CVE-2026-99307: ceph: require the dentry lease record to cover struct ceph_mds_reply_lease
In the Linux kernel, the following vulnerability has been resolved:
ceph: require the dentry lease record to cover struct cephmdsreplylease
parsereplyinfolease() reads a 32-bit structlen off the wire and only checks that many bytes are present, without requiring structlen to cover sizeof(struct cephmdsreplylease) (10 bytes). A malicious MDS can declare a shorter record and still get the pointer published: the following p += sizeof(lease) overshoot is undone by p = lend, so nothing downstream rejects the reply. cephfilltrace() -> updatedentrylease() then reads 10 bytes through that pointer on any lookup on a mounted CephFS, running past the end of the message front.
Require the record to cover the structure, matching the sibling parsereplyinfodir(). A conforming MDS always encodes the full 10 bytes.
BUG: KASAN: slab-out-of-bounds in updatedentrylease.constprop.0 (fs/ceph/inode.c:1459) Read of size 4 at addr ffff888026c63d65 by task kworker/0:3/2366 Workqueue: ceph-msgr cephconworkfn Call Trace: updatedentrylease.constprop.0 (fs/ceph/inode.c:1459) cephfilltrace (fs/ceph/inode.c:1474 fs/ceph/inode.c:1823) mdsdispatch (fs/ceph/mdsclient.c:4229 fs/ceph/mdsclient.c:7225) cephconprocessmessage (net/ceph/messenger.c:1424) cephconv1tryread (net/ceph/messengerv1.c:1430) cephconworkfn (net/ceph/messenger.c:1576) processonework (kernel/workqueue.c:3396) workerthread (kernel/workqueue.c:3479 kernel/workqueue.c:3560) kthread (kernel/kthread.c:436) retfromfork (arch/x86/kernel/process.c:158) retfromforkasm (arch/x86/entry/entry64.S:245)
The buggy address is located 357 bytes inside of allocated 360-byte region [ffff888026c63c00, ffff888026c63d68)