updated docu based on new findings regarding yarn output to tmp folder

This commit is contained in:
Ulf Gebhardt 2023-05-02 11:39:37 +02:00
parent 4ef9efca5c
commit 51900af67d
Signed by: ulfgebhardt
GPG Key ID: DA6B843E748679C9

View File

@ -196,6 +196,27 @@ Use it as pattern to do all steps manually in your terminal shell.
Follow the commands in `./install.sh` as installation pattern.
## Define Cronjob To Compensate Yarn Output In `/tmp`
`yarn` creates output in `/tmp` directory. This output is generated whenever `yarn start` is called. THis is especially problematic on staging systems where instable versions are automatically deployed which can lead to a ever resatrting service, hence generating alot of yarn output.
To solve this you can install the following hourly cron using `crontab` as `gradido` user.he `gradido` user.
Run:
```bash
crontab -e
```
This opens the crontab in edit-mode and insert the following entry:
```bash
0 * * * * find /tmp -name "yarn--*" -exec rm -r {} \; > /dev/null
```
For production systems this is not need by default since the yarn output is deleted when `start.sh` is executed. If the service runs stable and does not restart frequently the yarn output to the tmp folder scales with the amount of services running.
## Define Cronjob To start backup script automatically
At least at production stage we need a daily backup of our database. This can be done by adding a cronjob