

People reporting the issue and piling on clearly don't understand sparse files (which is fine in and of itself), and their claims that "the rest of the system treats the spare file as regular file" are just not very credible to me.

It now states avail 45Gb(above there were 40Gb).īut at least everything works now, but absolutely no understanding where is shadowed consumption hides.I took a brief look at the issue and it's very strange. I still see the same size for qemu file: %ls -lah 1 usss staff 21G May 27 00:46 /Users/usss/Library/Containers//Data/64-linux/Docker.qcow2īut usage form the container point of view changed a bit df -h Looks possible?Īnd installed qemu and added +5Gb to the image as was written in the article: %qemu-img resize ~/Library/Containers//Data/64-linux/Docker.qcow2 +5G Lets imagine, that despite df -h inside container and docker desktop ui settings reports about 64Gb, i still have 20Gb limitation. Was keep on reading all kinds if articles and after reading Īnd noticed that strange 20Gb limitation, default in 2017. Partial decision without qcow clear but no understanding I see that temp workaround is clear all data, but hope to find alternative, or at least some kind of explanation about disk usage metrics before i go brute clear. But seems this issue is about similar probem as my, cause issue author qcow2 file size is seems to be big enough. I saw before issues about misbehave, where deleting of qcow2 file was a fix, but always there really was a space issue where used and configured sizes where close, and the problem was in resizing. I also found something very similar as an issue on gh. So docker is not starting at all at the moment. I tried to restart docker app and it all ended in fatal space error. Where is configuration of max container disk size placed? How can i see the real container usage stats so i can then apply some tuning. The most usefull and detailed exmple I found is īut, it seems that, at my current docker version output is no more contains such fields as author uses for illustrations - no "base device size", no "DeviceSize"

Are those values are the things i should look for? None of those values look like container usage with du (shown above).

So seems i'am using those tools in a wrong way i interpret numbers in a worng way.ĬONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES SIZEģ1a71014ad95 nvm:nvm_tag "/bin/bash" 6 hours ago Exited (137) 23 minutes ago cr 3.52GB (virtual 5.44GB) I saw ps -s, inspect, system df as a recomended ways to estimate container disk usage stats, but has no success. Should be mentioned I'v read about general image file size on macos, it seems ok and far from default 64Gb size: ls -lah 1 coin staff 21G May 26 22:25 /Users/usss/Library/Containers//Data/64-linux/Docker.qcow2ĭocker container definetaly contains a big subdir: du -hs. ( shared_from_host - is shared volume for intepoeration with mother host) I have a container with following df stats inside it: df -hįilesystem Size Used Avail Use% Mounted on The worst thing is I can't determine actual container size from the mother host system point of view (info/inpsect/system df tools), in fact stats from docker and mother hosts does not equal :/ Faced with no space left on device error inside container and can't determine where does overflow come from.
