Note: Before you try to deploy file search office documents preview, make sure other parts of your seafile cluster are already working, e.g upload/download files in a web browser. Make sure memcached is configured as described in "Deploy in a cluster".
In the seafile cluster, only one server should run the background tasks, including:
- indexing files for search
- email notification
- office documents converts service
- LDAP sync
- virus scan
Let's assume you have three nodes in your cluster: A, B, and C.
- Node A is backend node that run background tasks.
- Node B and C are frontend nodes that serving requests from clients.
If you following the steps on settings up a cluster, node B and node C should have already be configed as frontend node. You can copy the configuration of node B as a base for node A. Then do the following steps:
On Ubuntu/Debian:
sudo apt-get install openjdk-8-jre libreoffice poppler-utils python-uno # or python3-uno for ubuntu 16.04+
On CentOS/Red Hat:
sudo yum install java-1.8.0-openjdk
sudo yum install libreoffice libreoffice-headless libreoffice-pyuno
sudo yum install poppler-utils
Edit seafevents.conf and ensure this line does NOT exist:
external_es_server = true
Edit seahub_settings.py and add a line:
OFFICE_CONVERTOR_NODE = True
Edit seafile.conf to enable virus scan according to virus scan document
In your firewall rules for node A, you should open the port 9200 (for search requests). For versions older than 6.1, es_port
was 9500.
On nodes B and C, you need to:
- Edit
seafevents.conf
, add the following lines:
[INDEX FILES]
external_es_server = true
es_host = <ip of node A>
es_port = 9200
Edit seahub_settings.py and add a line:
OFFICE_CONVERTOR_ROOT = 'http://<ip of node A>'
Make sure requests to http:// is also handled by Seahub. For example, you may need to add this Nginx configuration in the background node:
server {
listen 80;
server_name <IP of node A>;
location / {
proxy_pass http://127.0.0.1:8000;
...
}
As a simple test, you can use this command to test if you set it up correctly.
curl -v http://<IP of node A>/office-convert/internal/status/
It should say "400 Bad Request" when you have Nginx config updated.
Type the following commands to start the background node (Note, one additional command seafile-background-tasks.sh
is needed)
./seafile.sh start
./seahub.sh start # or "./seahub.sh start-fastcgi" if you're using fastcgi
./seafile-background-tasks.sh start
To stop the background node, type:
./seafile-background-tasks.sh stop
./seafile.sh stop
./seahub.sh stop
You should also configure Seafile background tasks to start on system bootup. For systemd based OS, you can add /etc/systemd/system/seafile-background-tasks.service
:
[Unit]
Description=Seafile Background Tasks Server
After=network.target seahub.service
[Service]
Type=forking
ExecStart=/opt/seafile/seafile-server-latest/seafile-background-tasks.sh start
ExecStop=/opt/seafile/seafile-server-latest/seafile-background-tasks.sh stop
User=root
Group=root
[Install]
WantedBy=multi-user.target
Then enable this task in systemd:
systemctl enable seafile-background-tasks.service
Here is the summary of configurations at the background node that related to clustering setup.
For seafile.conf:
[cluster]
enabled = true
memcached_options = --SERVER=<IP of memcached node> --POOL-MIN=10 --POOL-MAX=100
For seahub_settings.py:
OFFICE_CONVERTOR_NODE = True
AVATAR_FILE_STORAGE = 'seahub.base.database_storage.DatabaseStorage'
COMPRESS_CACHE_BACKEND = 'django.core.cache.backends.locmem.LocMemCache'
For seafevents.conf:
[INDEX FILES]
enabled = true
interval = 10m
[OFFICE CONVERTER]
enabled = true
workers = 1
## how many pages are allowed to be previewed online. Default is 50 pages
max-pages = 50
## the max size of documents allowed to be previewed online, in MB. Default is 10 MB
## Previewing a large file (for example >30M) online is likely going to freeze the browser.
max-size = 10