An issue was discovered in aiopoll() in fs/aio.c in the Linux kernel through 5.0.4. A file may be released by aiopollwake() if an expected event is triggered immediately (e.g., by the close of a pair of pipes) after the return of vfspoll(), and this will cause a use-after-free.

systemd 242 changes the VT1 mode upon a logout, which allows attackers to read cleartext passwords in certain circumstances, such as watching a shutdown, or using Ctrl-Alt-F1 and Ctrl-Alt-F2. This occurs because the KDGKBMODE (aka current keyboard mode) check is mishandled.

A flaw was found in the Linux kernel’s implementation of the SAS expander subsystem, where a race condition exists in the smptasktimedout() and smptaskdone() in drivers/scsi/libsas/sasexpander.c. An attacker could abuse this flaw to corrupt memory and escalate privileges.

The TLS protocol 1.2 and earlier supports the rsafixeddh, dssfixeddh, rsafixedecdh, and ecdsafixedecdh values for ClientCertificateType but does not directly document the ability to compute the master secret in certain situations with a client secret key and server public key but not a server secret key, which makes it easier for man-in-the-middle attackers to spoof TLS servers by leveraging knowledge of the secret key for an arbitrary installed client X.509 certificate, aka the "Key Compromise Impersonation (KCI)" issue.

An issue was discovered in rdstcpkillsock in net/rds/tcp.c in the Linux kernel before 5.0.8. There is a race condition leading to a use-after-free, related to net namespace cleanup.

A race condition was found in the way Linux kernel's memory subsystem handled breakage of the read only private mappings COW situation on write access.

An unprivileged local user could use this flaw to gain write access to otherwise read only memory mappings and thus increase their privileges on the system.

Red Hat is aware of this issue and if you have questions about the affectedness of your system please contact Red Hat Support. For additional information see https://access.redhat.com/security/vulnerabilities/2706661

A flaw was found in the Linux kernels with commit b6a2fea39318 ("mm: variable length argument support", from July 19, 2007) but without commit da029c11e6b1 ("exec:Limit arg stack to at most 75% of STKLIM", from July 7, 2017). An integer overflow in the Linux kernel's createelftables() function. A local attacker can exploit this vulnerability via a SUID-root binary and obtain full root privileges.

Referenced commits: b6a2fea39318 ("mm: variable length argument support", from July 19, 2007) https://github.com/torvalds/linux/commit/b6a2fea39318e43fee84fa7b0b90d68bed92d2ba

da029c11e6b1 ("exec: Limit arg stack to at most 75% of STKLIM", from July 7, 2017) https://github.com/torvalds/linux/commit/da029c11e6b12f321f36dac8771e833b65cec962

Additional references: https://www.qualys.com/2018/09/25/cve-2018-14634/mutagen-astronomy-integer-overflow-linux-createelftables-cve-2018-14634.txt

It was discovered that a systemd service that uses DynamicUser property can create a SUID/SGID binary that would be allowed to run as the transient service UID/GID even after the service is terminated. A local attacker may use this flaw to access resources that will be owned by a potentially different service in the future, when the UID/GID will be recycled.

It was discovered that a systemd service that uses DynamicUser property can get new privileges through the execution of SUID binaries, which would allow to create binaries owned by the service transient group with the setgid bit set. A local attacker may use this flaw to access resources that will be owned by a potentially different service in the future, when the GID will be recycled.

An issue where a provided address with accessok() is not checked was discovered in i915gemexecbuffer2ioctl in drivers/gpu/drm/i915/i915gemexecbuffer.c in the Linux kernel through 4.19.13. A local attacker can craft a malicious IOCTL function call to overwrite arbitrary kernel memory, resulting in a Denial of Service or privilege escalation.

In the Linux kernel before 4.20.5, attackers can trigger a drivers/char/ipmi/ipmimsghandler.c use-after-free and OOPS by arranging for certain simultaneous execution of the code, as demonstrated by a "service ipmievd restart" loop.

In the Linux kernel before 4.20.12, net/ipv4/netfilter/nfnatsnmpbasicmain.c in the SNMP NAT module has insufficient ASN.1 length checks (aka an array index error), making out-of-bounds read and write operations possible, leading to an OOPS or local privilege escalation. This affects snmpversion and snmphelper.

An infinite loop issue was found in the vhostnet kernel module in Linux Kernel up to and including v5.1-rc6, while handling incoming packets in handlerx(). It could occur if one end sends packets faster than the other end can process them. A guest user, maybe remote one, could use this flaw to stall the vhostnet kernel thread, resulting in a DoS scenario.

The Siemens R3964 line discipline driver in drivers/tty/nr3964.c in the Linux kernel before 5.0.8 has multiple race conditions.

It was found that the cgroup limitation of system resources used by Kubernetes can be bypassed. A guest pod can be used to consume a large amount of system memory.

A suggested upstream patch set:

https://lore.kernel.org/netdev/20190401113110.GA20717@hmswarspite.think-freely.org/T/#u

A race condition in perfeventopen() allows local attackers to leak sensitive data from setuid programs. As no relevant locks (in particular the credguardmutex) are held during the ptracemayaccess() call, it is possible for the specified target task to perform an execve() syscall with setuid execution before perfeventalloc() actually attaches to it, allowing an attacker to bypass the ptracemayaccess() check and the perfeventexittask(current) call that is performed in installexeccreds() during privileged execve() calls.

A vulnerability was found in OpenSSL 1.0.2. When an application encounters a fatal protocol error and then calls SSLshutdown() twice, OpenSSL can respond differently to the calling application if a 0 byte record is received with invalid padding compared to if a 0 byte record is received with an invalid MAC. This difference in behaviour can be detected by a remote peer, then this amounts to a padding oracle that could be used to decrypt data. In order for this to be exploitable "non-stitched" ciphersuites must be in use. Also the application must call SSLshutdown() twice even if a protocol error has occurred (applications should not do this but some do anyway). AEAD ciphersuites are not impacted. This issue does not impact OpenSSL 1.1.1 or 1.1.0.

Upstream bug: https://www.openssl.org/news/secadv/20190226.txt

Upstream Patch: https://github.com/openssl/openssl/commit/e9bbefbf0f24c57645e7ad6a5a71ae649d18ac8e

A flaw was found in the Linux kernel vfio interface implementation that permits violation of the user's locked memory limit. If a device is bound to a vfio driver, such as vfio-pci, and the local attacker is administratively granted ownership of the device, it may cause a system memory exhaustion and thus a denial of service (DoS).

References:

https://seclists.org/oss-sec/2019/q2/6

A suggested fix:

https://lore.kernel.org/lkml/155414977872.12780.13728555131525362206.stgit@gimli.home/T/#u