will also need to copy your data in a comma-separated value (CSV) formatted file to ```/srv/docker/ecdc/data```. Make sure it is readable by all. Only files with the ending ```.csv``` will be read in, all other files will be ignored.
If this is not an ECDC/EHDS2 installation, then download your site specific configuration repository:
```shell
sudo mkdir -p /etc/bridgehead/
sudo git clone <REPO_URL_FROM_EMAIL> /etc/bridgehead/
@ -149,12 +199,6 @@ Pay special attention to:
- OPERATOR_LAST_NAME
- OPERATOR_EMAIL
Clone the bridgehead repository:
```shell
sudo mkdir -p /srv/docker/
sudo git clone https://github.com/samply/bridgehead.git /srv/docker/bridgehead
```
Then, run the installation script:
```shell
@ -173,8 +217,38 @@ sudo ./bridgehead enroll <PROJECT>
... and follow the instructions on the screen. Please send your default Collection ID and the display name of your site together with the certificate request when you enroll. You should then be prompted to do the next step:
Note: if you are doing an ECDC/EHDS2 installation, you will need to perform the Beam certificate signing yourself. Do not send an email to either of the email addreesses suggested by the bridgehead enroll procedure. Instead, log on to the VM where Beam is running and perform the following (you will need root permissions):
```shell
cd /srv/docker/beam-broker
sudo mkdir -p csr
sudo vi csr/ecdc-bridgehead-<national node name>.csr # Copy and paste the certificate printed during the enroll
sudo pki-scripts/managepki sign --csr-file csr/ecdc-bridgehead-<national node name>.csr --common-name=ecdc-bridgehead-<national node name>.broker.bbmri.samply.de
```
You can check that the Bridgehead has connected to Beam with the following command:
```shell
pki-scripts/managepki list
```
### Starting and stopping your Bridgehead
For an ECDC/EHDS2 installation, this is done with the help of specialized scripts:
To start:
```shell
sudo /srv/docker/bridgehead/run.sh
```
To stop (you generally won't need to do this):
```shell
sudo /srv/docker/bridgehead/stop.sh
```
For regular installations, read on.
If you followed the above steps, your Bridgehead should already be configured to autostart (via systemd). If you would like to start/stop manually:
To start, run
@ -320,6 +394,19 @@ There will be a delay before the effects of Directory sync become visible. First
The data accessed by the federated search is held in the Bridgehead in a FHIR store (we use Blaze).
For an ECDC/EHDS2 installation, you need to provide your data as a table in a CSV (comma-separated value) files and place it in the directory /srv/docker/ecdc/data. You can provide as many data files as you like, and you can add new files incrementally over time.
In order for this new data to be loaded, you will need to execute the ```run.sh``` script with the appropriate arguments:
- To read just the most recently added data files: ```/srv/docker/bridgehead run.sh --upload```.
- To read in all data from scratch: ```/srv/docker/bridgehead run.sh --upload-all```.
These two variants give you the choice between uploading data in an incremental way that preserves the date used for statistics or as a single upload that date stamps everything with the current date.
The Bridgehead can be started without data, but obviously, any searches run from the Explorer will return zero results for your site if you do that. Note that an empty data directory will automatically be inserted on the first start of the Bridgehead if you don't set one up yourself.
For non-ECDC setups, read on.
You can load data into this store by using its FHIR API:
```

View File

@ -1,10 +1,13 @@
version: "3.7"
# This includes only the shared persistence for BBMRI-ERIC and GBN. Federation components are included as modules, see vars.
# This includes only the shared persistence for BBMRI-ERIC and GBN and EHDS2. Federation components are included as modules, see vars.
services:
blaze:
image: docker.verbis.dkfz.de/cache/samply/blaze:latest
#image: docker.verbis.dkfz.de/cache/samply/blaze:latest
# Blaze versions 0.26 and 0.27 do not return anything when you run a
# CQL query, so I am pinning the version at 0.25.
image: samply/blaze:0.25
container_name: bridgehead-bbmri-blaze
environment:
BASE_URL: "http://bridgehead-bbmri-blaze:8080"
@ -20,6 +23,8 @@ services:
- "traefik.http.services.blaze_ccp.loadbalancer.server.port=8080"
- "traefik.http.routers.blaze_ccp.middlewares=ccp_b_strip,auth"
- "traefik.http.routers.blaze_ccp.tls=true"
ports:
- "8081:8080"
volumes:
blaze-data:

81
bbmri/modules/bbmri.conf Normal file
View File

@ -0,0 +1,81 @@
### DO NOT EDIT THIS FILE DIRECTLY.
###
### This file is collaboratively managed by yourself and the CCP-IT team at DKFZ.
### The Bridgehead will pull it from git every night and restart if required.
### To make any changes (or review changes by CCP-IT), please login here:
### [URL_TO_SITE_SPECIFIC_GIT_REPO]
###
### DO NOT EDIT THIS FILE DIRECTLY.
### A note on Secrets:
###
### Variable with a value of <VAULT> will be fetched from a central component
### upon each bridgehead startup.
### Using the proven Vaultwarden password manager puts you in full control of
### who can read the passwords. In particular, as long as you don't declare a
### secret as shared ("SITE+DKFZ"), DKFZ cannot read these strings.
### We recommend putting credentials such as local passwords into the password
### store, not the git repo. Please keep your master password safe (vault.conf).
### Common Configuration of all Components
## This is a descriptive human readable name of your site (e.g. Belgium)
SITE_NAME=<National node>
## This is the id for your site used in machine to machine communication (should be
## lower-case, e.g. belgium)
SITE_ID=<National node>
## This server's hostname, for access from other computers within your institution
## (e.g. mybridgehead.intern.myinstitution.org)
## Optional. If left empty, this is auto-generated via the `hostname` command.
HOST=
## Proxy Configuration
# leave empty if not applicable
# eg.: http://my-proxy-host:my-proxy-port
HTTP_PROXY_URL=
HTTP_PROXY_USERNAME=
HTTP_PROXY_PASSWORD=
HTTPS_PROXY_URL=$HTTP_PROXY_URL
HTTPS_PROXY_USERNAME=$HTTP_PROXY_USERNAME
HTTPS_PROXY_PASSWORD=$HTTP_PROXY_PASSWORD
## Maintenance Configuration
# By default, the bridgehead regularly performs certain housekeeping tasks such as pruning of old docker images to not run out of disk space.
# Set the following to false to opt-out. (Default: true)
#AUTO_HOUSEKEEPING=
### Connector Configuration
## The operator of the specific site.
OPERATOR_FIRST_NAME=
OPERATOR_LAST_NAME=
OPERATOR_EMAIL=
OPERATOR_PHONE=
## SMTP Server
# ex.: mailhost.intern.klinik.de
MAIL_HOST=
MAIL_PORT=
# ex.: no-reply@bridgehead.intern.klinik.de
MAIL_FROM_ADDRESS=
MAIL_FROM_NAME=
### Monitoring
# The apikey used for reporting to the central DKFZ monitoring. Leave empty to opt out.
MONITOR_APIKEY=
### Biobanking (BBMRI) specifics
## We consider BBMRI as BBMRI-ERIC (European) and German Biobank Node (Germany).
## Obviously, all German biobanks are by definition also European. Thus,
## any Bridgehead will by default connect to the BBMRI-ERIC services but not
## the national ones. We aim to proceed similarly for other BBMRI-ERIC National Nodes.
##
## The default values are correct for biobanks outside Germany.
## For a biobank inside Germany, set ENABLE_GBN=true.
# Connect to the European services, e.g. BBMRI-ERIC Sample Locator (Default: true)
ENABLE_ERIC=false
# Connect to the German services, e.g. Biobank Node Sample Locator (Default: false)
# Set this to true in German biobanks!
ENABLE_GBN=false
# Connect to the ECDC services, e.g. ECDC Sample Locator (Default: false)
# Set this to true in ECDC national nodes!
ENABLE_EHDS2=true

View File

@ -0,0 +1,82 @@
version: "3.7"
services:
focus-ehds2:
#image: docker.verbis.dkfz.de/cache/samply/focus:${FOCUS_TAG}
image: samply/focus
container_name: bridgehead-focus-ehds2
environment:
API_KEY: ${EHDS2_FOCUS_BEAM_SECRET_SHORT}
BEAM_APP_ID_LONG: focus.${EHDS2_PROXY_ID}
PROXY_ID: ${EHDS2_PROXY_ID}
BLAZE_URL: "http://blaze:8080/fhir/"
BEAM_PROXY_URL: http://beam-proxy-ehds2:8081
RETRY_COUNT: ${FOCUS_RETRY_COUNT}
OBFUSCATE: "no"
depends_on:
- "beam-proxy-ehds2"
- "blaze"
beam-proxy-ehds2:
image: docker.verbis.dkfz.de/cache/samply/beam-proxy:develop
container_name: bridgehead-beam-proxy-ehds2
environment:
BROKER_URL: ${EHDS2_BROKER_URL}
PROXY_ID: ${EHDS2_PROXY_ID}
APP_focus_KEY: ${EHDS2_FOCUS_BEAM_SECRET_SHORT}
PRIVKEY_FILE: /run/secrets/proxy.pem
ALL_PROXY: http://forward_proxy:3128
TLS_CA_CERTIFICATES_DIR: /conf/trusted-ca-certs
ROOTCERT_FILE: /conf/root.crt.pem
secrets:
- proxy.pem
depends_on:
- "forward_proxy"
volumes:
- /etc/bridgehead/trusted-ca-certs:/conf/trusted-ca-certs:ro
- /srv/docker/bridgehead/bbmri/modules/${EHDS2_ROOT_CERT}.root.crt.pem:/conf/root.crt.pem:ro
# Convert ECDC CSV file into FHIR and push to Blaze
transfair:
container_name: transfair
image: samply/transfair
environment:
FHIR_INPUT_URL: "http://source_blaze:8080/fhir"
FHIR_OUTPUT_URL: "http://bridgehead-bbmri-blaze:8080/fhir"
PROFILE: "amr2fhir"
#WRITE_BUNDLES_TO_FILE: "true"
AMR_FILE_PATH: "/app/data"
restart: on-failure
# The start up logic for TransFAIR is kind of complicated for the ECDC/EHDS2
# pilot. This is because we only want to run it if 1. there are source data
# files to be transformed and 2. if there is no lock file. We also need to
# wait for Blaze to start, TransFAIR does not check for this. And finally,
# once TransFAIR has finished loading data, a lock file is created, to stop
# a time-consuming repeat run.
command: bash -c " \
echo listing /app/data && \
ls -la /app/data && \
ls /app/data/*.[cC][sS][vV] 1> /dev/null 2>&1 && \
[ ! -f /app/data/lock ] && \
( \
echo 'Wait for Blaze to finish initializing' ; \
sleep 360 ; \
echo 'Remove old output files' ; \
rm -rf /app/test/* ; \
cd /app ; \
echo 'Run TransFAIR' ; \
java -jar transFAIR.jar ; \
echo 'Touching lock file' ; \
touch /app/data/lock \
) & tail -f /dev/null"
# If you put .csv files into ./../ecdc/data, TransFAIR will try to process them.
volumes:
- ../../ecdc/test:/app/test/
- ../../ecdc/data:/app/data/
# Report on the data pushed to Blaze by TransFAIR
test-data-loader:
container_name: test-data-loader
image: samply/test-data-loader
command: sh -c "sleep 420 && echo Listing all resources in FHIR store && blazectl --server http://bridgehead-bbmri-blaze:8080/fhir count-resources && tail -f /dev/null"

View File

@ -0,0 +1,28 @@
#!/bin/bash
if [ "${ENABLE_EHDS2}" == "true" ]; then
log INFO "EHDS2 setup detected -- will start services for German Biobank Node."
OVERRIDE+=" -f ./$PROJECT/modules/ehds2-compose.yml"
# The environment needs to be defined in /etc/bridgehead
case "$ENVIRONMENT" in
"production")
export EHDS2_BROKER_ID=ecdc-vm-ehds-test1.swedencentral.cloudapp.azure.com
export EHDS2_ROOT_CERT=ehds2
;;
"test")
export EHDS2_BROKER_ID=ecdc-vm-ehds-test1.swedencentral.cloudapp.azure.com
export EHDS2_ROOT_CERT=ehds2.test
;;
*)
report_error 6 "Environment \"$ENVIRONMENT\" is unknown. Assuming production. FIX THIS!"
export EHDS2_BROKER_ID=ecdc-vm-ehds-test1.swedencentral.cloudapp.azure.com
export EHDS2_ROOT_CERT=ehds2
;;
esac
EHDS2_BROKER_URL=https://${EHDS2_BROKER_ID}
EHDS2_PROXY_ID=${SITE_ID}.${EHDS2_BROKER_ID}
EHDS2_FOCUS_BEAM_SECRET_SHORT="$(cat /proc/sys/kernel/random/uuid | sed 's/[-]//g' | head -c 20)"
EHDS2_SUPPORT_EMAIL=feedback@germanbiobanknode.de
fi

View File

@ -0,0 +1,22 @@
# EHDS2/ECDC certificate
-----BEGIN CERTIFICATE-----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-----END CERTIFICATE-----

View File

@ -0,0 +1,22 @@
# EHDS2/ECDC certificate
-----BEGIN CERTIFICATE-----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-----END CERTIFICATE-----

View File

@ -4,7 +4,10 @@
# Makes only sense for German Biobanks
: ${ENABLE_GBN:=false}
FOCUS_RETRY_COUNT=32
# Makes only sense for EHDS2 project
: ${ENABLE_EHDS2:=false}
FOCUS_RETRY_COUNT=128
PRIVATEKEYFILENAME=/etc/bridgehead/pki/${SITE_ID}.priv.pem
for module in $PROJECT/modules/*.sh
@ -14,12 +17,16 @@ do
done
SUPPORT_EMAIL=$ERIC_SUPPORT_EMAIL
BROKER_URL_FOR_PREREQ="${ERIC_BROKER_URL:-$GBN_BROKER_URL}"
BROKER_URL_FOR_PREREQ="https://ecdc-vm-ehds-test1.swedencentral.cloudapp.azure.com"
if [ -n "$GBN_SUPPORT_EMAIL" ]; then
SUPPORT_EMAIL=$GBN_SUPPORT_EMAIL
fi
if [ -n "$EHDS2_SUPPORT_EMAIL" ]; then
SUPPORT_EMAIL=$EHDS2_SUPPORT_EMAIL
fi
function do_enroll {
COUNT=0
if [ "$ENABLE_ERIC" == "true" ]; then
@ -30,6 +37,10 @@ function do_enroll {
do_enroll_inner $GBN_PROXY_ID $GBN_SUPPORT_EMAIL
COUNT=$((COUNT+1))
fi
if [ "$ENABLE_EHDS2" == "true" ]; then
do_enroll_inner $EHDS2_PROXY_ID $EHDS2_SUPPORT_EMAIL
COUNT=$((COUNT+1))
fi
if [ $COUNT -ge 2 ]; then
echo
echo "You just received $COUNT certificate signing requests (CSR). Please send $COUNT e-mails, with 1 CSR each, to the respective e-mail address."

View File

@ -88,7 +88,20 @@ case "$ACTION" in
start)
loadVars
hc_send log "Bridgehead $PROJECT startup: Checking requirements ..."
chown -R bridgehead ${BASE}
checkRequirements
# Note: changes to "bridgehead" script will only take effect after next start.
su bridgehead -c "git pull"
chown -R bridgehead ${BASE}
# Local versions of focus and transfair are needed by EHDS2
clone_focus_if_nonexistent ${BASE}/..
build_focus ${BASE}/..
clone_transfair_if_nonexistent ${BASE}/..
build_transfair ${BASE}/..
# Location for input data and results for EHDS2
mkdir -p ${BASE}/../ecdc/test
mkdir -p ${BASE}/../ecdc/data
chown -R bridgehead ${BASE}/../ecdc
hc_send log "Bridgehead $PROJECT startup: Requirements checked out. Now starting bridgehead ..."
exec $COMPOSE -p $PROJECT -f ./minimal/docker-compose.yml -f ./$PROJECT/docker-compose.yml $OVERRIDE up --abort-on-container-exit
;;

14
ecdc.service Normal file
View File

@ -0,0 +1,14 @@
[Unit]
Description=Start ECDC Bridgehead
[Service]
Type=simple
ExecStart=/srv/docker/bridgehead/restart_service.sh
ExecStop=/srv/docker/bridgehead/shutdown_service.sh
Restart=always
RestartSec=36000
KillMode=mixed
[Install]
WantedBy=default.target

View File

@ -239,3 +239,67 @@ add_basic_auth_user() {
log DEBUG "Saving clear text credentials in $FILE. If wanted, delete them manually."
sed -i "/^$NAME/ s|$|\n# User: $USER\n# Password: $PASSWORD|" $FILE
}
function clone_repo_if_nonexistent() {
local repo_url="$1" # First argument: Repository URL
local target_dir="$2" # Second argument: Target directory
local branch_name="$3" # Third argument: Branch name
echo Repo directory: $target_dir
# Check if the target directory exists
if [ ! -d "$target_dir" ]; then
echo "Directory '$target_dir' does not exist. Cloning the repository..."
# Clone the repository
git clone "$repo_url" "$target_dir"
fi
# Change to the cloned directory
cd "$target_dir"
# Checkout the specified branch
chown -R bridgehead .
su bridgehead -c "git checkout $branch_name"
cd -
}
function clone_transfair_if_nonexistent() {
local base_dir="$1"
clone_repo_if_nonexistent https://github.com/samply/transFAIR.git $base_dir/transfair ehds2
}
function clone_focus_if_nonexistent() {
local base_dir="$1"
clone_repo_if_nonexistent https://github.com/samply/focus.git $base_dir/focus ehds2
}
function build_transfair() {
local base_dir="$1"
# We only take the touble to build transfair if:
#
# 1. There is data available (any CSV files) and
# 2. There is no data lock file (which means that no ETL has yet been run).
if ls ../ecdc/data/*.[cC][sS][vV] 1> /dev/null 2>&1 && [ ! -f ../ecdc/data/lock ]; then
cd $base_dir/transfair
su bridgehead -c "git pull"
docker build --progress=plain -t samply/transfair --no-cache .
chown -R bridgehead .
cd -
fi
}
function build_focus() {
local base_dir="$1"
cd $base_dir/focus
su bridgehead -c "git pull"
docker build --progress=plain -f DockerfileWithBuild -t samply/focus --no-cache .
chown -R bridgehead .
cd -
}

View File

@ -84,7 +84,7 @@ else
SYNCTEXT="For example, consider entering a correct NTP server (e.g. your institution's Active Directory Domain Controller in /etc/systemd/timesyncd.conf (option NTP=) and restart systemd-timesyncd."
if [ $SKEW -ge 300 ]; then
report_error 5 "Your clock is not synchronized (${SKEW}s off). This will cause Samply.Beam's certificate will fail. Please setup time synchronization. $SYNCTEXT"
exit 1
log WARN "Server Time Error"
elif [ $SKEW -ge 60 ]; then
log WARN "Your clock is more than a minute off (${SKEW}s). Consider syncing to a time server. $SYNCTEXT"
fi

View File

@ -45,7 +45,7 @@ services:
landing:
container_name: bridgehead-landingpage
image: docker.verbis.dkfz.de/cache/samply/bridgehead-landingpage:main
image: docker.verbis.dkfz.de/cache/samply/bridgehead-landingpage:master
labels:
- "traefik.enable=true"
- "traefik.http.routers.landing.rule=PathPrefix(`/`)"

30
restart_service.sh Executable file
View File

@ -0,0 +1,30 @@
#!/bin/bash
# Start a running Bridgehead. If there is already a Bridgehead running,
# stop it first.
# This is intended to be used by systemctl.
cd /srv/docker/bridgehead
echo "git status before stop"
git status
echo "Stopping running Bridgehead, if present"
./bridgehead stop bbmri
# If "flush_blaze" is present, delete the Blaze volume before starting
# the Bridgehead again. This allows a user to upload all data, if
# requested.
if [ -f "/srv/docker/ecdc/data/flush_blaze" ]; then
docker volume rm bbmri_blaze-data
rm -f /srv/docker/ecdc/data/flush_blaze
fi
echo "git status before start"
git status | systemd-cat -p info
echo "Start the Bridgehead anew"
./bridgehead start bbmri
echo "Bridgehead has unexpectedly terminated"

83
run.sh Executable file
View File

@ -0,0 +1,83 @@
#!/usr/bin/env bash
# Start a Bridgehead from the command line. Upload data if requested.
# Behind the scenes we use systemctl to do the work.
# Function to print usage
print_usage() {
echo "Start a Bridghead, optionally upload data"
echo "Usage: $0 [--upload | --upload-all | --help | -h]"
echo "Options:"
echo " --upload Run Bridgehead and upload just the new CSV data files."
echo " --upload-all Run Bridgehead and upload all CSV data files."
echo " --help, -h Display this help message."
echo " No options Run Bridgehead only."
}
# Initialize variables
UPLOAD=false
UPLOAD_ALL=false
# Parse arguments
while [[ "$#" -gt 0 ]]; do
case $1 in
--upload)
UPLOAD=true
;;
--upload-all)
UPLOAD_ALL=true
;;
--help|-h)
print_usage
exit 0
;;
*)
echo "Error: Unknown argument '$1'"
print_usage
exit 1
;;
esac
shift
done
# Check for conflicting options
if [ "$UPLOAD" = true ] && [ "$UPLOAD_ALL" = true ]; then
echo "Error: you must specify either --upload or --upload-all, specifying both is not permitted."
print_usage
exit 1
fi
# Disable/stop standard Bridgehead systemctl services, if present
sudo systemctl disable bridgehead@bbmri.service >& /dev/null
sudo systemctl disable system-bridgehead.slice >& /dev/null
sudo systemctl disable bridgehead-update@bbmri.timer >& /dev/null
sudo systemctl stop bridgehead@bbmri.service >& /dev/null
sudo systemctl stop system-bridgehead.slice >& /dev/null
sudo systemctl stop bridgehead-update@bbmri.timer >& /dev/null
# Set up systemctl for EHDS2/ECDC if necessary
cp /srv/docker/bridgehead/ecdc.service /etc/systemd/system
systemctl daemon-reload
systemctl enable ecdc.service
# Use systemctl to stop the Bridgehead if it is running
sudo systemctl stop ecdc.service
# Use files to tell the Bridgehead what to do with any data present
if [ "$UPLOAD" = true ] || [ "$UPLOAD_ALL" = true ]; then
if [ -f /srv/docker/ecdc/data/lock ]; then
rm /srv/docker/ecdc/data/lock
fi
fi
if [ "$UPLOAD_ALL" = true ]; then
echo "All CSV files in /srv/docker/ecdc/data will be uploaded"
touch /srv/docker/ecdc/data/flush_blaze
fi
# Start up the Bridgehead
sudo systemctl start ecdc.service
# Show status of Bridgehead service
sleep 10
systemctl status ecdc.service

13
shutdown_service.sh Executable file
View File

@ -0,0 +1,13 @@
#!/bin/bash
# Shut down a running Bridgehead.
# This is intended to be used by systemctl.
cd /srv/docker/bridgehead
echo "git status before stop"
git status
echo "Stopping running Bridgehead, if present"
./bridgehead stop bbmri

43
stop.sh Executable file
View File

@ -0,0 +1,43 @@
#!/usr/bin/env bash
# Shut down a running Bridgehead.
# Behind the scenes we use systemctl to do the work.
# Function to print usage
print_usage() {
echo "Stop the running Bridgehead"
echo "Usage: $0 [--help | -h]"
echo "Options:"
echo " --help, -h Display this help message."
echo " No options Stop Bridgehead only."
}
# Parse arguments
while [[ "$#" -gt 0 ]]; do
case $1 in
--help|-h)
print_usage
exit 0
;;
*)
echo "Error: Unknown argument '$1'"
print_usage
exit 1
;;
esac
shift
done
# Set up systemctl for EHDS2/ECDC if necessary
cp /srv/docker/bridgehead/ecdc.service /etc/systemd/system
systemctl daemon-reload
systemctl enable ecdc.service
# Use systemctl to stop the Bridgehead if it is running
sudo systemctl stop ecdc.service
# Show status of Bridgehead service
sleep 20
systemctl status ecdc.service
docker ps