File _patchinfo of Package patchinfo.9278

<patchinfo incident="9278">
  <issue tracker="bnc" id="1108835">SES5: application cores are not being written too /var/lib/systemd/coredump by systemd.</issue>
  <issue tracker="bnc" id="1106923">systemd fails to execute ExecStopPost command from service unit file</issue>
  <issue tracker="bnc" id="1109252">audit/systemd-journal allocates lots of unfreeable slabs</issue>
  <issue tracker="bnc" id="1111278">udev: IMPORT{cmdline} produces highly unexpected results</issue>
  <issue tracker="bnc" id="1113665">VUL-0: CVE-2018-15686: systemd: Line splitting via fgets() allows for state injection during daemon-reexec</issue>
  <issue tracker="bnc" id="1110445">shutdown hangs if swap is disabled before (full) tmpfs file systems are unmounted</issue>
  <issue tracker="bnc" id="1112024">logind does not properly recognize that a client is already inside a session</issue>
  <issue tracker="bnc" id="1113632">VUL-0: CVE-2018-15688: systemd:  Out-of-bounds heap write in systemd-networkd dhcpv6 option handling</issue>
  <issue tracker="bnc" id="1113083">systemd: emergency.service fails if /root mount is missing</issue>
  <issue tracker="cve" id="2018-15688"/>
  <issue tracker="cve" id="2018-15686"/>
  <category>security</category>
  <rating>important</rating>
  <packager>fbui</packager>
  <description>
This update for systemd fixes the following issues:

Security issues fixed:

- CVE-2018-15688: A buffer overflow vulnerability in the dhcp6 client of systemd allowed a malicious dhcp6 server to overwrite heap memory in systemd-networkd. (bsc#1113632)
- CVE-2018-15686: A vulnerability in unit_deserialize of systemd allows an attacker to supply arbitrary state across systemd re-execution via NotifyAccess. This can be used to improperly influence systemd execution and possibly lead to root privilege escalation. (bsc#1113665)

Non-security issues fixed:

- dhcp6: split assert_return() to be more debuggable when hit
- core: skip unit deserialization and move to the next one when unit_deserialize() fails
- core: properly handle deserialization of unknown unit types (#6476)
- core: don't create Requires for workdir if "missing ok" (bsc#1113083)
- logind: use manager_get_user_by_pid() where appropriate
- logind: rework manager_get_{user|session}_by_pid() a bit
- login: fix user@.service case, so we don't allow nested sessions (#8051) (bsc#1112024)
- core: be more defensive if we can't determine per-connection socket peer (#7329)
- socket-util: introduce port argument in sockaddr_port()
- service: fixup ExecStop for socket-activated shutdown (#4120)
- service: Continue shutdown on socket activated unit on termination (#4108) (bsc#1106923)
- cryptsetup: build fixes for "add support for sector-size= option"
- udev-rules: IMPORT cmdline does not recognize keys with similar names (bsc#1111278)
- core: keep the kernel coredump defaults when systemd-coredump is disabled
- core: shorten main() a bit, split out coredump initialization
- core: set RLIMIT_CORE to unlimited by default (bsc#1108835)
- core/mount: fstype may be NULL
- journald: don't ship systemd-journald-audit.socket (bsc#1109252)
- core: make "tmpfs" dependencies on swapfs a "default" dep, not an "implicit" (bsc#1110445)
- mount: make sure we unmount tmpfs mounts before we deactivate swaps (#7076)
- tmp.mount.hm4: After swap.target (#3087)

- Ship systemd-sysv-install helper via the main package
  This script was part of systemd-sysvinit sub-package but it was
  wrong since systemd-sysv-install is a script used to redirect
  enable/disable operations to chkconfig when the unit targets are
  sysv init scripts. Therefore it's never been a SySV init tool.
</description>
  <summary>Security update for systemd</summary>
</patchinfo>