quick and comprehensive setup

README.md

@@ -1 +1,2 @@
-# immich-podman-quadlets
+# immich-podman-quadlets
+If you're using podman, this is the easy way to setup immich.

docs/install/hwa.md

@@ -0,0 +1,78 @@
+---
+sidebar_position: 100
+---
+
+# HardWare Acceleration setup for immich with podman
+
+If you are using NVIDIA on a redhat OS such as fedora or almalinux, reboots sometimes cause the nvidia-uvm and nvidia-uvm-tools to just... not load. Use the following script and systemd file to correct this behaviour:
+
+You can ultimately place this wherever you want as long as it's referenced appropriately in the systemd file:
+nv_init.sh
+```bash
+#!/bin/bash
+
+echo "modprobe nvidia"
+/sbin/modprobe nvidia
+
+if [ "$?" -eq 0 ]; then
+  # Count the number of NVIDIA controllers found.
+  NVDEVS=`lspci | grep -i NVIDIA`
+  N3D=`echo "$NVDEVS" | grep "3D controller" | wc -l`
+  NVGA=`echo "$NVDEVS" | grep "VGA compatible controller" | wc -l`
+
+  N=`expr $N3D + $NVGA - 1`
+  echo "mknode nvidia"
+  for i in `seq 0 $N`; do
+    if [ ! -e /dev/nvidia$i ]; then
+      mknod -m 666 /dev/nvidia$i c 195 $i
+    fi
+  done
+
+  echo "mknode nvidiactl"
+  if [ ! -e /dev/nvidiactl ]; then
+    mknod -m 666 /dev/nvidiactl c 195 255
+  fi
+else
+  echo "exit NVIDIA controller"
+  exit 1
+fi
+
+echo "modprobe nvidia-uvm"
+/sbin/modprobe nvidia-uvm
+
+if [ "$?" -eq 0 ]; then
+  # Find out the major device number used by the nvidia-uvm driver
+  D=`grep nvidia-uvm /proc/devices | awk '{print $1}'`
+
+  echo "mknode nvidia-uvm"
+  if [ ! -e /dev/nvidia-uvm ]; then
+    mknod -m 666 /dev/nvidia-uvm c $D 0
+  fi
+  echo "mknode nvidia-uvm-tools"
+  if [ ! -e /dev/nvidia-uvm-tools ]; then
+    mknod -m 666 /dev/nvidia-uvm-tools c $D 0
+  fi
+else
+  echo "exit nvidia-uvm"
+  exit 1
+fi
+```
+
+/etc/systemd/system/init_nvidia.service
+```bash
+[Unit]
+Description=NVIDIA Init for CUDA
+After=network.target
+ConditionPathExists=|!/dev/nvidia
+ConditionPathExists=|!/dev/nvidiactl
+ConditionPathExists=|!/dev/nvidia-uvm
+ConditionPathExists=|!/dev/nvidia-uvm-tools
+
+[Service]
+Type=oneshot
+ExecStart=/bin/bash ${your_path_to_script}/nv_init.sh
+RemainAfterExit=yes
+
+[Install]
+WantedBy=multi-user.target
+```

docs/install/podman-quadlet.md

@@ -0,0 +1,136 @@
+---
+sidebar_position: 90
+---
+
+# Podman deploy with quadlets
+
+You can deploy Immich on Podman using quadlets.
+
+Here are some sample rootless quadlet container files that can be placed in /etc/containers/systemd/users/${ID} where ID is the uid of whatever your rootless user is.
+
+Please note you'll need :z or :Z for selinux enabled hosts.
+
+If you are doing hardware acceleration with NVIDIA make sure you follow the immich documentation including the appropriate NVIDIA Container Device Interface (https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/cdi-support.html) and adding the device (AddDevice=nvidia.com/gpu=0).
+
+immich-database.container
+```bash
+[Unit]
+Description=Immich Database
+Requires=immich-redis.service
+
+[Container]
+AutoUpdate=registry
+EnvironmentFile=${location_of_env_file}
+Image=registry.hub.docker.com/tensorchord/pgvecto-rs:pg16-v0.2.1
+Label=registry
+Network=slirp4netns:port_handler=slirp4netns
+PublishPort=5432:5432
+Volume=${host_database_directory}:/var/lib/postgresql/data:z
+Volume=/etc/localtime:/etc/localtime:ro
+
+[Service]
+Restart=always
+
+[Install]
+WantedBy=multi-user.target default.target
+```
+
+immich-microservices.container
+```bash
+[Unit]
+Description=Immich Microservices
+Requires=immich-redis.service immich-database.service
+
+[Container]
+#AddDevice=/dev/dri:/dev/dri #Needed for HWA
+#AddDevice=nvidia.com/gpu=0 #Needed for nvidia HWA, after setting up container tools
+AutoUpdate=registry
+EnvironmentFile=${location_of_env_file}
+Image=ghcr.io/immich-app/immich-server:release
+Label=registry
+Network=slirp4netns:port_handler=slirp4netns
+PublishPort=3002:3002
+Volume=${host_upload_directory}:/usr/src/app/upload:z
+Volume=/etc/localtime:/etc/localtime:ro
+Exec=start.sh microservices
+#Unmask=/dev/dri:/dev/dri #May be needed if doing HWA
+#UserNS=keep-id #May be needed if doing HWA
+
+[Service]
+Restart=always
+
+[Install]
+WantedBy=multi-user.target default.target
+```
+
+immich-ml.container
+```bash
+
+[Unit]
+Description=Immich Machine Learning
+Requires=immich-redis.service immich-database.service
+
+[Container]
+#AddDevice=/dev/dri:/dev/dri #Needed for HWA
+#AddDevice=nvidia.com/gpu=0 #Needed for nvidia HWA, after setting up container tools
+AutoUpdate=registry
+EnvironmentFile=${location_of_env_file}
+Image=ghcr.io/immich-app/immich-machine-learning:release
+Label=registry
+Network=slirp4netns:port_handler=slirp4netns
+PublishPort=3003:3003
+Volume=${cache_directory}:/cache:z
+Volume=/etc/localtime:/etc/localtime:ro
+#Unmask=/dev/dri:/dev/dri #May be needed for HWA
+
+[Service]
+Restart=always
+
+[Install]
+WantedBy=multi-user.target default.target
+```
+
+immich-redis.container
+```bash
+[Unit]
+Description=Immich Redis
+
+[Container]
+AutoUpdate=registry
+Image=registry.hub.docker.com/library/redis:6.2-alpine@sha256:51d6c56749a4243096327e3fb964a48ed92254357108449cb6e23999c37773c5
+Label=registry
+Network=slirp4netns:port_handler=slirp4netns
+PublishPort=6379:6379
+Timezone=America/Montreal
+
+[Service]
+Restart=always
+
+[Install]
+WantedBy=multi-user.target default.target
+```
+
+immich-server.container
+```bash
+[Unit]
+Description=Immich Server
+Requires=immich-redis.service immich-database.service
+
+[Container]
+AutoUpdate=registry
+EnvironmentFile=${location_of_env_file}
+Image=ghcr.io/immich-app/immich-server:release
+Label=registry
+Network=slirp4netns:port_handler=slirp4netns
+Exec=start.sh immich
+PublishPort=3000:3000
+PublishPort=3001:3001
+Volume=${host_upload_directory}:/usr/src/app/upload
+Volume=/etc/localtime:/etc/localtime:ro
+
+[Service]
+Restart=always
+
+[Install]
+WantedBy=multi-user.target default.target
+```

podman-quadlet.md

@@ -0,0 +1 @@
+docs/install/podman-quadlet.md