Closed ahathaway closed 4 years ago
Could you please provide your playbook? Make sure you cross out any passwords!
For sure, please also be aware that I have only added the other steps before your role as a way to troubleshoot the issue. I thought perhaps because I didn't have enough swap space, so I added that step, and then I thought perhaps I needed to update apt repos, and update installed software first. The issue occured first with just your role being used, the same as it is now. I began on AWS free tier, t2.micro, w/ 30gb SSD, and also db.t2.micro RDS instance. I have since upgraded the ec2 instance to t3.small, and 4gb swap, still just churns until out of memory error. I let it run overnight last night and was still running 8 hours later when I checked it this morning, creeping up on virtual memory, full on hard memory. I'm sure something is erroring out but I'm not sure how to see what it is.
I have "redacted" any sensitive information. Anywhere you see *-redacted was something legitimate.
Here it is:
---
- name: Converge
hosts: hosts-redacted
become: true
become_user: root
tasks:
- name: update apt cache
apt:
update_cache: yes
- name: upgrade all packages
apt:
name: "*"
state: latest
force_apt_get: true
- name: pass vars to role for swapfile setup
import_role:
name: "ansible-swapfile"
vars:
swapfile_size: 4GB
swapfile_swappiness: 60
- name: Pass variables to role for ispconfig
import_role:
name: ansible-role-ispconfig3
vars:
certbot_auto_renew_user: root
certbot_auto_renew_minute: 20
certbot_auto_renew_hour: 5
mail:
admin_email: email-redacted
admin_password: password-redacted
base_domain: domain-redacted
mailing_list_name: Mailman
phpmyadmin:
admin_password: password-redacted
roundcube:
admin_password: password-redacted
ispconfig:
language: en
install_mode: standard
hostname: domain-redacted
mysql_hostname: instace-redacted.cceqss5ahxtt.us-west-2.rds.amazonaws.com
mysql_root_user: mysql1root
mysql_root_password: password-redacted
mysql_database: dbispconfig
mysql_port: 3306
mysql_charset: utf8
http_server: apache
ispconfig_port: 8080
ispconfig_use_ssl: y
ispconfig_admin_password: password-redacted
ssl_cert_country: US
ssl_cert_state: state-redacted
ssl_cert_locality: city-redacted
ssl_cert_organisation: company-redacted
ssl_cert_organisation_unit: "Hosting Services"
ssl_cert_common_name: domain-redacted
mysql_ispconfig_user: ispconfig
mysql_ispconfig_password: password-redacted
join_multiserver_setup: n
mysql_master_hostname: localhost
mysql_master_root_password: password-redacted
mysql_master_database: dbispconfig
configure_mail: y
configure_jailkit: y
configure_ftp: y
configure_dns: y
configure_nginx: y
configure_apache: y
configure_firewall: y
install_ispconfig_web_interface: y
do_backup: yes
reconfigure_permissions_in_master_database: no
reconfigure_services: yes
create_new_ispconfig_ssl_cert: no
reconfigure_crontab: yes
port: 8080
quota:
mounts:
certbot:
admin_email: email-redacted
create_standalone_stop_services: apache
install_from_source: false
staging: no
certs:
- domains:
- domain-redacted
Can you also please post which AMI you are using on your instance? Then i can try to reproduce it!
ubuntu/images/hvm-ssd/ubuntu-bionic-18.04-amd64-server-20190212.1 (ami-005bdb005fb00e791) Canonical, Ubuntu, 18.04 LTS, amd64 bionic image build on 2019-02-12
Just the standard Ubuntu ami that is offered when you make a new instance.
Thank you. I've been looking into this. No clue yet. Still looking. Meanwhile could you please check the file /var/log/ispconfig_install.log
to see if you get any hints?
We did change a few smaller things, that might get you going. Please make sure you update the role before going ahead. Please also try the following playbook with a minimal example first:
---
- name: Converge
hosts: hosts-redacted
become: true
become_user: root
tasks:
- name: update apt cache
apt:
update_cache: yes
- name: upgrade all packages
apt:
name: "*"
state: latest
force_apt_get: true
- name: Pass variables to role for ispconfig
import_role:
name: ansible-role-ispconfig3
vars:
mail_admin_email: email@your-company.com
mail_base_domain: your-company.com
ispconfig_hostname: ispconfig.your-company.com
certbot_admin_email: certificate@your-company.de
certbot_certs:
- domains:
- ispconfig.your-company.de
post_hook:
- ln -s -f /etc/letsencrypt/live/ispconfig.your-company.de/fullchain.pem /usr/local/ispconfig/interface/ssl/ispserver.crt
- ln -s -f /etc/letsencrypt/live/ispconfig.your-company.de/privkey.pem /usr/local/ispconfig/interface/ssl/ispserver.key
If it doesn't work please post your /var/log/ispconfig_install.log
.
Thanks!
Closed because of inactivity.
I've been able to make it to the task "Start ISPConfig Installation" without issue several times. Each time, it eventually errors out, once the machine runs out of both hard and virtual memory. It seems to spawn several /usr/sbin/mysqld processes which then go into sleep state. I am trying to use localhost for mysql_master_hostname and AWS RDS instance for mysql_hostname. I have installed ISPConfig before by hand and it didn't take this long. Wondering if you have any advice on this?