summaryrefslogtreecommitdiff
path: root/libpod/container_internal.go
diff options
context:
space:
mode:
authorDaniel J Walsh <dwalsh@redhat.com>2019-03-14 08:33:53 -0400
committerDaniel J Walsh <dwalsh@redhat.com>2019-03-15 10:44:44 -0400
commit9d81be961491ac965f6cc9fd96884833c0165334 (patch)
treef107bb824b7f8af5168c33dee8d946f9d16d6d0d /libpod/container_internal.go
parent37dcc0a305a1606de7c0f5521d11250a4318bb51 (diff)
downloadpodman-9d81be961491ac965f6cc9fd96884833c0165334.tar.gz
podman-9d81be961491ac965f6cc9fd96884833c0165334.tar.bz2
podman-9d81be961491ac965f6cc9fd96884833c0165334.zip
Make sure buildin volumes have the same ownership and permissions as image
When creating a new image volume to be mounted into a container, we need to make sure the new volume matches the Ownership and permissions of the path that it will be mounted on. For example if a volume inside of a containre image is owned by the database UID, we want the volume to be mounted onto the image to be owned by the database UID. Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Diffstat (limited to 'libpod/container_internal.go')
-rw-r--r--libpod/container_internal.go4
1 files changed, 4 insertions, 0 deletions
diff --git a/libpod/container_internal.go b/libpod/container_internal.go
index 872802016..ac2d65342 100644
--- a/libpod/container_internal.go
+++ b/libpod/container_internal.go
@@ -1429,5 +1429,9 @@ func (c *Container) copyWithTarFromImage(src, dest string) error {
}
a := archive.NewDefaultArchiver()
source := filepath.Join(mountpoint, src)
+
+ if err = c.copyOwnerAndPerms(source, dest); err != nil {
+ return err
+ }
return a.CopyWithTar(source, dest)
}