CVE-2026-99306 in Linuxinfo

Summary

by MITRE • 11/16/2026

In the Linux kernel, the following vulnerability has been resolved:

ceph: reject mdsmaps that claim more than CEPH_MAX_MDS ranks

check_new_map() sizes its export-target bitmap from CEPH_MAX_MDS but walks it with newmap->possible_max_rank, which comes straight off the wire: ceph_mdsmap_decode() sets it to max(m_num_active_mds, m_max_mds), then overwrites it with the size of the "in" set. Neither is bounded by CEPH_MAX_MDS, only by m_info allocating, so a monitor announcing more ranks than the bitmap holds makes test_bit() read past the end of targets[]. The
clear_bit() loop above indexes the same bitmap, though it is additionally bounded by mdsc->max_sessions. The map arrives on the monitor connection, which is unauthenticated in legacy msgr mode.

CEPH_MAX_MDS is a protocol constant from src/mds/mdstypes.h, and the reserved inode ranges in super.h are laid out one rank apart over that span, so a map claiming more ranks is corrupt. Reject it where the rank space is established rather than letting the count reach check_new_map().

BUG: KASAN: stack-out-of-bounds in check_new_map (fs/ceph/mds_client.c:5968) Read of size 8 at addr ffffc90000ac7798 by task kworker/1:2/127 Workqueue: ceph-msgr ceph_con_workfn Call Trace: <TASK> dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) print_report (mm/kasan/report.c:378 mm/kasan/report.c:482) kasan_report (mm/kasan/report.c:595) kasan_check_range (mm/kasan/generic.c:186 mm/kasan/generic.c:200) check_new_map (fs/ceph/mds_client.c:5968) ceph_mdsc_handle_mdsmap (fs/ceph/mds_client.c:7060) extra_mon_dispatch (fs/ceph/super.c:802) mon_dispatch (net/ceph/mon_client.c:1483) ceph_con_process_message (net/ceph/messenger.c:1424) ceph_con_v1_try_read (net/ceph/messenger_v1.c:1430) ceph_con_workfn (net/ceph/messenger.c:1576) process_one_work (kernel/workqueue.c:3396) worker_thread (kernel/workqueue.c:3479 kernel/workqueue.c:3560) kthread (kernel/kthread.c:436) ret_from_fork (arch/x86/kernel/process.c:158) ret_from_fork_asm (arch/x86/entry/entry_64.S:245) </TASK> The buggy address belongs to stack of task kworker/1:2/127 and is located at offset 288 in frame: check_new_map (fs/ceph/mds_client.c:5846) This frame has 1 object: [32, 288) 'targets'

Be aware that VulDB is the high quality source for vulnerability data.

Analysis

by VulDB Data Team • 11/16/2026

The Linux kernel's Ceph filesystem client contains a critical out-of-bounds read vulnerability within the check_new_map function located in fs/ceph/mds_client.c. This flaw arises from an inconsistency between how the export-target bitmap is allocated and how it is subsequently accessed during the processing of Monitor Daemon maps. The bitmap, referred to as targets, is sized based on CEPH_MAX_MDS, a protocol constant that defines the maximum number of Metadata Server ranks supported by the implementation. However, the iteration logic relies on newmap->possible_max_rank, which is derived directly from network data without sufficient validation against this hard limit. Specifically, during the decoding process in ceph_mds_map_decode, the possible_max_rank field is initially set to the maximum of active MDS count and configured max MDS count, but it is subsequently overwritten by the size of the in set received from the monitor. Neither of these source values is strictly bounded by CEPH_MAX_MDS before being used to index into the fixed-size bitmap.

This architectural mismatch allows a malicious or compromised Monitor Daemon to announce a map claiming more ranks than the kernel client can safely handle. When check_new_map executes, it attempts to read from and clear bits in the targets array using indices derived from this unbounded rank count. If the announced number of ranks exceeds CEPH_MAX_MDS, the test_bit operation reads past the end of the allocated stack memory for the targets array. This results in a kernel stack out-of-bounds read, as confirmed by Kernel Address Sanitizer reports indicating an eight-byte read at an address located beyond the valid frame boundaries. The vulnerability is particularly severe because it occurs during the processing of messages on the monitor connection, which operates in legacy msgr mode and lacks authentication mechanisms. Consequently, any entity capable of injecting or manipulating network traffic to the Ceph cluster can trigger this condition without needing prior credentials.

The operational impact of this vulnerability includes potential information disclosure through the leakage of kernel stack contents that may contain sensitive data such as pointers, cryptographic keys, or other process-specific state. Furthermore, depending on how the out-of-bounds read interacts with subsequent memory operations or control flow logic, it could potentially lead to a denial of service by causing a kernel panic or crash via an oops condition triggered by KASAN reporting or general protection faults in production environments where sanitizers are not enabled. The vulnerability affects the integrity and availability of Ceph storage clusters that rely on unauthenticated legacy messaging protocols for monitor communication. It represents a classic case of improper input validation where external data is trusted implicitly without verifying it against internal capacity constraints before use.

From a security taxonomy perspective, this issue aligns with CWE-125, Out-of-bounds Read, as the code accesses memory beyond its intended boundary due to insufficient bounds checking on array indices derived from untrusted sources. It also relates to CWE-20, Improper Input Validation, because the system fails to validate that the number of MDS ranks conforms to the expected protocol limits before processing. In terms of attack vectors and tactics, this vulnerability is exploitable via network-based attacks targeting the Ceph monitor service, fitting within MITRE ATT&CK techniques related to remote code execution or denial of service through memory corruption if exploited further, though primarily it facilitates reconnaissance and disruption in unauthenticated contexts.

Mitigation strategies should prioritize upgrading to kernel versions where this patch has been applied, ensuring that CEPH_MAX_MDS limits are strictly enforced during the decoding phase rather than at later processing stages. Administrators operating Ceph clusters must disable legacy msgr authentication if possible, moving to authenticated communication channels such as msgr2 which provides integrity and confidentiality protections against network-level manipulation. Additionally, implementing strict firewall rules to restrict access to monitor ports from untrusted networks reduces the attack surface for injecting malformed mdsmap messages. Regular auditing of kernel updates and applying security patches promptly is essential to maintain the resilience of distributed storage infrastructure against such protocol-level exploitation attempts.

Responsible

Linux

Reservation

09/25/2026

Disclosure

11/16/2026

Moderation

accepted

CPE

ready

EPSS

0.00000

KEV

no

Activities

very low

Sources

Do you know our Splunk app?

Download it now for free!