Vulnerabilities

Velociraptor versions prior to 0.76.4 contain a cross organization authorization bypass in the HTTP API. A user with only the reader role in the root organization (the lowest authenticated role, holding only READ_RESULTS permission ) can issue a single authenticated HTTP GET that can read any files from other orgs - even if they have no explicit permissions in the target org.<br /> <br /> <br /> <br /> However, the problem does not occur in reverse - a user with read access to a sub org is unable to read from other org or the root org.

The MongoDB C Driver&amp;#39;s Cyrus SASL integration performs unsafe string copying during username canonicalization, enabling a heap buffer overflow before any authentication or network traffic. This may be triggered by passing untrusted input in the username of a MongoDB URI with authMechanism=GSSAPI.

Cross Site Scripting vulnerability in Juzaweb CMS v.5.0.0 allows a remote attacker via execute arbitrary code via a crafted script to the Add Banner Ads function

A type confusion vulnerability in Qt SVG allows an attacker to cause an application crash via a crafted SVG image.<br /> <br /> <br /> <br /> When processing SVG marker references, the renderer retrieves a node by its id attribute and casts it to QSvgMarker* without verifying the node type. A non-marker element (such as a element) that references itself as a marker triggers an out-of-bounds heap read due to the object size difference between QSvgLine and QSvgMarker,<br /> followed by an endless recursion that bypasses the marker recursion <br /> guard through incorrect virtual dispatch. The result is an application <br /> crash (denial of service).<br /> <br /> <br /> <br /> This issue affects Qt SVG: <br /> from 6.7.0 before 6.8.8, from 6.9.0 before 6.11.1.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> APEI/GHES: ensure that won&amp;#39;t go past CPER allocated record<br /> <br /> The logic at ghes_new() prevents allocating too large records, by<br /> checking if they&amp;#39;re bigger than GHES_ESTATUS_MAX_SIZE (currently, 64KB).<br /> Yet, the allocation is done with the actual number of pages from the<br /> CPER bios table location, which can be smaller.<br /> <br /> Yet, a bad firmware could send data with a different size, which might<br /> be bigger than the allocated memory, causing an OOPS:<br /> <br /> Unable to handle kernel paging request at virtual address fff00000f9b40000<br /> Mem abort info:<br /> ESR = 0x0000000096000007<br /> EC = 0x25: DABT (current EL), IL = 32 bits<br /> SET = 0, FnV = 0<br /> EA = 0, S1PTW = 0<br /> FSC = 0x07: level 3 translation fault<br /> Data abort info:<br /> ISV = 0, ISS = 0x00000007, ISS2 = 0x00000000<br /> CM = 0, WnR = 0, TnD = 0, TagAccess = 0<br /> GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0<br /> swapper pgtable: 4k pages, 52-bit VAs, pgdp=000000008ba16000<br /> [fff00000f9b40000] pgd=180000013ffff403, p4d=180000013fffe403, pud=180000013f85b403, pmd=180000013f68d403, pte=0000000000000000<br /> Internal error: Oops: 0000000096000007 [#1] SMP<br /> Modules linked in:<br /> CPU: 0 UID: 0 PID: 303 Comm: kworker/0:1 Not tainted 6.19.0-rc1-00002-gda407d200220 #34 PREEMPT<br /> Hardware name: QEMU QEMU Virtual Machine, BIOS unknown 02/02/2022<br /> Workqueue: kacpi_notify acpi_os_execute_deferred<br /> pstate: 214020c5 (nzCv daIF +PAN -UAO -TCO +DIT -SSBS BTYPE=--)<br /> pc : hex_dump_to_buffer+0x30c/0x4a0<br /> lr : hex_dump_to_buffer+0x328/0x4a0<br /> sp : ffff800080e13880<br /> x29: ffff800080e13880 x28: ffffac9aba86f6a8 x27: 0000000000000083<br /> x26: fff00000f9b3fffc x25: 0000000000000004 x24: 0000000000000004<br /> x23: ffff800080e13905 x22: 0000000000000010 x21: 0000000000000083<br /> x20: 0000000000000001 x19: 0000000000000008 x18: 0000000000000010<br /> x17: 0000000000000001 x16: 00000007c7f20fec x15: 0000000000000020<br /> x14: 0000000000000008 x13: 0000000000081020 x12: 0000000000000008<br /> x11: ffff800080e13905 x10: ffff800080e13988 x9 : 0000000000000000<br /> x8 : 0000000000000000 x7 : 0000000000000001 x6 : 0000000000000020<br /> x5 : 0000000000000030 x4 : 00000000fffffffe x3 : 0000000000000000<br /> x2 : ffffac9aba78c1c8 x1 : ffffac9aba76d0a8 x0 : 0000000000000008<br /> Call trace:<br /> hex_dump_to_buffer+0x30c/0x4a0 (P)<br /> print_hex_dump+0xac/0x170<br /> cper_estatus_print_section+0x90c/0x968<br /> cper_estatus_print+0xf0/0x158<br /> __ghes_print_estatus+0xa0/0x148<br /> ghes_proc+0x1bc/0x220<br /> ghes_notify_hed+0x5c/0xb8<br /> notifier_call_chain+0x78/0x148<br /> blocking_notifier_call_chain+0x4c/0x80<br /> acpi_hed_notify+0x28/0x40<br /> acpi_ev_notify_dispatch+0x50/0x80<br /> acpi_os_execute_deferred+0x24/0x48<br /> process_one_work+0x15c/0x3b0<br /> worker_thread+0x2d0/0x400<br /> kthread+0x148/0x228<br /> ret_from_fork+0x10/0x20<br /> Code: 6b14033f 540001ad a94707e2 f100029f (b8747b44)<br /> ---[ end trace 0000000000000000 ]---<br /> <br /> Prevent that by taking the actual allocated are into account when<br /> checking for CPER length.<br /> <br /> [ rjw: Subject tweaks ]

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> dm: clear cloned request bio pointer when last clone bio completes<br /> <br /> Stale rq-&gt;bio values have been observed to cause double-initialization of<br /> cloned bios in request-based device-mapper targets, leading to<br /> use-after-free and double-free scenarios.<br /> <br /> One such case occurs when using dm-multipath on top of a PCIe NVMe<br /> namespace, where cloned request bios are freed during<br /> blk_complete_request(), but rq-&gt;bio is left intact. Subsequent clone<br /> teardown then attempts to free the same bios again via<br /> blk_rq_unprep_clone().<br /> <br /> The resulting double-free path looks like:<br /> <br /> nvme_pci_complete_batch()<br /> nvme_complete_batch()<br /> blk_mq_end_request_batch()<br /> blk_complete_request() // called on a DM clone request<br /> bio_endio() // first free of all clone bios<br /> ...<br /> rq-&gt;end_io() // end_clone_request()<br /> dm_complete_request(tio-&gt;orig)<br /> dm_softirq_done()<br /> dm_done()<br /> dm_end_request()<br /> blk_rq_unprep_clone() // second free of clone bios<br /> <br /> Fix this by clearing the clone request&amp;#39;s bio pointer when the last cloned<br /> bio completes, ensuring that later teardown paths do not attempt to free<br /> already-released bios.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ALSA: usb-audio: Add sanity check for OOB writes at silencing<br /> <br /> At silencing the playback URB packets in the implicit fb mode before<br /> the actual playback, we blindly assume that the received packets fit<br /> with the buffer size. But when the setup in the capture stream<br /> differs from the playback stream (e.g. due to the USB core limitation<br /> of max packet size), such an inconsistency may lead to OOB writes to<br /> the buffer, resulting in a crash.<br /> <br /> For addressing it, add a sanity check of the transfer buffer size at<br /> prepare_silent_urb(), and stop the data copy if the received data<br /> overflows. Also, report back the transfer error properly from there,<br /> too.<br /> <br /> Note that this doesn&amp;#39;t fix the root cause of the playback error<br /> itself, but this merely covers the kernel Oops.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> drm/xe: Add bounds check on pat_index to prevent OOB kernel read in madvise<br /> <br /> When user provides a bogus pat_index value through the madvise IOCTL, the<br /> xe_pat_index_get_coh_mode() function performs an array access without<br /> validating bounds. This allows a malicious user to trigger an out-of-bounds<br /> kernel read from the xe-&gt;pat.table array.<br /> <br /> The vulnerability exists because the validation in madvise_args_are_sane()<br /> directly calls xe_pat_index_get_coh_mode(xe, args-&gt;pat_index.val) without<br /> first checking if pat_index is within [0, xe-&gt;pat.n_entries).<br /> <br /> Although xe_pat_index_get_coh_mode() has a WARN_ON to catch this in debug<br /> builds, it still performs the unsafe array access in production kernels.<br /> <br /> v2(Matthew Auld)<br /> - Using array_index_nospec() to mitigate spectre attacks when the value<br /> is used<br /> <br /> v3(Matthew Auld)<br /> - Put the declarations at the start of the block<br /> <br /> (cherry picked from commit 944a3329b05510d55c69c2ef455136e2fc02de29)

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> mailbox: Prevent out-of-bounds access in fw_mbox_index_xlate()<br /> <br /> Although it is guided that `#mbox-cells` must be at least 1, there are<br /> many instances of `#mbox-cells = ;` in the device tree. If that is<br /> the case and the corresponding mailbox controller does not provide<br /> `fw_xlate` and of_xlate` function pointers, `fw_mbox_index_xlate()` will<br /> be used by default and out-of-bounds accesses could occur due to lack of<br /> bounds check in that function.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> RDMA/ionic: Fix potential NULL pointer dereference in ionic_query_port<br /> <br /> The function ionic_query_port() calls ib_device_get_netdev() without<br /> checking the return value which could lead to NULL pointer dereference,<br /> Fix it by checking the return value and return -ENODEV if the &amp;#39;ndev&amp;#39; is<br /> NULL.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> net: ethernet: ec_bhf: Fix dma_free_coherent() dma handle<br /> <br /> dma_free_coherent() in error path takes priv-&gt;rx_buf.alloc_len as<br /> the dma handle. This would lead to improper unmapping of the buffer.<br /> <br /> Change the dma handle to priv-&gt;rx_buf.alloc_phys.

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> drm/atmel-hlcdc: fix memory leak from the atomic_destroy_state callback<br /> <br /> After several commits, the slab memory increases. Some drm_crtc_commit<br /> objects are not freed. The atomic_destroy_state callback only put the<br /> framebuffer. Use the __drm_atomic_helper_plane_destroy_state() function<br /> to put all the objects that are no longer needed.<br /> <br /> It has been seen after hours of usage of a graphics application or using<br /> kmemleak:<br /> <br /> unreferenced object 0xc63a6580 (size 64):<br /> comm "egt_basic", pid 171, jiffies 4294940784<br /> hex dump (first 32 bytes):<br /> 40 50 34 c5 01 00 00 00 ff ff ff ff 8c 65 3a c6 @P4..........e:.<br /> 8c 65 3a c6 ff ff ff ff 98 65 3a c6 98 65 3a c6 .e:......e:..e:.<br /> backtrace (crc c25aa925):<br /> kmemleak_alloc+0x34/0x3c<br /> __kmalloc_cache_noprof+0x150/0x1a4<br /> drm_atomic_helper_setup_commit+0x1e8/0x7bc<br /> drm_atomic_helper_commit+0x3c/0x15c<br /> drm_atomic_commit+0xc0/0xf4<br /> drm_atomic_helper_set_config+0x84/0xb8<br /> drm_mode_setcrtc+0x32c/0x810<br /> drm_ioctl+0x20c/0x488<br /> sys_ioctl+0x14c/0xc20<br /> ret_fast_syscall+0x0/0x54