kernel KVM before versions kernel 4.16, kernel 4.16-rc7, kernel 4.17-rc1, kernel 4.17-rc2 and kernel 4.17-rc3 is vulnerable to a flaw in the way the Linux kernel's KVM hypervisor handled exceptions delivered after a stack switch operation via Mov SS or Pop SS instructions. During the stack switch operation, the processor did not deliver interrupts and exceptions, rather they are delivered once the first instruction after the stack switch is executed. An unprivileged KVM guest user could use this flaw to crash the guest or, potentially, escalate their privileges in the guest.

An issue was discovered in HAProxy before 1.8.8. The incoming H2 frame length was checked against the max_frame_size setting instead of being checked against the bufsize. The max_frame_size only applies to outgoing traffic and not to incoming, so if a large enough frame size is advertised in the SETTINGS frame, a wrapped frame will be defragmented into a temporary allocated buffer where the second fragment may overflow the heap by up to 16 kB. It is very unlikely that this can be exploited for code execution given that buffers are very short lived and their addresses not realistically predictable in production, but the likelihood of an immediate crash is absolutely certain.

389-ds-base before version 1.3.6 is vulnerable to an improperly NULL terminated array in the uniqueness_entry_to_config() function in the "attribute uniqueness" plugin of 389 Directory Server. An authenticated, or possibly unauthenticated, attacker could use this flaw to force an out-of-bound heap memory read, possibly triggering a crash of the LDAP service.

Unbounded memory allocation in Google Guava 11.0 through 24.x before 24.1.1 allows remote attackers to conduct denial of service attacks against servers that depend on this library and deserialize attacker-provided data, because the AtomicDoubleArray class (when serialized with Java serialization) and the CompoundOrdering class (when serialized with GWT serialization) perform eager allocation without appropriate checks on what a client has sent and whether the data size is reasonable.

The DPDK vhost-user interface does not check to verify that all the requested guest physical range is mapped and contiguous when performing Guest Physical Addresses to Host Virtual Addresses translations. This may lead to a malicious guest exposing vhost-user backend process memory. All versions before 18.02.1 are vulnerable.

pcs before version 0.9.164 and 0.10 is vulnerable to a privilege escalation via authorized user malicious REST call. The REST interface of the pcsd service did not properly sanitize the file name from the /remote/put_file query. If the /etc/booth directory exists, an authenticated attacker with write permissions could create or overwrite arbitrary files with arbitrary data outside of the /etc/booth directory, in the context of the pcsd process.

rhnreg_ks in Red Hat Network Client Tools (aka rhn-client-tools) on Red Hat Gluster Storage 2.1 and Enterprise Linux (RHEL) 5, 6, and 7 does not properly validate hostnames in X.509 certificates from SSL servers, which allows remote attackers to prevent system registration via a man-in-the-middle attack.

In Ruby before 2.2.10, 2.3.x before 2.3.7, 2.4.x before 2.4.4, 2.5.x before 2.5.1, and 2.6.0-preview1, an attacker controlling the unpacking format (similar to format string vulnerabilities) can trigger a buffer under-read in the String#unpack method, resulting in a massive and controlled information disclosure.

In Ruby before 2.2.10, 2.3.x before 2.3.7, 2.4.x before 2.4.4, 2.5.x before 2.5.1, and 2.6.0-preview1, an attacker can pass a large HTTP request with a crafted header to WEBrick server or a crafted body to WEBrick server/handler and cause a denial of service (memory consumption).

Directory traversal vulnerability in the Dir.mktmpdir method in the tmpdir library in Ruby before 2.2.10, 2.3.x before 2.3.7, 2.4.x before 2.4.4, 2.5.x before 2.5.1, and 2.6.0-preview1 might allow attackers to create arbitrary directories or files via a .. (dot dot) in the prefix argument.

In Apache httpd 2.2.0 to 2.4.29, when generating an HTTP Digest authentication challenge, the nonce sent to prevent reply attacks was not correctly generated using a pseudo-random seed. In a cluster of servers using a common Digest authentication configuration, HTTP requests could be replayed across servers by an attacker without detection.

A specially crafted request could have crashed the Apache HTTP Server prior to version 2.4.30, due to an out of bound access after a size limit is reached by reading the HTTP header. This vulnerability is considered very hard if not impossible to trigger in non-debug mode (both log and build level), so it is classified as low risk for common server usage.

In Apache httpd 2.4.0 to 2.4.29, when mod_session is configured to forward its session data to CGI applications (SessionEnv on, not the default), a remote user may influence their content by using a "Session" header. This comes from the "HTTP_SESSION" variable name used by mod_session to forward its data to CGIs, since the prefix "HTTP_" is also used by the Apache HTTP Server to pass HTTP header fields, per CGI specifications.

In Apache httpd 2.4.0 to 2.4.29, the expression specified in <FilesMatch> could match '$' to a newline character in a malicious filename, rather than matching only the end of the filename. This could be exploited in environments where uploads of some files are are externally blocked, but only by matching the trailing portion of the filename.

In Apache httpd 2.0.23 to 2.0.65, 2.2.0 to 2.2.34, and 2.4.0 to 2.4.29, mod_authnz_ldap, if configured with AuthLDAPCharsetConfig, uses the Accept-Language header value to lookup the right charset encoding when verifying the user's credentials. If the header value is not present in the charset conversion table, a fallback mechanism is used to truncate it to a two characters value to allow a quick retry (for example, 'en-US' is truncated to 'en'). A header value of less than two characters forces an out of bound write of one NUL byte to a memory location that is not part of the string. In the worst case, quite unlikely, the process would crash which could be used as a Denial of Service attack. In the more likely case, this memory is already reserved for future use and the issue has no effect at all.

org.slf4j.ext.EventData in the slf4j-ext module in QOS.CH SLF4J before 1.8.0-beta2 allows remote attackers to bypass intended access restrictions via crafted data. EventData in the slf4j-ext module in QOS.CH SLF4J, has been fixed in SLF4J versions 1.7.26 later and in the 2.0.x series.

Samba before versions 4.6.1, 4.5.7 and 4.4.11 are vulnerable to a malicious client using a symlink race to allow access to areas of the server file system not exported under the share definition.

Buffer overflow in the PixarLogDecode function in tif_pixarlog.c in LibTIFF 4.0.6 and earlier allows remote attackers to cause a denial of service (application crash) or possibly have unspecified other impact via a crafted TIFF image, as demonstrated by overwriting the vgetparent function pointer with rgb2ycbcr.

Apache HTTP Server mod_cluster before version httpd 2.4.23 is vulnerable to an Improper Input Validation in the protocol parsing logic in the load balancer resulting in a Segmentation Fault in the serving httpd process.

It was found that when Artemis and HornetQ before 2.4.0 are configured with UDP discovery and JGroups discovery a huge byte array is created when receiving an unexpected multicast message. This may result in a heap memory exhaustion, full GC, or OutOfMemoryError.