Not an airport, no need to announce your departure
Not an airport, no need to announce your departure
Incoming wall of text
Here is my install script to set up Ubuntu since it has a bit of extra steps for privileged ports https://gitlab.meme.beer/-/snippets/1
Docker compose example, note that my config has a shared network with containers in another compose called nginx
to keep traffic inside docker.
name: "gitlab"
services:
gitlab:
image: 'gitlab/gitlab-ce:latest'
#command: update-permissions
restart: always
hostname: 'gitlab.example.com'
environment:
GITLAB_OMNIBUS_CONFIG: |
external_url 'https://gitlab.example.com'
pages_external_url 'https://pages.example.com'
pages_nginx['enable'] = true
pages_nginx['listen_port'] = 6000
pages_nginx['listen_https'] = false
pages_nginx['redirect_http_to_https'] = false
#puma['per_worker_max_memory_mb'] = 2048 # 2GB
gitlab_rails['gitlab_email_from'] = 'gitlab@mailer.example.com'
gitlab_rails['gitlab_email_display_name'] = 'GitLab'
gitlab_rails['smtp_enable'] = true
gitlab_rails['smtp_address'] = "smtp.sendgrid.net"
gitlab_rails['smtp_port'] = 587
gitlab_rails['smtp_user_name'] = 'apikey'
gitlab_rails['smtp_password'] = '$SENDGRID_API_KEY_HERE'
gitlab_rails['smtp_domain'] = "smtp.sendgrid.net"
gitlab_rails['smtp_authentication'] = "login"
gitlab_rails['smtp_enable_starttls_auto'] = true
gitlab_rails['smtp_tls'] = false
gitlab_rails['gitlab_default_theme'] = 2
gitlab_rails['gitlab_shell_ssh_port'] = 2224
gitlab_rails['gitlab_default_projects_features_container_registry'] = true
gitlab_rails['registry_enabled'] = true
gitlab_rails['registry_api_url'] = 'https://registry.example.com'
gitlab_rails['registry_issuer'] = 'gitlab-issuer'
registry['log_level'] = 'info'
registry_external_url 'https://registry.example.com'
registry_nginx['enable'] = true
registry_nginx['listen_port'] = 5050
registry_nginx['listen_https'] = false
registry_nginx['redirect_http_to_https'] = false
gitlab_shell['log_level'] = 'INFO'
letsencrypt['enable'] = false
nginx['error_log_level'] = 'info'
nginx['listen_https'] = false
#nginx['proxy_protocol'] = true
#nginx['trusted_proxies'] = ["10.0.0.0/8", "172.16.0.0/12", "192.168.0.0/16"]
# Workhorse
gitlab_workhorse['enable'] = true
gitlab_workhorse['ha'] = false
gitlab_workhorse['listen_network'] = "tcp"
gitlab_workhorse['listen_addr'] = "127.0.0.1:8181"
gitlab_workhorse['log_directory'] = "/var/log/gitlab/gitlab-workhorse"
# Errors
# for sentry error logging the GitLab service
#gitlab_rails['sentry_enabled'] = true
#gitlab_rails['sentry_dsn'] = ''
#gitlab_rails['sentry_clientside_dsn'] = ''
#gitlab_rails['sentry_environment'] = 'production'
# Add any other gitlab.rb configuration here, each on its own line
networks:
- nginx
ports:
# gitlab loves https on 443
#- '80:80'
#- '443:443'
- '2224:22'
volumes:
- ./config:/etc/gitlab
- ./logs:/var/log/gitlab
- ./data:/var/opt/gitlab
shm_size: '256m'
#deploy:
# resources:
# limits:
# cpus: '6'
# memory: 12G
# reservations:
# cpus: '4'
# memory: 6G
# disable healthcheck for restoring backup
#healthcheck:
# disable: true
networks:
nginx:
external: true
name: nginx
The VM is a 6 thread 16gb
OS is currently Ubuntu 22.04.5 LTS (cloud image which is lightweight) just running a very simple docker engine install using the script (plus a few other options since I script the install)
The load averages as of this current moment are 0.12, 0.15, 0.10
so not even a full thread is being used.
I let the container run unmetered on the CPU and memory.
I can provide both the compose and my install script (which is on the GitLab instance) if you are curious.
I run GitLab with docker compose and watchtower, all the updates are automated and have never caused any issues for me.
That being said my setup uses about 7-8gb of ram.
I got a home server with a Nvidia Tesla P4, not the most power or the most vram (8gb), but can be gotten for ~$100usd (it is a headless GPU so no video outputs)
I’m using ollama with dolphin-mistral and recently deepseek coder
My point being I don’t want to update that sort of stuff that can be automated.
Naw I mostly do it for my own personal shit, can’t be fucked to update Plex 3 times a week and so on with other homelab stuff. Everything production is tagged with gitops version managed kubernetes manifests
Edit: should also mention I build quite a bit of the software being deployed
I run latest with watchtower on so much shit
It wasn’t in a markdown format, also I’m on jerboa
The ) at the end of the url is breaking it and sending me to a 404 page
Proxmox has a full webui with almost every feature you could want, except for some more advanced zfs features
GitLab CE self hosted
I’ve been buying Water Panther refurbished drives.
Both Arsenal (had them for a while) and SaveGreen (they just released recently) and have any had a good experience with them for how cheap they are with warranty.
Only filed 1 RMA, and the turnaround was fairly quick.
They are also Seagate drives, and the SaveGreens have different firmware
I almost exclusively use whoogle
deleted by creator
I haven’t heard any such thing unless you are paying for their video CDN which isn’t cheap
🤮 Asslassian
You would change it from orange cloud to gray cloud to disable the proxy in your domain DNS settings.
Specifically for the jellyfin subdomain.
That would be the correct solution IMO.
Be careful about using video over the cloudflare proxy, they will come after you if you are using a TON of bandwidth
DARE to do drugs
DARE to keep threads defed’d