Same for me. Seems like it’s related to the latest 1.11.1-beta13 as I didn’t have such problem for few weeks now I have been testing the native docker for Mac.
MacBook-Pro:com.docker.driver.amd64-linux nzelenkov$ ls -lh log/docker.log
-rw------- 1 nzelenkov staff 43G May 27 13:10 log/docker.log
MacBook-Pro:com.docker.driver.amd64-linux nzelenkov$ ls -lh Docker.qcow2
-rw-r–r-- 1 nzelenkov staff 73G May 27 13:21 Docker.qcow2
+1. Confirming what most other folks said, recent beta update seems to have caused this. Running on OS X 10.11.4. My system monitor indicates that docker is writing at about 43 MB/s pretty much the whole time it’s running.
@aleveille You have mutated this ticket, the original issue was about the size of the image that contained the docker layers being too small and not growing to the size of the physical disk. The issues you’ve directed here are about excessive logging.
Same issue. more than 100G stolen by logs and Docker.qcow2. Relaunch does not help - logs and qcow2 grows 100M per second. Problem appears after last beta update.
I assumed the fan was related to corporate antivirus or my VMware virtual machines and didn’t even think to check the Docker beta I’d just installed. The moment I ran out of disk space and VMware notified me my virtual machines were stopping I rushed to track down the culprit.
Thank you, I’ve blown away the qcow2 archive and restarted, got a watch on the directory so I can see if this explodes again. Presently at 700M, much better than 60G.
Update: Holding steady at 1.1G (1154678784 bytes) just as expected
I was wondering if there were any updates on this. Right now my only fix to recover the free space is to wipe the Docker.qcow2 and let the app re-create it.
3 question i have is:
1 Is there a way to specify a size for the qcow file to be larger then 64G. That’s too small for my use cases and usually run out of space after a few weeks.
2 My previous pattern to recover free space was using this script:
#!/usr/bin/env bash # remove untagged images docker stop $(docker ps -aq) docker rm $(docker ps -aq) docker rmi $(docker images --filter dangling=true -q) # remove unused volumes # remove stopped + exited containers, I skip Exit 0 as I have old scripts using data containers. docker rm -v $(docker ps -a | grep "Exit [1-255]" | awk '{ print $1 }')
Is there a better way of doing this or is delete the image file and resetting the only valid solution atm?
3 Is there a way to ssh into the VM? that’s running the linux kernel? Or is that abstracted away ?