简体   繁体   中英

flink on docker disk full

I created a very small cluster in docker (1 job manager 3 task manager). I have 2 jobs (1 streaming and 1 batch which is restarted when it's finished).

I have 58G free space (approx. 30G used). When I start the aggregator, the job manager started to eat the disk. After some hour my disk is full. But when I check the FS ( du -su /*/ ), it only contains about 30G of files.

I do not understand why it's happened, and where is the data. I try to check inside the jobmanager container but no success.

Any idea?

I had the same problem, it appears it was the .out file in the /usr/local/flink-{1-0.3}/log folder which was over 6GB. I had to delete the file and redeploy the jobs.

Are you using Docker Meta for Mac? I had a "full disk space" issue and I couldn't build my projects anymore. The steps that worked for me:

1) df -h --> check that the host has enough available space

If that's the case, do the following:

1) docker exec -it your-container bash --> connect to your container

2) df -h /var/www --> check disk space

If there's no more available space, exit the container and do the following:

1) docker rm $(docker ps -aq) --> stop and delete all containers

2) docker rmi --> delete all images

3) ls -lh ~/Library/Containers/com.docker.docker/Data/com.docker.driver.amd64-linux/Docker.qcow2 --> check this file size and delete it

4) quit Docker and reopen it

5) quit the terminal and reopen it

6) Build again

The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM