After implementing #86 , we have evidence that the /mnt disk may not be empty when it is mounted for a CI job. We need to find a solution for this problem to improve the CI stability.
See this run for example.
There is a 65G disk0.img file, which seems to be a left-over from other jobs.
+ sudo ls -lah /mnt
total 69G
drwxr-xr-x 4 root root 4.0K Oct 18 14:24 .
drwxr-xr-x 23 root root 4.0K Oct 18 14:44 ..
-rw-r--r-- 1 root root 333 Oct 18 14:44 DATALOSS_WARNING_README.txt
-rw-r--r-- 1 root root 65G Oct 18 14:30 disk0.img
drwx------ 2 root root 16K Oct 18 13:10 lost+found
-rw------- 1 root root 4.0G Oct 18 13:11 swapfile
drwxrwxrwt 3 root root 4.0K Oct 18 14:30 tmp.VTSeRwLoad
+ sudo du -h -d1 /mnt
504K /mnt/tmp.VTSeRwLoad
16K /mnt/lost+found
69G /mnt
After implementing #86 , we have evidence that the /mnt disk may not be empty when it is mounted for a CI job. We need to find a solution for this problem to improve the CI stability.
See this run for example.
There is a 65G
disk0.imgfile, which seems to be a left-over from other jobs.