CVE-2024-49569: nvme-rdma: unquiesce admin_q before destroy it
In the Linux kernel, the following vulnerability has been resolved:
nvme-rdma: unquiesce admin_q before destroy it
Kernel will hang on destroy admin_q while we create ctrl failed, such
as following calltrace:
PID: 23644 TASK: ff2d52b40f439fc0 CPU: 2 COMMAND: "nvme"
#0 [ff61d23de260fb78] __schedule at ffffffff8323bc15
#1 [ff61d23de260fc08] schedule at ffffffff8323c014
#2 [ff61d23de260fc28] blk_mq_freeze_queue_wait at ffffffff82a3dba1
#3 [ff61d23de260fc78] blk_freeze_queue at ffffffff82a4113a
#4 [ff61d23de260fc90] blk_cleanup_queue at ffffffff82a33006
#5 [ff61d23de260fcb0] nvme_rdma_destroy_admin_queue at ffffffffc12686ce
#6 [ff61d23de260fcc8] nvme_rdma_setup_ctrl at ffffffffc1268ced
#7 [ff61d23de260fd28] nvme_rdma_create_ctrl at ffffffffc126919b
#8 [ff61d23de260fd68] nvmf_dev_write at ffffffffc024f362
#9 [ff61d23de260fe38] vfs_write at ffffffff827d5f25
RIP: 00007fda7891d574 RSP: 00007ffe2ef06958 RFLAGS: 00000202
RAX: ffffffffffffffda RBX: 000055e8122a4d90 RCX: 00007fda7891d574
RDX: 000000000000012b RSI: 000055e8122a4d90 RDI: 0000000000000004
RBP: 00007ffe2ef079c0 R8: 000000000000012b R9: 000055e8122a4d90
R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000004
R13: 000055e8122923c0 R14: 000000000000012b R15: 00007fda78a54500
ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b
This due to we have quiesced admi_q before cancel requests, but forgot
to unquiesce before destroy it, as a result we fail to drain the
pending requests, and hang on blk_mq_freeze_queue_wait() forever. Here
try to reuse nvme_rdma_teardown_admin_queue() to fix this issue and
simplify the code.
Security readout for executives and security teams
Plain-English summary
A failure while creating a Linux NVMe-over-RDMA storage controller can leave its administrative queue stuck during cleanup, potentially hanging the kernel indefinitely. The documented impact is loss of availability, not data theft or modification. Business risk is concentrated in systems using NVMe/RDMA storage rather than general Linux hosts.
Executive priority
Treat this as a high-priority availability issue for NVMe/RDMA-dependent infrastructure, where a kernel hang could interrupt storage-dependent services. Remediation urgency is lower for Linux systems that do not use this subsystem. No evidence supplied supports emergency incident response for active exploitation.
Technical view
The driver quiesces the NVMe/RDMA administrative queue before cancelling requests but previously failed to unquiesce it before destruction. Pending requests therefore cannot drain, and cleanup can block indefinitely in blk_mq_freeze_queue_wait(). The referenced Linux stable commits reuse the administrative-queue teardown path to correct the sequence.
Likely exposure
Prioritize Linux systems that actively use NVMe over RDMA, particularly where controller creation may fail. The supplied version data identifies affected and unaffected releases or commits, but does not clearly describe complete version ranges. Distribution kernels must be checked against vendor advisories or the referenced fixes.
Exploitation context
The CVSS 3.1 assessment is 7.5, describing network-reachable, low-complexity, unauthenticated availability impact. However, the supplied sources do not demonstrate a practical remote attack path or identify prerequisites beyond the NVMe/RDMA controller failure condition. This CVE is not listed as KEV, and no active exploitation evidence is provided.
Researcher notes
The failure occurs on the controller-creation error path while destroying admin_q. The queue remains quiesced, preventing pending requests from draining and causing an indefinite wait during queue cleanup. The fix changes teardown sequencing. The bundle contains no CWE assignment, proof of exploitation, or sufficient detail to confirm attacker control over the triggering failure.
Mitigation direction
Update to a vendor-supported kernel that incorporates the applicable referenced stable fix.
Confirm the running kernel, not merely the installed package, contains the fix.
Prioritize remediation on production hosts actively using NVMe-over-RDMA storage.
Check Linux distribution guidance before interpreting the supplied upstream version list.
Validation and detection
Inventory kernel builds on hosts configured for NVMe over RDMA.
Map each distribution kernel to the referenced upstream fixes or vendor advisory.
Review system logs for hangs involving nvme_rdma_destroy_admin_queue or blk_mq_freeze_queue_wait.
After updating, verify hosts rebooted into the intended fixed kernel.
Test controller-failure recovery in a controlled environment without affecting production storage.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cve · low confidence lookup
CVE-2024-49569 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
1CVSS vectors
3Timeline events
0ADP providers
4Source links
CVSS vector scores
1 official score
We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.