diff options
| author | Matthew Heon <matthew.heon@pm.me> | 2020-02-23 13:25:12 -0500 |
|---|---|---|
| committer | Matthew Heon <matthew.heon@pm.me> | 2020-02-24 09:29:34 -0500 |
| commit | 4004f646cd7a99f86b2098587c256071a89a4dc7 (patch) | |
| tree | 2dd8da33094ab8cbac7012473bfccd916ad2ae4f /libpod/runtime_ctr.go | |
| parent | 18dcb84d641f4e8ae7fa31fa446d9461e0c915ab (diff) | |
| download | podman-4004f646cd7a99f86b2098587c256071a89a4dc7.tar.gz podman-4004f646cd7a99f86b2098587c256071a89a4dc7.tar.bz2 podman-4004f646cd7a99f86b2098587c256071a89a4dc7.zip | |
Add basic deadlock detection for container start/remove
We can easily tell if we're going to deadlock by comparing lock
IDs before actually taking the lock. Add a few checks for this in
common places where deadlocks might occur.
This does not yet cover pod operations, where detection is more
difficult (and costly) due to the number of locks being involved
being higher than 2.
Also, add some error wrapping on the Podman side, so we can tell
people to use `system renumber` when it occurs.
Signed-off-by: Matthew Heon <matthew.heon@pm.me>
Diffstat (limited to 'libpod/runtime_ctr.go')
| -rw-r--r-- | libpod/runtime_ctr.go | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/libpod/runtime_ctr.go b/libpod/runtime_ctr.go index 3ad09f27c..39284026c 100644 --- a/libpod/runtime_ctr.go +++ b/libpod/runtime_ctr.go @@ -412,6 +412,9 @@ func (r *Runtime) removeContainer(ctx context.Context, c *Container, force bool, } // Lock the pod while we're removing container + if pod.config.LockID == c.config.LockID { + return errors.Wrapf(define.ErrWillDeadlock, "container %s and pod %s share lock ID %d", c.ID(), pod.ID(), c.config.LockID) + } pod.lock.Lock() defer pod.lock.Unlock() if err := pod.updatePod(); err != nil { |