The entrypoint executes some commands on MySQL:

    print_log "initial_setup:info" "Applying changes to MySQL database... (MySQL error will be printed to console as guidance)"
    exec_cmd "mysql --protocol=socket -u $AUTO_DB_SETUP_USER -p$AUTO_DB_SETUP_PASS -h$SIMPLERISK_DB_HOSTNAME -P$SIMPLERISK_DB_PORT <<EOSQL
    CREATE DATABASE `${SIMPLERISK_DB_DATABASE}`;
    USE `${SIMPLERISK_DB_DATABASE}`;
    \. ${SCHEMA_FILE}
    CREATE USER ${SIMPLERISK_DB_USERNAME}@'%' IDENTIFIED BY '${SIMPLERISK_DB_PASSWORD}';
    GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, REFERENCES, INDEX, ALTER ON `${SIMPLERISK_DB_DATABASE}`.* TO ${SIMPLERISK_DB_USERNAME}@'%';
EOSQL" "Was not able to apply settings on database. Check error above. Exiting."

The problem is that, as this is a bash script, the content inside the backticks are being interpreted as commands.

I've installed SimpleRisk with the FIRST_TIME_SETUP parameter, everything runs okay.
But after I kill the container and start it again, this varible in config.php is set the "false" again and when I access the application, it goes back to installation page.
This shouldn't happen and it should let me in the application instead.
Can you add another parameter like SIMPLERISK_INSTALLED and set the value in config.php

NVD Info

First detection

Local scan (get only critical, discard CVEs with "Won't fix" status):

> grype -q simplerisk/simplerisk-minimal | grep Critical | grep -v "won't fix"
linux-libc-dev             5.10.178-3                            deb     CVE-2020-36694       Critical
> grype -q php:8.1-apache | grep Critical | grep -v "won't fix"
linux-libc-dev             5.10.178-3                            deb     CVE-2020-36694    Critical

The simplerisk image is published to a repository on Docker Hub. The simplerisk-minimal image, however, has a repository on Docker Hub but is not published to it. Please publish this image as well.

Environment:

• Ubuntu 20.04 (x86_64)
• Docker version 20.10.17, build 100c701
• Container: simplerisk/simplerisk:latest (d074297a490c)

To reproduce:

• Pull docker image
• Start container
• Stop container
• Start container

Expected result:

Container remains running

Observed behavior:

Container crashes after about 10-15s with the following logs:

 * Starting MySQL database server mysqld
su: warning: cannot change directory to /nonexistent: No such file or directory
   ...done.
 * Starting periodic command scheduler cron
   ...done.
AH00558: apache2: Could not reliably determine the server's fully qualified domain name, using 172.17.0.2. Set the 'ServerName' directive globally to suppress this message
 * Starting MySQL database server mysqld
su: warning: cannot change directory to /nonexistent: No such file or directory
   ...fail!
sed: no input files

Possible issue

It seems as though the MySQL process is dying due to not having a homedir set for the mysql user.

Suggested Solution

Add a line to the dockerfile to create the expected homedir. I'll submit a PR which seems to rectify this issue.

version: '3'

services:
simplerisktest:
image: phsa-simplerisknew:v1.0
hostname: risktest

ports:

- 80:80

- 443:443

 volumes:
   - "./bashrc:/root/.bashrc:rw"
   - "html:/var/www/simplerisk:rw"
   - "html-etc-ssl:/etc/apache2/ssl:rw"
   - "mysql-data:/var/lib/mysql:rw"
   - "mysql-etc:/etc/mysql:rw"
   - "var-log:/var/log:rw"
   - "config:/configurations:rw"
   - "pw:/passwords:rw"
   - /etc/localtime:/etc/localtime:ro

phpmyadmintest:
image: phpmyadmin/phpmyadmin
container_name: phpmyadmintest
environment:
- PMA_ARBITRARY=1
- PMA_ABSOLUTE_URI= https://phsarisktest.phsabc.ehcnet.ca/phpmyadmin/
- PMA_HOST=simplerisktest
restart: always

ports:

- 8080:80

- 443:443

 volumes:
  - /sessions
  - "./config.inc.php:/etc/phpmyadmin/config.inc.php:rw"
  - ./ssl-cert:/ssl-cert
  - ./nginx.conf:/etc/nginx.conf

volumes:
html:
html-etc-ssl:
mysql-data:
mysql-etc:
var-log:
config:
pw:

Kubernetes Ingress expects TLS to terminate at the ingress controller, all communication from that point on is plaintext (port 80). The rewrite rule included in the common/000-default.conf files causes a redirect loop when deployed to a kubernetes cluster with an ingress controller.

Mapping the 6 volumes to local disk (so that the config and database is stored outside of the docker container) breaks the simplerisk container.

Please advise

This image can’t be compatible with the user having the extras folder (for those paid features) under /var/www/simple risk. It will failed to upgrade process after importing the DB backup from old system. Dorian knows more details.

Was building the 18.04 image and received this error

"No releases available for package "pecl.php.net/mcrypt"
install failed"

Mcrypt has been depreciated and was wondering if its a dependent module or if I can use something else.

installation procedure does not work. can you help me ?