Bug Fixes

BZ#720458

If a server sent a response to an unbind request and the client simply closed the connection, Directory Server 8.2 logged "Netscape Portable Runtime error -5961 (TCP connection reset by peer.)".

BZ#752155

An incorrect SELinux context caused AVC errors in /var/log/audit/audit.log.

BZ#711266

The DS could not restart after a new object class was created which used the entryUSN attribute.

BZ#697663, BZ#700665, BZ#711533, BZ#711241, BZ#726136, BZ#700215

A number of memory leaks and performance errors were fixed.

BZ#712167

The ns-slapd process segfaulted if suffix referrals were enabled.

BZ#711513

A high volume of TCP traffic could cause the slapd process to quit responding to clients.

BZ#714298

Attempting to delete a VLV index caused the server to hang.

BZ#720051

Connections to the DS by an RSA authentication server using simple paged results by default would timeout.

BZ#735217

Running a simple paged search against a subtree with a host-based ACI would hang the server.

BZ#733443

If the target attribute list for an ACI had syntax errors and more than five attributes, the server crashed.

BZ#734267

It was not possible to set account lockout policies after upgrading from RHDS 8.1.

BZ#720452

Adding an entry with an RDN containing a % caused the server to crash.

BZ#709868

Only FIPS-supported ciphers can be used if the server is running in FIPS mode.

BZ#711265

It is possible to disable SSLv3 and only allow TLS.

BZ#713317, BZ#713318

If the changelog was encrypted and the certificate became corrupt, the server crashed.

BZ#733434

If the passwordisglobalpolicy attribute was enabled on a chained server, a secure connection to the master failed.

BZ#714310

If a chained database was replicated, the server could segfault.

BZ#694571

Editing a replication agreement to use SASL/GSS-API failed with GSS-API errors.

BZ#742611

In replication, a msgid may not be sent to the right thread, which caused "Bad parameter to an LDAP routine" errors. This causes failures to propagate up and halt replication.

BZ#701057

Password changes were replicated among masters replication, but not to consumers.

BZ#717066

If an entry was modified on RHDS and the corresponding entry was deleted on the Windows side, the sync operation attempts to use the wrong entry.

BZ#734831

Some changes were not properly synced over to RHDS from Windows.

BZ#726273

RHDS entries were not synced over to Windows if the user's CN had a comma.

BZ#718351

Intensive update loads on master servers could break the cache on the consumer, causing it to crash.

BZ#699458

Syncing a multi-valued attribute could delete all the other instances of that attribute when a new value was added.

BZ#729817

If a synced user subtree on Windows was deleted and then a user password was changed on the RHDS, the DS would crash.

Enhancements

BZ#742382

The nsslapd-idlistscanlimit configuration attribute can be set dynamically, instead of requiring a restart.

BZ#742661

Separate resource limits can be set for paged searches, independent of resource limits for regular searches.

BZ#720459

The sudo schema has been updated.

BZ#739959

A new configuration attribute sets a different list of replicated attributes for a total update versus an incremental update.

BZ#733440

A new configuration option allows the server to be started with an expired certificate.

BZ#720461

New TLS/SSL error messages have been added to the replication error log level.

Bug Fixes

BZ#701554

Password changes did not replicate because the method used to pass the changes to consumer servers was rejected on the consumer. This issue has been corrected, and password changes now replicate as expected.

BZ#701556

Values could be lost when group memberships were synchronized between 389 Directory Server and Active Directory with the Windows Sync feature. The synchronization and modify operations have been altered to prevent this issue, allowing group updates to synchronize with Active Directory.

BZ#701558

The ldclt command-line testing tool crashed during LDAP ADD operations because an LDAP attribute was not set correctly, preventing the creation of entries that did not already exist. This update allows the LDAP ADD to proceed correctly.

BZ#701559

The server crashed if a long running task was started using the cn=tasks,cn=config interface and then the server was shut down before the task completed. This update prevents the server from crashing, but does not gracefully terminate the task, which can leave the server database in an inconsistent state. For example, the fixup-memberof.pl script invokes a tasks to fix up the memberOf attribute in group member entries. If the server is shut down before the task can complete, some entries may not have the correct memberOf values. Users should ensure that tasks are complete before shutting down the server to avoid inconsistency.

BZ#701560

When using the Entry USN feature, deleting an entry caused a memory leak via the entryusn attribute. This update fixes the memory leak.

BZ#610603

The abrt-gui application used to list plug-ins multiple times if they were configured in the configuration file. This is now fixed.

BZ#627621

In the previous version of ABRT, a daemon restart was required for any changes in the configuration to take effect. In the new version, most of the options in the configuration file no longer require a restart.

BZ#653872

Support for retrace server has been added. Refer to https://fedorahosted.org/abrt/wiki/AbrtRetraceServer for more information about this new feature.

BZ#671354

By default, ABRT stores all problem information in the /var/spool/abrt/ directory. Previously, this path was hard coded and could not be changed in the configuration. With this update, this path can be changed in the /etc/abrt/abrt.conf configuration file.

BZ#671359

The previous documentation failed to cover some customer use cases. This error has been fixed, and all of these use cases are now covered in the Red Hat Enterprise Linux 6 Deployment Guide.

BZ#673173

In ABRT version 1, it was not possible to use wildcards to specify that some action should happen for any user. ABRT version 2 adds support for this functionality.

BZ#695416

The lacking information about configuring a proxy has been added to the Red Hat Enterprise Linux 6 Deployment Guide.

BZ#707950

Previously, a bug in ABRT version 1 was preventing a local Python build to finish. This is now fixed.

BZ#725660

The previous report tool and report library have been obsoleted by abrt and libreport. Users can notice the change in the problem reporting user interface of Anaconda, setroubleshoot, and ABRT.

BZ#720318

Prior to this update, the ACL library did not provide any function to check for extended ACLs of a file without following symbolic links. The only available function, acl_extended_file(), used to cause unnecessary mounts of autofs. This update introduces a new function, acl_extended_file_nofollow(), that checks for extended ACLs of a file without following symbolic links.

BZ#723998

Previously, the ACL library was linked without support for RELRO (read-only relocations) flags. With this update, the library is now linked with partial RELRO support.

BZ#674883

Prior to this update, the setfacl.1 man page was not intelligible in that it did not state that removing a non-existent ACL entry is not considered to be an error. With this update, the setfacl.1 man page has been updated so that its content is now intelligible and exactly specifies the aforementioned behavior with regard to removing a non-existent ACL entry.

BZ#702638

Prior to this update, the package specification did not reflect a change of the upstream project web page address. This update corrects the respective address in the package specification.

BZ#704772

Prior to this update, the alsa output plugin for the Audacious Audio Player did not work correctly. As a result, Audacious could under certain circumstances fail to generate any sound and display error messages. With this update, alsa-lib is modified so that Audacious can now generate sound as expected.

BZ#641861

Issues with "interactive" mode partitioning are fixed.

BZ#731274

The network command is parsed correctly.

BZ#689996

The /boot partition on EFI systems is handled correctly.

BZ#705274

Files that are necessary for libreport and SSL installation mode have been added.

BZ#676404

Symbolic links to LVM commands have been added to the rescue image.

BZ#730650

The /sbin/cio_ignore command is added to initrd.img for IBM System z.

BZ#689029

Support for dracut-style "rdloaddriver=" and "rdblacklist=" parameters is added.

BZ#679108

Support for static addresses in "ipv6=" is added.

BZ#706099

A testing framework for stub commands is added.

BZ#699745

Driver disks support multiple kernel versions and are also built for Red Hat Enterprise Linux 6.0 and 6.1.

BZ#668570

Network connection is brought up before saving a bug report.

BZ#715130

Errors in .treeinfo are detected.

BZ#698282

The xhost authentication is changed when performing live installation.

BZ#664981, BZ#726804

Debugging improvements in loader and package installation code have been made.

BZ#679810

The dialog box focus and initialization have been corrected.

BZ#701220

The iSCSI Login button is disabled when no nodes are selected.

BZ#695362

When a mount point is set to /boot, the file system type is no longer changed.

BZ#728280, BZ#725777, BZ#723194, BZ#723344, BZ#694800, BZ#621175

EDD handling improvements have been made, including Xen and CCISS.

BZ#698429

Extended partitions are handled correctly.

BZ#681803

Handling of "network --device=bootif" is corrected.

BZ#750764

Centering of the Anaconda window when an external display is present is corrected.

BZ#605938

Encrypted device lines written to kickstart files are corrected.

BZ#618535

zFCP multipath devices can be added in the user interface as expected.

BZ#732380

iSCSI discovery that returns no devices is handled correctly.

BZ#704593

Systems with more than 2147483647 kB of memory are handled properly.

BZ#712487

The header image is hidden on all but 800x600 displays.

BZ#690058

The "noprobe" parameter for driver disks is honored.

BZ#713991

The "linksleep=" boot parameter is honored.

BZ#699640

Installation sources (including NFS ISO storages) are mounted correctly.

BZ#679397

Processes in the anaconda process group are killed when the system is shut down.

BZ#693271

Partitioning alignment is corrected.

BZ#616641

Progress indicator improvements for device discovery and command line mode have been made.

BZ#691817, BZ#690748

Kickstart network failures and device name collisions are handled properly.

BZ#691910

The "crashkernel=" parameter in a kickstart file is handled properly.

BZ#712195

Support for the "ext4migrate" parameter has been removed.

BZ#706675

The language and keyboard selection screens are now skipped in stage2 when possible.

BZ#614504

Device capacity values are sorted as numbers, not characters.

BZ#695740

Swap partitions are handled correctly.

BZ#676118

The "--target" option is used in kickstart files for iSCSI devices.

BZ#701371, BZ#696876, BZ#674241, BZ#734374, BZ#729716

Various multipath and raid storage bugs are fixed.

BZ#679073

Anaconda verifies that devices specified with "part" can be partitioned.

BZ#659790

Vendor-provided tools on driver disks are now allowed.

BZ#694198

The initrd.img file is compressed with LZMA.

BZ#696696

The "noverifyssl" boot parameter is added.

BZ#697419

The tboot package is configured when it is installed.

BZ#709653

Multipath device can now be specified using WWID.

Security Fix

CVE-2011-1928

The fix for CVE-2011-0419 (released via RHSA-2011:0507) introduced an infinite loop flaw in the apr_fnmatch() function when the APR_FNM_PATHNAME matching flag was used. A remote attacker could possibly use this flaw to cause a denial of service on an application using the apr_fnmatch() function.

BZ#651119

Prior to this update, the setfattr utility could not restore the original values of the attributes when the "getfattr -e text" or "getfattr --encoding=text" command was used to dump attributes with embedded null characters. This update fixes the encoding of these values in getfattr to prevent information loss.

BZ#665049

Prior to this update, the getfattr utility followed symbolic links to directories even if the "-h" or "--no-dereference" option was specified. Additionally, the description in the getfattr(1) man page that related to this functionality was misleading. This update fixes getfattr with the "-h" option so that it no longer follows the symbolic links and the related content of the getfattr(1) man page is now correct.

BZ#665050

Prior to this update, the getfattr utility did not return a non-zero exit code when an attribute specified in the "getfattr" command did not exist. This update fixes getfattr so that it now returns a non-zero exit code when an attribute does not exist.

BZ#674870

Prior to this update, supported methods for encoding values of the extended attributes were not properly described in the setfattr(1) man page. This update adds the appropriate descriptions of the encoding methods to the setfattr(1) man page.

BZ#702639

Prior to this update, the project web page address as stated in the package specification did not reflect the change of the upstream project web page address. This update corrects the project web page address in the package specification.

BZ#727307

Prior to this update, the attr library was built without support for read-only relocations (RELRO) flags. With this update, the library is now built with partial RELRO support.

BZ#715279

Previously, the audit daemon was logging messages even when configured to ignore "disk full" and "disk error" actions. With this update, audit now does nothing if it is set to ignore these actions, and no messages are logged in the described scenario.

BZ#715315

Previously, the Audit remote logging client received a "disk error" event instead of "disk full" event from a server when the server's disk space ran out. This bug has been fixed and the logging client now returns the correct event in the described scenario.

BZ#748124

Prior to this update, the audit system was identifying the accept4() system call as the now deprecated paccept() system call. Now, the code has been fixed and audit uses the correct identifier for the accept4() system call.

BZ#709345

Previously, the "auditctl -l" command returned 0 even if it failed because of dropped capabilities. This bug has been fixed and a non-zero value is now returned if the operation is not permitted.

BZ#728475

When Kerberos support was disabled, some configuration options in the audisp-remote.conf file related to Kerberos 5 generated warning messages about GSSAPI support during boot. With this update, the options are now commented out in the described scenario and the messages are no longer returned.

BZ#700005

On i386 and IBM System z architectures, the "autrace -r /bin/ls" command returned error messages even though all relevant rules were added correctly. This bug has been fixed and no error messages about sending add rule data requests are now returned in the described scenario.

BZ#693539

Previously, due to a bug in the source code, parsing invalid files failed silently without any error message. With this update, error messages are provided to inform users about the problem.

Bug Fixes

BZ#704935

The autofs utility did not reset the map entry status on a reload request. As a result, newly added map entries that had previously recorded a mount failure failed to work. With this update, autofs resets the map entry status on a reload request and map entries are mounted as expected.

BZ#704939

The autofs utility could have terminated with a segmentation fault when attempting certain mounts. This occurred due to a race condition between mount handling threads for mounts that had previously recorded a mount failure. The automount cache map entry is now verified to be valid.

BZ#704940

The automount(8) man page referred to a non-existent man page. This was caused by a typographical error in the code. With this update, the man page reference has been corrected and the man page is displayed as expected.

BZ#704929

Due to a deadlock, autofs could stop responding when attempting to mount map entries that were nested within maps. With this update, the underlying code has been changed and, where possible, nested map entries mount correctly.

BZ#704933

Prior to this update, automount could terminate unexpectedly with a pthreads error. This occurred because attempts to acquire the master map lock occasionally failed as the lock was held by another thread. With this update, the underlying code has been adapted to wait for a short time before failing.

BZ#704927, BZ#704928

When retrieving paged results from an LDAP (Lightweight Directory Access Protocol) server, autofs handled certain cases incorrectly, which caused the query to not obtain all results. This update adds the code that handles these additional cases.

BZ#704937

Prior to this update, if a key entry of an automount map began with an asterisk (*) sign, the daemon failed with a segmentation fault because the sign was not matched correctly. With this update, such asterisk signs are handled correctly.

BZ#704228

When using GSSAPI authentication, the fact that an incorrect authentication host name was being used caused the connection to fail. This update now gets the correct host name for the connection.

BZ#692816

automount was not performing sufficient sanity checks of server names in its configuration. This update corrects the configuration entry parsing.

BZ#700136

Error reporting for invalid mount locations was unclear. This update improves the error reporting.

BZ#703332

When an automount map key is present in a file map and is also present in an included map source, if the file map entry was removed and a lookup performed before a re-load was issued, the map lookup would have failed. This update corrects the logic used to determine if the lookup needs to continue into included maps.

BZ#718927

When reloading maps that include a combination of direct and indirect maps, it was possible for automount to deadlock due to incorrect lock ordering.

BZ#725931

There was inadvertent use of a small amount of GPLv3-licensed code from Samba in autofs. While this was permissible, it would have entailed explicitly relicensing autofs from "GPLv2 or later" to "GPLv3", which is not intended for autofs at this time. Therefore, the Samba-derived code has been replaced in order to maintain the "GPLv2 or later" licensing status of autofs.

Enhancements

BZ#704416

This update adds the "--dumpmaps" option to the automount command, which allows you to dump the maps from their source as seen by the automount daemon.

BZ#704932

This update adds simple Base64 encoding for LDAP and thus allows hashing of the password entries in the /etc/autofs_ldap_auth.conf configuration file.

Bug Fix

BZ#658057

When installing autotrace-devel multilib RPM packages from the optional repository, file conflicts between these packages appeared, causing the installation transaction to abort. This problem has been fixed and the installation transaction now proceeds without conflicts.

BZ#651776

Prior to this update, the bacula packages were not distributed with the applybaculadate file. As a result, the logwatch cron script failed. The problem has been fixed by including the applybaculadate file in the bacula packages so that the logwatch cron script now works as expected.

BZ#651780

Prior to this update, the make_catalog_backup.pl script created a MySQL configuration file, which had the file permissions set to world-writeable and world-readable so that MySQL did not accept the configuration file with these permissions and the MySQL database login configuration was not used. As a result, it was not possible to complete a MySQL database dump. With this update, the configuration file is now created with correct permissions, and the MySQL database login configuration is used by MySQL so that it is now possible to complete the MySQL database dump as expected.

BZ#651786

Prior to this update, there was no option to change Bacula's runtime user. As a consequence, Bacula was always run under the root user. The problem has been fixed by adding support for the bacula-dir, bacula-fd, and bacula-sd files in the /etc/sysconfig/ directory; these files can be used for specifying a non-root user and group with the DIR_USER, FD_USER, SD_USER, and DIR_GROUP, FD_GROUP and SD_GROUP options, respectively. With this update, Bacula can be run under the specified user.

BZ#651787

Prior to this update, when creating a symbolic link to the "bscan" utility, the new link was erroneously named "dbcheck". As a result, the already existing "dbcheck" symbolic link was overwritten by the erroneous one. Thus the "dbcheck" command ran the "bscan" utility so that it was not possible to execute the "bscan" utility with the "bscan" command. The problem has been fixed in this update so that the "dbcheck" and "bscan" utilities now work as expected.

BZ#657297

Prior to this update, Bacula's default configuration missed a required option. As a result, the Bacula tray monitor component terminated unexpectedly. The problem has been fixed by adding the "Address" option to the "Director" section in the Bacula tray monitor configuration file so that the Bacula tray monitor now works as expected with the default configuration file. Note that this bug fix does not alter any existing Bacula tray monitor configuration file. As a consequence, the Bacula tray monitor can terminate unexpectedly if the existing Bacula tray monitor configuration is incorrect.

BZ#689400

Prior to this update, the backup size was computed incorrectly under certain circumstances. As a consequence, the reported size of the incremental backup could have been wrong. The problem has been fixed by correcting the backup size computation process so that the size of the incremental backup is now reported correctly.

BZ#712794

Prior to this update, the shadow-utils package was not listed among the package dependencies for Bacula. As a result, the bacula user and bacula group were not created when the shadow-utils package was not present on the system, and a warning message was displayed during the bacula packages installation. This bug has been fixed by adding shadow-utils to the package dependencies.

BZ#712804

Prior to this update, the chkconfig package, which contains the "alternatives" utility, was not listed among the package dependencies for Bacula. As a result, the bacula-dir and bacula-sd services were not configured, the "alternatives" utility was not found, and Bacula's symbolic links were not created. These problems have been fixed by adding chkconfig to the package dependencies.

Security Fix

CVE-2011-1910

An off-by-one flaw was found in the way BIND processed negative responses with large resource record sets (RRSets). An attacker able to send recursive queries to a BIND server that is configured as a caching resolver could use this flaw to cause named to exit with an assertion failure.

Security Fix

CVE-2011-2464

A flaw was discovered in the way BIND handled certain DNS requests. A remote attacker could use this flaw to send a specially-crafted DNS request packet to BIND, causing it to exit unexpectedly due to a failed assertion.

Security Fix

CVE-2011-4313

A flaw was discovered in the way BIND handled certain DNS queries, which caused it to cache an invalid record. A remote attacker could use this flaw to send repeated queries for this invalid record, causing the resolvers to exit unexpectedly due to a failed assertion.

BZ#699951

Prior to this update, the code in libdns which sends DNS requests was not robust enough and suffered from a race condition. If a race condition occurred, the "named" name service daemon logged an error message in the format "zone xxx.xxx.xxx.in-addr.arpa/IN: refresh: failure trying master xxx.xxx.xxx.xxx#53 (source xxx.xxx.xxx.xxx#0): operation canceled" even when zone refresh was successful. This update improves the code to prevent a race condition in libdns and the error no longer occurs in the scenario described.

BZ#700097

A command or script traditionally gives a non-zero exit status to indicate an error. Prior to this update, the nsupdate utility incorrectly returned the exit status "0" (zero) when the target DNS zone did not exist. Consequently, the nsupdate command returned "success" even though the update failed. This update corrects this error and nsupdate now returns the exit status "2" in the scenario described.

BZ#725577

Prior to this update, named did not unload the bind-dyndb-ldap plugin in the correct places in the code. Consequently, named sometimes terminated unexpectedly during reload or stop when the bind-dyndb-ldap plugin was used. This update corrects the code, the plug-in is now unloaded in the correct places, and named no longer crashes in the scenario described.

BZ#693982

A non-writable working directory is a long time feature on all Red Hat systems. Previously, named wrote "the working directory is not writable" as an error to the system log. This update changes the code so that named now writes this information only into the debug log.

BZ#717468

The named initscript lacked the "configtest" option that was available in earlier releases. Consequently, users of the bind initscript could not use the "service named configtest" command. This update adds the option and users can now test their DNS configurations for correct syntax using the "service named configtest" command.

BZ#742368

Previously, the bind-dyndb-ldap plug-in could faile to honor the selected authentication method because it did not call the ldap_bind() function on reconnection. Consequently, the plug-in connected to the LDAP server anonymously. With this update, the ldap_bind() function is executed on reconnection and the plug-in uses the correct authentication method in the described scenario.

BZ#707255

The bind-dyndb-ldap plug-in failed to load new zones from the LDAP server runtime. This update adds the zone_refresh parameter to the plug-in which controls how often the zone check is performed.

BZ#745045

The bind-dyndb-ldap plug-in could fail to connect to the LDAP server. This happened when the LDAP server was using localhost and FreeIPA installation was using a name different from the machine hostname. This update adds to the plug-in the ldap_hostname option, which can be used to set the correct LDAP server hostname.

BZ#727856

The "named" process could have remained unresponsive due to a race condition in the bind-dyndb-ldap plug-in. With this update, the race condition has been resolved and the problem no longer occurs.

Bug Fixes

BZ#664640

Prior to this update, the readelf utility added 0x40 into a character in order to display a non-printing character but did not do so when processing a multibyte character. As a result, the readelf utility did not display a multibyte character in the ELF header correctly. The code has been corrected and readelf no longer displays garbled characters when processing multibyte, or non-ASCII, characters.

BZ#674925

An unneeded patch to bineutils caused a large link time degradation when using the binutils --build-id command. This update removes that patch.

BZ#689829

An Operating System (OS) Application Binary Interface (ABI) describes the low-level interface between a program and the operating system (OS/ABI). The indirect meta-function, ifunc(), whose value can be determined at load time, allows for architecture dependent optimization. Prior to this update, the OS/ABI preprocessor macro was erroneously set to UNIX - Linux instead of UNIX - System V in an ELF header by a dynamic executable which used ifunc(). This update applies a backported patch which corrects the code and the error no longer occurs.

BZ#698005

Prior to this update, the binutils' strip command, which is run as part of the RPM build process, did not copy the EI_OSABI value in the ELF file header properly, it set the value to zero. Consequently, if the EI_OSABI field of the debug file had a value of 3 (ABI tag for GNU/Linux), in the stripped file it was erroneously set to 0 (UNIX - System V). This update corrects the problem and strip now leaves the field intact.

BZ#701586

On 64-bit PowerPC platforms, the position of -ldl in the list compiler options caused unexpected behavior when compiling C++ code. If -ldl was not placed at the end of parameter list, the GNU C Compiler (GCC) failed with an error in the format:

libtest.a(some_object_file.o): undefined reference to `.dlerror'

With this update, the code has been corrected and the GCC compiler functions as expected.

BZ#707387

When compiling C source code using the GNU C Compiler (GCC), a Table Of Contents (TOC), is created for every executable file. Prior to this update, compiling C++ code using GCC for 64-bit PowerPC, using -mcmodel=small -mno-minimal-toc as options, GNU linker, (ld), erroneously decided that if a section did not make use of the TOC it could belong to any TOC group. Consequently, when a local function call was made from one section of code to another section in the same object file, due to the two sections being assigned to different TOC groups, a failure occurred and an error message in the following format was logged.

libbackend.a(cse.o)(.text.unlikely+0x60): sibling call optimization to `.opd' does not allow automatic multiple TOCs; recompile with -mminimal-toc or -fno-optimize-sibling-calls, or make `.opd' extern

This update applies an upstream patch to improve the partitioning of sections of code, which make local function calls, into multiple TOC groups. As a result the error no longer occurs in the scenario described.

Note

It is necessary to relink executables and shared libraries containing objects which were compiled with -mcmodel=small -mno-minimal-toc. Therefore code should be recompiled by running these commands again after applying the update.

BZ#714824

Prior to this update, after compiling a kernel from source code with debugging information, some debug information was missing. Consequently, when using the GNU Project's debugger (GDB) utility, if a user issued the command l setup_arch to determine the target architecture, the following error was displayed.

No line number known for setup_arch

This update corrects the code and the GDB utility now correctly displays the architecture for which the code was compiled.

BZ#721079

Compilers used for producing code optimized for 64-bit PowerPC platforms use the default Red Hat Enterprise Linux system linker, ld, provided with the operating system to produce executables and libraries. Some object code generated by the IBM XL compiler caused ld to terminate unexpectedly with a segmentation fault. Consequently, users were not able to produce optimized executables or libraries. With this update, a backported patch has been applied to correct the problem and ld no longer crashes in the scenario described.

BZ#733122

When linking FORTRAN programs with the IBM XL compiler and the default Red Hat Enterprise Linux 6.1 system linker, ld sometimes terminated unexpectedly with a segmentation fault. This updates applies an upstream patch to correct the problem and ld no longer crashes in the scenario described.

BZ#747695

The assembler, as, when generating a memory reference to a local symbol plus or minus an offset, did not include the constant offset when generating 32-bit x86 code. Consequently, when the local symbol being referenced was defined before the instruction using the symbol with an offset, an error would occur. This update corrects the code and the problem no longer occurs.

Enhancements

BZ#696368

With this update, backported patches have been included to support new AMD processors.

BZ#696494

Certain Intel processors support a new RdRand instruction to generate a true random number in a short time. This update includes support for this new instruction.

BZ#721079

Prior to this update, an input object file could have a non-empty .toc section but no references to the .toc entries because of a problem in the 64-bit PowerPC linker TOC editing code. As a result, various utilities of the binutils package terminated unexpectedly with a segmentation fault under certain conditions. This update handles local symbols in .toc sections correctly. Now, no more crashes occur.

BZ#700248

When NPAR (NIC Partitioning) is enabled, the partition number should be appended as a suffix to the interface name. Previously, biosdevname did not add partition numbers to interface names, for example, instead of naming an interface "em3_1", the interface was named "em3". Consequently, partitioned network interfaces were missing the suffix necessary to describe the partition. Now, biosdevname correctly recognizes the VPD (Vital Product Data) suffix and full interface names are created correctly.

BZ#700251

When biosdevname ran in a guest environment, it suggested names to new network interfaces as if it was in a host environment. Consequently, affected network interfaces were incorrectly renamed. Now, biosdevname no longer suggests names in the described scenario.

BZ#729591

When biosdevname was reading VPD information to retrieve NPAR-related data, the read operations failed or became unresponsive on certain RAID controllers. Additionally, biosdevname sometimes attempted to read beyond the VPD boundary in the sysfs VPD file, which also resulted in a hang. This bug has been fixed and biosdevname now performs the read operation correctly in the described scenarios.

BZ#739592

Previously, the "--smbios" and "--nopirq" command-line parameters were missing in the biosdevname binary. Consequently, consistent network device naming could not be enabled because biosdevname exited without suggesting a name. This update adds support for these parameters and enables the device naming.

BZ#740532

Previously, NICs (Network Interface Cards) on biosdevname-compatible machines were given traditional "eth*" names instead of "em*" or "p*p*" names. This bug has been fixed and biosdevname now provides correct names for the NICs.

BZ#696252

With this update, "--smbios" and "--nopirq" command-line parameters have been added to biosdevname.

BZ#736442

The biosdevname man page has been updated to explain the functionality of the "--smbios" and "--nopirq" command-line parameters.

BZ#705128

Prior to this update, the blkparse code contained a misprint. As a result, blkparse used the wrong variable when printing the PC Writes Completed. This update modifies the code so that blkparse now prints the correct value for PC Writes Completed.

BZ#736399

This update adds FLUSH/FUA support to blktrace.

BZ#618308

Prior to this update, the bltk tree was corrupted. As a result, the bltk_report script failed. This update modifies the settings of the bltk root path. Now, the report script works as expected.

BZ#679028

Prior to this update, bltk could be installed without requiring the gnuplot binary. As a result, the bltk_plot script exited with an error message when the gnuplot package was not installed and the charts were shown from measured data. This update requires the gnuplot package for its installation. Now, the bltk_plot script no longer exits with an error.

BZ#723503

Prior to this update, the cyclic redundancy check (CRC) was not correctly computed on 64-bit architectures during decompression of gzip archives. In this update, constant-width integer types are used to compute CRC to make the results stable across all architectures.

Security Fix

BZ#734381

It was found that a Certificate Authority (CA) issued fraudulent HTTPS certificates. This update removes that CA's root certificate from the ca-certificates package, rendering any HTTPS certificates signed by that CA as untrusted.

BZ#660347

Prior to this update, cachefilesd used the wrong log level for cull info messages. As a result, the /var/log/messages file could become overloaded. This update reduces the messages to the debug level. Now, /var/log/messages no longer becomes overloaded.

BZ#723890

Prior to this update, cachefilesd depended on a specific version of the SELinux policy package. As a result, only the nominated version was allowed. This update permits the nominated version and any later versions. Now, the SELinux policy dependency works as expected.

BZ#692766

Previously, the certmonger service could access a Network Security Services (NSS) database without a password, despite being configured to use a password to access that database. This behavior was not recognized as an error. This update correctly diagnoses this inconsistency as an error.

BZ#694184

Previously, if the certmonger service could not generate a key pair in an NSS database because it did not have the password that was required for accessing the database, the certmonger service did not recover when it was subsequently given the correct password. This update handles this case correctly.

BZ#697058

Previously, the certmonger service did not correctly diagnose a missing token if the name of the token to use was specified when the service was instructed to generate a key pair for storage in an NSS database. This update corrects this error.

BZ#712500

Previously, the certmonger service encountered an assertion failure if the D-Bus message bus service was not already running when certmonger was started. This update modifies the certmonger service so that no more assertion problems occur in such a situation.

BZ#721392

Previously, when the getcert command needed to report an error message which it received from the certmonger service, it exited unexpectedly due to a logic error. This update corrects the logic so that the error message is correctly reported.

BZ#727863

Previously, the certmonger service was not fully compatible with newer versions of the xmlrpc-c and libcurl packages. As a result, credentials could not be delegated when using GSSAPI authentication with a CA that was accessed via XML-RPC. This update includes the necessary changes to continue to be able to delegate credentials when using GSSAPI authentication with a CA that is accessed using XML-RPC, such as IPA.

BZ#699059, BZ#739903

Previously, when the getcert request command was given a location for key or certificate storage using a relative path, and the location did not exist, the error was only reported after multiple warnings during which the command attempted to convert the relative path to an absolute path. This update suppresses these warnings.

BZ#741262

Previously, an incorrect error message was displayed if the getcert resubmit command was invoked with the -i flag to specify which request should be resubmitted to a CA but no request that matched the provided value was present. This update displays the correct error message.

BZ#742348

Due to a logic error, attempts to save a newly-obtained certificate to an NSS database could fail intermittently. This update corrects the error.

BZ#698772

Previously, the getcert list command only printed information about every certificate and enrollment request being managed by certmonger, and there was no way to narrow down the results. This update includes an updated version of the command which can narrow the result set if the invoking user provides information about the location of the certificate or key in which the user is interested

BZ#750617

This update now includes an HTTP "Referer:" header value when submitting requests to CAs which are accessed using XML-RPC, as is expected to be required by future releases of the IPA CA.

BZ#729803

When submitting a signing request to a Red Hat IPA (Identity, Policy, Audit) CA, certmonger is expected to authenticate using the client's host credentials, and to delegate the client's credentials to the server. Recent updates to libraries on which certmonger depends changed delegation of client credentials from a mandatory operation to an optional operation that is no longer enabled by default, which effectively broke certmonger's support for IPA CAs. This update gives certmonger the ability to explicitly request credential delegation when used with newer versions of these libraries, which introduce an API that allows certmonger to explicitly request that credential delegation be performed.

BZ#676439

Prior to this update, mount.cifs dropped the CAP_DAC_READ_SEARCH flag together with most of the other capability flags before it performed a mount. As a result, mounting onto a directory without execute permissions failed if mount.cifs was installed as a setuid program and the user mount was configured in the /etc/fstab file. This update reinstates the CAP_DAC_READ_SEARCH flag before calling mount. Now, mounting no longer fails.

BZ#719363

Prior to this update, several mount options were missing from the mount.cifs(8) man page. With this update, the man page documents all mount options.

BZ#682650

Prior to this update, when viewing the U+4190 CJK character with the AR PL UMing font and the font size 10, this character was not displayed properly. This bug has been corrected in this update so that the character is now correctly displayed as expected.

Bug Fixes

BZ#707115

The cluster and gfs2-utils packages have been upgraded to upstream version 3.0.12.1, which provides a number of bug fixes over the previous version.

BZ#713977

Previously, when a custom multicast address was configured, the configuration parser incorrectly set the default value of the time-to-live (TTL) variable for multicast packet to 0. As a consequence, cluster nodes were not able to communicate with each other. With this update, the default TTL value is set to 1, which fixes the problem.

BZ#726777

A section describing the "suborg" option for the fence_cisco_usc agent was not present in the RELAX NG schema which is used to validate the cluster.conf file. As a consequence, validation of cluster.conf failed even if the file was valid. The suborg section has been added to the RELAX NG schema and cluster.conf is now validated correctly.

BZ#707091

Building the resource group index for a new GFS2 file system using the mkfs.gfs2 utility used all the space allocated. If the file system filled up completely, no room was left to write a new rindex entry. As a consequence, the gfs2_grow utility was unable to expand the file system. The mkfs.gfs2 utility has been modified so that enough space is now allocated for the entire rindex file, and one extra rindex entry. The gfs2_grow source code has been modified to utilize the unused rindex space. As a result, gfs2_grow is now able to expand a completely full GFS2 file system.

BZ#678585

GFS2 POSIX (Portable Operating System Interface) lock operations (implemented in Distributed Lock Manager, also known as DLM) are not interruptible when they wait for another POSIX lock. Previously, processes that created a deadlock with POSIX locks could not be killed to resolve the problem, and one node had to be reset. DLM now uses a new kernel feature that allows the waiting process to be killed, and information about the killed process is now passed to the dlm_controld daemon to be cleaned up. Processes deadlocked on GFS2 POSIX locks can now be recovered by killing one or more of them.

BZ#719135

Prior to this update, boundaries for the locktable and label fields in the GFS2 superblock were not properly checked by the tunegfs2 tool. As a consequence, running the "gfs2_tool sb" command could terminate unexpectedly with buffer overflow. In addition, invalid characters could be printed when using tunegfs2 to change locktable or label to a minimum or maximum length (63 characters). The tunegfs tool has been modified to check the correct boundaries of the locktable and label fields. As a result, tunegfs2 no longer creates invalid locktables or labels, and therefore gfs2_tool prints the superblock values properly.

BZ#740385

When executing the cman utility by using the init script with enabled debugging, a file descriptor leaked. The file pointed to the file descriptor would continue to grow endlessly, filling up the /tmp file system. This update ensures that the file descriptor is closed after a successful cman startup. Space in /tmp is now released correctly.

BZ#695795

The cman utility implements a complex set of checks to configure the Totem protocol. One of the checks that copies the configuration data was incorrect and the transport protocol option was not handled correctly as a consequence. A patch has been applied to address this issue and cman now handles the transport option properly.

BZ#679566

When the user executed the "gfs2_edit savemeta" command to save the metadata for a target GFS2 file system, not all of the directory information was saved for large directories. If the metadata was restored to another device, the fsck.gfs2 tool found directory corruption because of a missing leaf block. This was due to gfs2_edit treating the directory leaf index (also known as the directory hash table) like a normal data file. With this update, gfs2_edit's savemeta function is modified to actually read all the data (the directory hash table) for large directories and traverse the hash table, saving all the leaf blocks. Now, all leaf blocks are saved properly.

BZ#679080

When the fsck.gfs2 tool was resolving block references and no valid reference was found, the reference list became empty. As a consequence, fsck.gfs2 check in pass1b terminated unexpectedly with a segmentation fault. With this update, pass1b is modified to check that the list is empty. The segmentation fault no longer occurs and fsck.gfs2 proceeds as expected.

BZ#731775

The dlm_controld daemon passed error results back to the kernel for POSIX unlock operations flagged with CLOSE. As a consequence, the kernel displayed the "dlm: dev_write no op" error messages, most of them when using non-POSIX locks, flocks. The dlm_controld daemon has been fixed to not pass error results to the kernel for POSIX unlock operations flagged with CLOSE. As a result, error messages no longer appear.

BZ#729071

Previously, the mount.gfs2 utility passed the "loop" option to the GFS2 kernel module which treated it as an invalid option. Mounting a GFS2 file system on loopback devices failed with an "Invalid argument" error message. With this update, mount.gfs2 is modified to avoid passing the "loop" option to the kernel. Mounting GFS2 systems on loopback devices now works as expected.

BZ#728230

Missing sanity checks related to the length of a cluster name caused the cman utility to fail to start. The correct sanity checks have been implemented with this update. The cman utility starts successfully and informs the user of the incorrect value of the cluster name, if necessary.

BZ#726065

The XML format requires special handling of certain special characters. Handling of these characters was not implemented correctly, which caused the cluster.conf file to not function as expected. Correct handling of the characters has been implemented and cluster.conf now works as expected.

BZ#706141

The exact device/mount paths were not compared due to incorrect logic in mount.gfs2 when trying to find mtab entries for deletion. The original entry was not found during remounts and therefore was not deleted. This resulted in double mtab entries. With this update, the realpath() function is used on the device/mount paths so that they match the content of mtab. As a result, the correct original mtab entry is deleted during a remount, and a replacement entry with the new mount options is inserted in its place.

BZ#720668

Previously, mkfs.gfs2 treated normal files incorrectly as if they were block devices. Attempting to create a GFS2 file system on a normal file caused mkfs.gfs2 to fail with a "not a block device" error message. Additional checks have been added so that mkfs.gfs2 does not call functions specific for block devices on normal files. GFS2 file systems can now be created on normal files. However, use of GFS2 in such cases is not recommended.

BZ#719126

The tunegfs2 command line usage message was not updated to reflect the available arguments which are documented in the man page. As a consequence, tunegfs2 printed an inaccurate usage message. The usage message has been updated and tunegfs2 now prints an accurate message.

BZ#719124

Previously, certain argument validation functions did not return error values, and tunegfs2 therefore printed confusing error messages instead of exiting quietly. Error handling has been improved in these validation functions, and tunegfs2 now exits quietly instead of printing the confusing messages.

BZ#694823

Previously, the gfs2_tool command printed the UUID (Universally Unique Identifier) output in uppercase. Certain applications expecting the output being in lowercase (such as mount) could have malfunctioned as a consequence. With this update, gfs2_tool is modified to print UUIDs in lowercase so that they are in a commonly accepted format.

BZ#735917

The qdisk daemon did not allow cman to upgrade the quorum disk device name. The quorum disk device name was not updated when the device was changed and, in very rare cases, the number of qdiskd votes would therefore not be correct. A new quorum API call has been implemented to update the name and votes of a quorum device. As a result, quorum disk device names and votes are updated consistently and faster than before.

BZ#683104

Prior to this update, the fsck.gfs2 utility used the number of entries in the journal index to look for missing journals. As a consequence, if more than one journal was missing, not all journals were rebuilt and subsequent runs of fsck.gfs2 were needed to recover all the journals. Each node needs its own journal; fsck.gfs2 has therefore been modified to use the "per_node" system directory to determine the correct number of journals to repair. As a result, fsck.gfs2 now repairs all the journals in one run.

BZ#663397

Previously, token timeout intervals of corosync were larger than the time it took a failed node to rejoin the cluster. Consequently, corosync did not detect that a node had failed until it rejoined. The failed node had been added again before the dlm_controld daemon asked corosync for the new member list, but dlm_controld did not notice this change. This eventually caused the DLM (Distributed Lock Manager) lockspace operations to get stuck. With this update, dlm_controld can notice that a node was removed and added between checks by looking for a changed incarnation number. Now, dlm_controld can properly handle nodes that are quickly removed and added again during large token timeouts.

BZ#732991

Previously, if a cluster was configured with a redundant corosync ring, the dlm_controld daemon would log harmless EEXIST errors, "mkdir failed: 17". This update removes these error messages so that they no longer appear.

Enhancements

BZ#733345

The corosync IPC port allows, when configured correctly, non-privileged users to access corosync services. Prior to this update, the cman utility did not handle such connections correctly. As a consequence, users were not able to configure unprivileged access to corosync when it was executed using cman. This update adds support to cman to configure unprivileged access. As a result, configured users and groups can now access corosync services without root privileges.

BZ#680930

This update introduces dynamic schema generation, which provides a lot of flexibility for end users. Users can plug into Red Hat Enterprise Linux High Availability Add-On custom resource and fence agents, and still retain the possibility to validate their cluster.conf file against those agents.

BZ#732635, BZ#735912

This update adds support for Redundant Ring Protocol, which aligns the default configuration of cman with corosync. Note that this enhancement is included as a Technology Preview.

BZ#702313

Previously, gfs2_edit saved GFS2 metadata uncompressed. Saved GFS2 metadata sets could have filled up a lot of storage space, and transferring them (for example, for support and debugging) would be slow. This update adds gzip compression to the metadata saving and restoring functions of gfs2_edit. GFS2 metadata sets are now compressed when saving and decompressed when restoring them. The user can specify the compression level with a command line option.

BZ#704178

With this update, the tunegfs2 utility replaces the superblock manipulating feature of gfs2_tool.

BZ#673575

Previously, the fence_scsi agent did not reboot a node when it was fenced. As a consequence, the node had to be rebooted manually before rejoining the cluster. This update provides a script for detecting loss of SCSI reservations. This can be used in conjunction with the watchdog package in order to reboot a failed host.

BZ#728247

Prior to this update, the "suborg" option was not allowed by the cluster configuration schema defined in the /usr/share/cluster/cluster.rng file. As a consequence, when the "suborg" option was specified for the fence_cisco_ucs agent, the cluster refused to validate the configuration schema. The "suborg" option is now properly recognized, which fixes the problem.

BZ#720100

Previously, when a custom multicast address was configured, the configuration parser incorrectly set the default value of the time-to-live (TTL) variable for multicast packet to 0. Consequently, cluster nodes could not communicate with each other. With this update, the default TTL value is set to 1, thus fixing this bug.

BZ#634373

Previously, the clustermon tool failed to shut down nodes if the user had mounted a GFS2 file system that was not listed in the /etc/fstab file. This was caused by clustermon relying on the rgmanager tool and the GFS2 init scripts to unmount all file systems, but the cluster stack would not stop properly if the user mounted the file system manually. This has been fixed: clustermon now ensures that there are no cluster file systems mounted and then attempts to stop the cluster stack.

BZ#724978

The "get_cluster_schema" function call has been added to allow users to easily get the XML cluster schema content.

BZ#578690

This update adds support for Personal Identity Verification (PIV) smart cards.

BZ#700907

Common Access Cards (CAC) are defined to have exactly three certificates. However, some cards that used the CAC interface supplied one or two certificates only, which may have caused the coolkey utility to fail. CAC smart cards that contain less than three certificates are now supported.

BZ#691292

Prior to this update, SELinux appeared to be disabled when building coreutils in Mock. As a result, coreutils did not build. With this update, SELinux determines more precisely whether it is disabled or not. Now, the packages are built successfully.

BZ#703712

Previously, incorrect signal handling could cause various problems for tcsh users logging into the root shell using the su utility. Signal masking in the subshell called by the su utility has been modified to respect the SIGTSTP signal as well as the SIGSTOP signal.

BZ#715557

When using the "-Z/--context" option in the cp utility, the SELinux context of a file was not changed if the file destination already existed. The utility has been modified and the context is changed as expected. However, this option is not portable to other systems.

BZ#720325

Prior to this update, the acl_extended_file() function could cause unnecessary mounts of autofs when using the ls command on a directory with autofs mounted. This update adds the new acl function, acl_extended_file_nofollow(), to prevent unnecessary autofs mounts.

BZ#725618

The description of the "--sleep-interval" option in the tail(1) manual page has been improved to be clearer about the behavior and to match the upstream version of coreutils.

Bug Fixes

BZ#677583

Prior to this update, the corosync-blackbox command could, under certain circumstances, produce a backtrace in the output and consequently terminate with a segmentation fault. With this update, Corosync creates correct fdata files and also corosync-fplay is more resistant when dealing with incorrect fdata files.

BZ#677583

Prior to this update, cpg did not use the "left_nodes" field in the downlist message. As a consequence, a node could miss a configuration change and report larger old_members than expected if one node was paused. This update modifies the downlist so that the "left_nodes" field is used. Now, the membership events are correct.

BZ#692620

Prior to this update, cpg did not use the "left_nodes" field in the downlist message. As a consequence, a node could miss a configuration change and report larger old_members than expected if one node was paused. This update modifies the downlist so that the "left_nodes" field is used. Now, the membership events are correct.

BZ#696883

Prior to this update, running Corosync could cause a segmentation fault on multiple nodes when executed via CMAN. This update modifies the code so that executing Corosync via CMAN no longer causes segmentation faults with the pacemaker test suite.

BZ#696887

Prior to this update, the reference counting on the configuration server in Corosync was incorrect. As a consequence, terminating the corosync-cfgtool -r command before completing caused a segmentation fault. This update adds the correct reference counting for each architecture. Now, Corosync no longer encounters segmentation faults in this situation.

BZ#707860

Prior to this update, Corosync could terminate with a segmentation fault if it ran out of available open files. This update handles the maximum number of open files more gracefully. Now, Corosync no longer crashes when going over open file limits.

BZ#707867

Prior to this update, corosync-objctl could not create a new object/key and display double or float values. This update adds float and double support to corosync-objctl. Now, corosync-objctl can display object values with double or float types.

BZ#707873

Prior to this update, Corosync could terminate with a segmentation fault if it encountered a negative value for the message type on systems where char is signed. This update improves the check of the message type for incoming messages.

BZ#707875

Prior to this update, an error message was wrongly displayed if files in the service.d directory differed from the service key. With this update, Corosync longer checks for sub parameters in files in the service.d directory. Now, files in service.d directory can contain every possible configuration option.

BZ#709758

Prior to this update, Corosync used a spinlock around I/O operations. As a consequence, Corosync consumed an extremely high portion of the central processing unit (CPU) when running a large amount of inter-process communication (IPC) operations because the spinlocks would spin during I/O. This update replaces the spinlock with a mutual exclusion (mutex), which releases the processor from spinning but enforces correct behavior.

BZ#712115

Prior to this update, an incorrect mutex in the internal confdb data storage system could, under certain circumstances, cause Corosync to terminate with a segmentation fault. This update corrects the mutex and objdb API iteration no longer causes Corosync to terminate with a segmentation fault.

BZ#712188

Prior to this update, Corosync became locked with contrived test cases when the tracking functionality of the internal object database was enabled if it was under heavy load. This update modifies Corosync so that the tracking functionality under heavy load no longer causes Corosync to lock up.

BZ#725058

Prior to this update, retransmit list errors could occur on slower hardware due to high multicast traffic and slow CPU usage. This update processes the multicast buffer queue more frequently and retransmit errors are now less probable.

BZ#732698

Prior to this update, Corosync sometimes terminated unexpectedly when Corosync ran the cman_tool join and cman_tool leave commands in a loop. This update modifies the code so that no more segmentation faults occur in such situations.

Enhancements

BZ#529136

Prior to this update, the protocol in Corosync unnecessarily copied memory on AMD64 and EM64T architectures to align data structures for architectures which do not handle alignment correctly. As a consequence, the utilization of the central processing unit (CPU) was increased. This update can conditionally avoid copies on unaligned safe architectures such as Intel 80386, AMD64, and EM64T architectures. Now the CPU utilization is reduced by around 20%.

BZ#599327

Prior to this update, no diagnostic message was available when the multicast was blocked. As a consequence, each partition lost quorum which never remerged. This update displays a diagnostic warning that the node can not exit the GATHER state when a local NIC (network interface card) fault occurs or the firewall prevents totem from forming a cluster. In addition, the runtime.totem.pg.mrp.srp.firewall_enabled_or_nic_failure key is now set to 1.

BZ#667652

Prior to this update, fenced nodes where not safely powered up due to issues with the boot sequence. As a consequence, users had to skip cluster services at boot to avoid problems such as long response times and fences in two-node clusters. With this update, setting the nocluster boot parameter prevents Corosync to start automatically.

BZ#688260

Prior to this update, configuring two rings with different IP subnets only duplicated the IP address data of one ring. This update adds support for the redundant ring functionality to Corosync as a Technology Preview.

BZ#707876

Prior to this update, the corosync init script did not depend on syslog. As a consequence, syslog logging did not work if the user turned off syslog. This update adds syslog as a dependency to the init script. Now, logging works in all cases.

BZ#722469

Prior to this update, configuring two rings with different IP subnets only duplicated the IP address data of one ring. This update adds support for the redundant ring functionality to Corosync as a Technology Preview.

BZ#726608

Previously, under heavy traffic, receive buffers sometimes overflowed, causing loss of packets. Consequently, retransmit list error messages appeared in the log files. This bug has been fixed, incoming messages are now processed more frequently, and the retransmit list error messages no longer appear in the described scenario.

BZ#727962

Previously, when a combination of a lossy network and a large number of configuration changes was used with corosync, corosync sometimes terminated unexpectedly. This bug has been fixed, and corosync no longer crashes in the described scenario.

BZ#734997

Prior to this update, when corosync ran the "cman_tool join" and "cman_tool leave" commands in a loop, corosync sometimes terminated unexpectedly. This bug has been fixed, and corosync no longer crashes in the described scenario.

BZ#675734

Prior to this update, the cpufreq-aperf utility did not run on 32-bit systems due to an incorrect argument passed to the read() call. This problem has been fixed: the buffer size is now used instead of the size of the pointer and the cpufreq-aperf utility runs as expected.

BZ#710193

The crash package has been upgraded to upstream version 5.1.8, which provides a number of enhancements and bug fixes over the previous version.

BZ#705142

Previously, compressed kdump dump files were handled incorrectly on AMD64 and Intel 64 architectures if a system contained more than 454 CPUs. In such a case, the crash session terminated during initialization with the "crash: compressed kdump: invalid nr_cpus value: [cpus]" error message. A patch has been provided to address this issue, and the compressed dump files are now handled properly, thus fixing this bug.

BZ#716931

When the first chunk of physical memory on a system was assigned to NUMA (Non-Uniform Memory Architecture) node 1 (typically it is assigned to NUMA node 0), the "kmem -s" or "kmem -S" command incorrectly showed all cache blocks allocated by the slab allocator as empty. This bug has been fixed, and the kmem command now shows populated kmem_cache slab data correctly.

BZ#712214

In a rare scenario, a non-crashing CPU received a shutdown NMI (non-maskable interrupt) immediately after receiving an interrupt from another source. Because the IRQ entry-point symbols "IRQ0x00_interrupt" through "IRQ0x##_interrupt" no longer existed, the bt command terminated with the "bt: cannot transition from exception stack to current process stack" error message on AMD64 and Intel 64 architectures. This bug has been fixed, and backtrace now properly transitions from the NMI stack back to the interrupted process stack.

BZ#695413

The crash.8 man page and the associated built-in "crash -h" output have been re-written. The crash.8 man page now clarifies the required invocation options, adds all of the rarely-used command line options that have proliferated over the years, and updates the ENVIRONMENT variables section. The "crash -h" output now closely mimics the relevant parts of the crash.8 man page.

BZ#703467

With this update, the new "--osrelease [dump_file]" command line option that displays the OSRELEASE vmcoreinfo string from a kdump dump file has been added.

BZ#609544

Prior to this update, an example included in the /etc/crontab file contained an omission. It did not state that defining a job in crontab requires a username to be defined. The missing information has been added to the /etc/crontab file in this update.

BZ#713410

When the cryptsetup or libcryptsetup utility was run in FIPS (Federal Information Processing Standards) mode, the "Running in FIPS mode." message was displayed during initialization of all commands. This sometimes caused minor issues with associated scripts. This bug has been fixed and the message is now displayed only in verbose mode.

BZ#732179

Prior to this update, several directives were missing in cryptsetup status command implementation. Therefore, the cryptsetup status command always returned the exit code 0 when verifying the status of a mapped device. To fix this issue, the code has been modified. The cryptsetup status command now returns the 0 value only if the device checked is active.

BZ#701936

Previously, the libcryptsetup crypt_get_volume_key() function allowed to perform an action not compliant with FIPS. To conform FIPS requirements, the function is now disabled in FIPS mode and returns an EACCES error code to indicate it. Note that the "luksDump --dump-master-key" command and the key escrow functionality of the volume_key package are also disabled in FIPS mode as a consequence of this update.

BZ#713456

When the cryptsetup or libcryptsetup utility was run in FIPS (Federal Information Processing Standards) mode, the "Running in FIPS mode." message was displayed during initialization of all commands. This sometimes caused minor issues with associated scripts. This bug has been fixed and the message is now displayed only in verbose mode.

BZ#709055

Previously, the libcryptsetup crypt_get_volume_key() function allowed to perform an action not compliant with FIPS. To conform FIPS requirements, the function is now disabled in FIPS mode and returns an EACCES error code to indicate it.

Note that the "luksDump --dump-master-key" command and the key escrow functionality of the volume_key package are also disabled in FIPS mode as a consequence of this update.

BZ#728545

Prior to this update, the ctdb daemon leaked a file descriptor to anon_inodefs. This update modifies ctdb so that this file discriptor can no longer leak.

BZ#672641

This update adds support for Clustered Samba on top of GFS2 as a Technology Preview.

Security Fix

CVE-2011-2896

A heap-based buffer overflow flaw was found in the Lempel-Ziv-Welch (LZW) decompression algorithm implementation used by the CUPS GIF image format reader. An attacker could create a malicious GIF image file that, when printed, could possibly cause CUPS to crash or, potentially, execute arbitrary code with the privileges of the "lp" user.

BZ#681836

Previously CUPS was not correctly handling the language setting LANG=en_US.ASCII. As a consequence lpadmin, lpstat and lpinfo binaries were not displaying any output when the LANG=en_US.ASCII environment variable was used. As a result of this update the problem is fixed and the expected output is now displayed.

BZ#706673

Previously the scheduler did not check for empty values of several configuration directives. As a consequence it was possible for the CUPS daemon (cupsd) to crash when a configuration file contained certain empty values. With this update the problem is fixed and cupsd no longer crashes when reading such a configuration file.

BZ#709896

Previously when printing to a raw print queue, when using certain printer models, CUPS was incorrectly sending SNMP queries. As a consequence there was a noticeable 4-second delay between queueing the job and the start of printing. With this update the problem is fixed and CUPS no longer tries to collect SNMP supply and status information for raw print queues.

BZ#712430

Previously when using the BrowsePoll directive it could happen that the CUPS printer polling daemon (cups-polld) began polling before the network interfaces were set up after a system boot. CUPS was then caching the failed hostname lookup. As a consequence no printers were found and the error, "Host name lookup failure", was logged. With this update the code that re-initializes the resolver after failure in cups-polld is fixed and as a result CUPS will obtain the correct network settings to use in printer discovery.

BZ#735505

The MaxJobs directive controls the maximum number of print jobs that are kept in memory. Previously, once the number of jobs reached the limit, the CUPS system failed to automatically purge the data file associated with the oldest completed job from the system in order to make room for a new print job. This bug has been fixed, and the jobs beyond the set limit are now properly purged.

BZ#744791

The cups init script (/etc/rc.d/init.d/cups) uses the daemon function (from /etc/rc.d/init.d/functions) to start the cups process, but previously it did not source a configuration file from the /etc/sysconfig/ directory. As a consequence, it was difficult to cleanly set the nice level or cgroup for the cups daemon by setting the NICELEVEL or CGROUP_DAEMON variables. With this update, the init script is fixed.

BZ#736304

The MaxJobs directive controls the maximum number of print jobs that are kept in memory. Previously, once the number of jobs reached the limit, the CUPS system failed to automatically purge the oldest completed job from the system to make room for a new one. This bug has been fixed, and the jobs beyond the limit are now properly purged in the described scenario.

Security Fix

CVE-2011-2192

It was found that cURL always performed credential delegation when authenticating with GSSAPI. A rogue server could use this flaw to obtain the client's credentials and impersonate that client to other servers that are using GSSAPI.

BZ#727884

As a solution to a security issue, GSSAPI credential delegation was disabled, which broke the functionality of the applications that were relying on delegation, which was incorrectly enabled by libcurl. To fix this issue, the CURLOPT_GSSAPI_DELEGATION libcurl option has been introduced in order to enable delegation explicitly when applications need it. All applications using GSSAPI credential delegation can now use this new libcurl option to be able to run properly.

Security Fix

CVE-2011-1926

It was discovered that cyrus-imapd did not flush the received commands buffer after switching to TLS encryption for IMAP, LMTP, NNTP, and POP3 sessions. A man-in-the-middle attacker could use this flaw to inject protocol commands into a victim's TLS session initialization messages. This could lead to those commands being processed by cyrus-imapd, potentially allowing the attacker to steal the victim's mail or authentication credentials.

Security Fix

CVE-2011-3208

A buffer overflow flaw was found in the cyrus-imapd NNTP server, nntpd. A remote user able to use the nntpd service could use this flaw to crash the nntpd child process or, possibly, execute arbitrary code with the privileges of the cyrus user.

Security Fixes

CVE-2011-3372

An authentication bypass flaw was found in the cyrus-imapd NNTP server, nntpd. A remote user able to use the nntpd service could use this flaw to read or post newsgroup messages on an NNTP server configured to require user authentication, without providing valid authentication credentials.

CVE-2011-3481

A NULL pointer dereference flaw was found in the cyrus-imapd IMAP server, imapd. A remote attacker could send a specially-crafted mail message to a victim that would possibly prevent them from accessing their mail normally, if they were using an IMAP client that relies on the server threading IMAP feature.

BZ#720451

Prior to this update, the ntlm plug-in did not work due to a code error. This update modifies the source code so that the plug-in now works as expected.

BZ#730242

Prior to this update, creating the user ID and the group ID of the saslauth daemon caused conflicts. This update corrects this behavior and now the saslauth daemon works as expected.

BZ#730246

Prior to this update, cyrus-sasl displayed redundant warnings during the compilation. With this update, cyrus-sasl has been modified and now works as expected.

BZ#727274

This update adds support of partial Relocation Read-Only (RELRO) for the cyrus-sasl libraries.

Security Fix

CVE-2011-2200

A denial of service flaw was found in the way the D-Bus library handled endianness conversion when receiving messages. A local user could use this flaw to send a specially-crafted message to dbus-daemon or to a service using the bus, such as Avahi or NetworkManager, possibly causing the daemon to exit or the service to disconnect from the bus.

Bug Fixes

BZ#677449

DM Multipath removed a device if it failed to check the device status due to insufficient memory. This happened because the command checking if the device map existed failed as the system returned an error. With this update, Multipath no longer returns an error under these circumstances and no devices are removed if the system runs out of memory while checking device status.

BZ#678673

If a device-mapper-multipath device was open but all attached device paths had been lost, the device was unable to create a new table with no device paths. As a concequence the multipath -ll command returned output indicating that no paths to the device were available with confusing "failed faulty running" rows presenting the missing paths. Multipath devices now reload tables with no device paths correctly.

BZ#689504

Device paths could fail even if unavailable only temporarily. This happened because the RDAC (Redundant Disk Array Controller) checker function did not recheck the status of hosts if it had received a temporary error code. The function now rechecks the path after it has received such error codes and the path failures are transient as expected.

BZ#697386

A previous bug fix introduced a race condition between the main thread and the thread running the checkerloop routine as the checkerloop thread was created with deferred cancellation type. The checkerloop thread continued running and attempted to access a property, which had been previously unallocated by the main thread. This caused the multipathd daemon to shutdown with a segmentation fault. Now the checkerloop thread checks if a shutdown is in progress and the deamon shuts down gracefully.

BZ#700169

The Multipath daemon failed to include some ghost paths when counting the number of active paths; however, when the ghost paths failed, they were subtracted from the number of active paths. This caused multipathd to fail IO requests even though some paths were still available. The Multipath daemon now counts ghost paths correctly and no longer fails IO requests while there are still active paths available.

BZ#705854

If the user set dev_loss_tmo to a value greater than 600 in multipath.conf without setting the fast_io_fail_tmo value, the multipathd daemon did not notify the user that fast_io_fail_tmo was not set. Multipath now issues a warning that fast_io_fail_tmo is not set under such circumstances.

BZ#706555

On shared-storage multipath setups that set failback to manual, multipath could keep alternating from the failover pathgroup to the primary pathgroup infinitely. This happened because multipath was incorrectly failing back to the primary pathgroup whenever a path priority changed. With this update, multipath no longer fails back to the primary pathgroup when a path's priority changes under such circumstances.

BZ#707560

If the multipath device was deleted while a path was being checked, multipathd did not abort the path check and terminated unexpectedly when trying to access the multipath device information. The Multipath daemon now aborts any path checks when the multipath device is removed and the problem no longer occurs.

BZ#714821

The Multipath daemon was removing a multipath device twice. This could cause multipathd to access memory already used for another purpose, and caused the multipathd daemon to terminate unexpectedly. The multipathd daemon now removes the device once and the problem no longer occurs.

BZ#719571

The kpartx utility built partition devices for invalid GUID partition tables (GPT) because it did not validate the size of GUID partitions. The kpartx utility now checks the partition size, and does not build devices for invalid GPTs.

BZ#723168

Multipath previously returned an unclear error message when it failed to find rport_id. The returned message and its severity have been adjusted.

BZ#725541

Several upstream commits have been included in the device-mapper-multipath package providing a number of bug fixes and enhancements over the previous version.

BZ#738298

Anaconda failed to recognize an existing filesystem on a zSeries Linux fibre-channel adapter (zFCP) LUN and marked it as 'Unknown' when reinstalling the system. This happened due to an incorrect setting of the DM_UDEV_DISABLE_DISK_RULES_FLAG property. Filesystem on a multipath zFCP LUN is now correctly recognized during the installation.

BZ#747604

The asynchronous TUR path checker caused multipathd to terminate unexpectedly due to memory corruption. This happened if multipathd attempted to delete a path while the asynchronous TUR checker was running on the path. The asynchronous TUR checker code has been removed, and multipathd no longer crashes on path removal.

Enhancements

BZ#636009

Multipath now supports up to 8000 device paths.

BZ#683616

To provide support for Asymmetric Logical Unit Access (ALUA), the RDAC checker has been modified to work better with devices in IOSHIP mode. The checker now sets the Task Aborted Status (TAS) bit to 1 if the TAS bit is set to 0 and changeable on a LUN (Logical Unit Number) discovery. The function now also reports PATH_UP for both the path groups in the RDAC storage in IOSHIP mode.

BZ#694602

To run multipath on IBM BladeCenter S-series with RAIDed Shared Storage Module (RSSM) demanded a manual multipath configuration to enable RSSM. Multipath now configures the server automatically.

BZ#699577

The text in the defaults multipaths devices sections of the multipath.conf man page has been improved to provide a better clarification.

BZ#713754

The rr_min_io_rq option has been added to the default, devices, and multipaths sections of the multipath.conf file. This option defines the number of I/O requests to route to a path before switching to the next path in the current path group. Note that the rr_min_io option is no longer used.

BZ#710478

UID, GID, and mode owner settings defined in /etc/multipath.conf for a multipath device are ignored. These access permissions are now set with the udev rules.

BZ#732384

When deleting a multipath device while checking a path, the multipathd daemon did not abort the path check. As a consequence, the daemon terminated when trying to access multipath device information. The problem has been fixed and the multipathd daemon now aborts the path check when deleting a multipath device.

BZ#625880

To allow administrators easily disable the suspend and hibernate actions on the system, DeviceKit-power now checks the PolicyKit authorization before deciding whether an action can be completed.

BZ#727544

This update introduces a new sub-package DeviceKit-power-devel-docs, which contains developer's documentation for DeviceKit-power, so that it is now possible to install the DeviceKit-power-devel package on machines with multiple architectures without file conflicts.

Security Fixes

CVE-2011-2748, CVE-2011-2749

Two denial of service flaws were found in the way the dhcpd daemon handled certain incomplete request packets. A remote attacker could use these flaws to crash dhcpd via a specially-crafted request.

BZ#694798

Previously, when multiple DHCP clients were launched at the same time to handle multiple virtual interfaces on the same network interface card (NIC), the clients used the same seed to choose when to renew their leases. Consequently, these virtual interfaces for some clients could have been removed over time. With this update, the dhclient utility uses the Process Identifier (PID) for seeding the random number generator, which fixes the bug.

BZ#694799

If a system was rebooted while a network switch was inoperative, the network connection would recover successfully. However, it was no longer configured to use DHCP even if the dhclient utility had been running in persistent mode. With this update, the dhclient-script file has been modified to refresh the ARP (Address Resolution Protocol) table and the routing table instead of bringing the interface down, which fixes the bug.

BZ#731990

If the system included network interfaces with no hardware address, the dhcpd scan could have experienced a segmentation fault when scanning such an interface. As a consequence, the dhcpd daemon unexpectedly terminated. To prevent this issue, dhcpd now tests a pointer which represents the hardware address of the interface for the NULL value. The dhcp daemon no longer crashes.

BZ#736999

Previously, all source files were compiled with the "-fpie" or "fPIE" flag. As a consequence, the libraries used by dhcp could not have been used to build Perl modules. To fix this problem, all respective dhcp Makefiles have been modified to compile libraries with the "-fpic" or "-fPIC" flag. The libraries used by dhcp are now built without the previous restrictions.

BZ#736194

Previously, both dhcp and dhclient packages included the dhcp-options(5) and dhcp-eval(5) man pages. As a consequence, a conflict could have occurred when any of these man pages were updated, because dhcp and dhclient packages could have been upgraded separately. To prevent the problem from occurring in future updates, shared files of dhcp and dhclient packages have been moved to the dhcp-common package that is required by both dhcp and dhclient as a dependency.

BZ#706974

A feature has been backported from dhcp version 4.2.0. This feature allows the DHCPv6 server to be configured to identify DHCPv6 clients in accordance with their link-layer address and their network hardware type. With this update, it is now possible to define a static IPv6 address for the DHCPv6 client with a known link-layer address.

BZ#693381

Previously, the dhcpd daemon ran as root. With this update, new "-user" and "-group" options can be used with dhcpd. These options allow dhcpd to change the effective user and group ID after it starts. The dhcpd and dhcpd6 services now run the dhcpd daemon with the "-user dhcpd -group dhcpd" parameters, which means that the dhcpd daemon runs as the dhcpd user and group instead root.

Bug Fix

BZ#744690

Previously, extended records for Memory Device (DMI type 17) and Memory Array Mapped Address (DMI type 19) DMI types were missing from the dmidecode utility output. With this update, dmidecode has been upgraded to upstream version 2.11, which updates support for the SMBIOS specification to version 2.7.1, thus fixing this bug.

Bug Fixes

BZ#584009

Three changes were made to /etc/init.d/dnsmasq, the dnsmasq startup script.

If dnsmasq was started or restarted by a non-privileged user, the startup script previously failed silently. With this update, the dnsmasq startup script now exits with a status code of 4 (user had insufficient privilege) and returns a "User has insufficient privilege" error to STD OUT when started or restarted by a non-privileged user.

A "force-reload" option was added: The "service force-reload dnsmasq" command now forces dnsmasq to reload. Previously, it did nothing.

If /etc/init.d/dnsmasq passed an invalid argument, previously the startup script exited with a status code of 1 (generic or unspecified error). With this update, the startup script now exits correctly, returning a status code of 2 (invalid or excess argument) in such a circumstance.

BZ#704073

If the virtual bridge interface (virbr0) was up and dnsmasq was started by default, dnsmasq could, in some circumstances, write a "DHCP packet received on eth(x) which has no address" message to /var/log/messages. Note: this message was not in error. The message was written if an actual interface (eg eth1) was up; did not have a configured IP address (eg was slaved to a logical bonded interface); and was in the same LAN as another host which generated a DHCP request. The message had little-to-no utility, however: it presented a warning where none was needed. With this update, this message is no longer written to /var/log/messages in these, and equivalent, circumstances.

BZ#624596

Previously, when the dosfsck and the dosfslabel utilities were executed on the IBM System z architecture using a FAT32 file system, they terminated with this error message: "Logical sector size is zero". This was caused by unaligned fields which were first byte-wise copied. With this fix, the fields are not pre-copied any more, but are accessed the same way as on the i686 architecture.

BZ#677789

The fsck.vfat utility terminated due to buffer overflow. This occurred when checking a device with the corrupted VFAT file system if there were any chains of orphaned clusters. The name of the newly created file that contained these clusters was printed directly into the name field, which led to an out of boundary write. The name is now printed into the buffer and individual parts are then correctly copied into the appropriate field.

BZ#688128

The dosfslabel utility displayed an error message when labeling the FAT32 file system due to some of its internal structures being initialized incorrectly. The dosfslabel utility now reads the FAT file system first, which fixes the problem.

BZ#709266

The mkfs.vfat utility did not correctly detect device partitions on RAID devices. As a consequence, formatting failed with an error message. This was caused by an invalid mask for the statbuf.st_rdev variable. The mask has been fixed to be at least four bytes long and the problem no longer occurs.

Security Fix

CVE-2011-1929

A denial of service flaw was found in the way Dovecot handled NULL characters in certain header names. A mail message with specially-crafted headers could cause the Dovecot child process handling the target user's connection to crash, blocking them from downloading the message successfully and possibly leading to the corruption of their mailbox.

BZ#690076

Prior to this update, Doxygen required invalid BuildRequires on the qt-devel package. With this update, packages with BuildRequires dependencies on the qt-devel package have been fixed. Now, these packages explicitly require qt4-devel.

Bug Fixes

BZ#659076

Previously, dracut incorrectly displayed that it loaded SELinux even if SELinux was disabled in the config file and "selinux=0" was not specified on the kernel command line. As a consequence, an error message could confuse the user when booting the system. With this update, the dracut utility is modified and the error message no longer appears.

BZ#696980

Due to an error in the dracut module script, the system could fail to find the root volume if a static IP address was specified. As a consequence, the system did not boot. With this update, the error is corrected, and the system is able to boot with a static IP address.

BZ#698160

When mounting the root device over the NFS (Network File System) protocol, the /var/lib/rpcbind directory created by initramfs was world-writable. The dracut tool has been modified to generate initramfs which now sets the ownership to the rpc user and the group.

BZ#698165

When auto-assembling an md RAID device, initramfs used an invalid parameter when calling the mdadm tool. This prevented the system from booting if the root device was on the RAID device. The invalid parameter has been removed and the system now boots properly.

BZ#698215

When auto-assembling an md RAID device, an error in the mdraid_start.sh script prevented the system from booting if the root device was on the RAID device. The error in the script has been fixed and the system now boots correctly.

BZ#701309

Prior to this update, the /var/lib/nfs/prc_pipefs partition could not be accessed on system boot. The problem occurred when booting the system with NFS set as the root partition with at least one separate /var partition. This was caused by initramfs mounting the /var partition over the existing rpc_pipefs partition. The initramfs file system now mounts entries in /etc/fstab.sys, which fixes the problem.

BZ#707609

The dm-mod and dm-crypt kernel modules were missing from the list of kernel modules, which are pre-loaded for the FIPS-140 check. These modules have been added to the list with this update.

BZ#712254

When loading SELinux from inside initramfs, the output of the SELinux commands could be garbled if the user used non-Latin locales. The initramfs file system has been modified to turn off localization for the SElinux commands, which results in readable messages.

BZ#737134

The QLogic qla4xxx iSCSI driver and the iSCSI (Internet Small Computer System Interface) transport layer now support iSCSI boot from Storage Area Network (SAN) using the iscsistart. With this update, dracut is modified to support these changes.

BZ#737593

If the user installed a system with rootfs on a RAID device where RAID members were encrypted, dracut failed to assemble the RAID device on reboot. As a consequence, the system did not boot. A patch has been applied to address this issue, and the RAID device is now assembled on every boot so that the system boots successfully.

BZ#741430

When applying SELinux labels for /dev in initramfs, the restorecon tool did not alter the MCS/MLS label only types. To fix this problem, the "-F" option has been added to all calls of restorecon.

BZ#742920

Prior to this update, the boot process timed out for network settings with DHCP involved. A patch has been applied to extend the timeout interval if DHCP is involved, which fixes the problem.

Enhancements

BZ#701864

This update adds support for iSCSI (Internet Small Computer System Interface) partial offload functionality for certain Broadcom network devices.

BZ#740487

This update adds the dracut-fips-aesni subpackage. Note that the package should be installed when using the aesni-intel module in FIPS mode.

BZ#723548

This update adds support for Logical Volume Management (LVM) mirror devices to serve as root devices. Additionally degraded mirrors are used after a certain timeout if the other half cannot be found at booting time.

BZ#729573

This update adds support for configuring an interface with automatic IPv6 and DHCP over IPv4 by using the "ip=[interface]:dhcp,auto6" command line parameter.

BZ#736094

With this update, the Broadcom FCoE (Fibre Channel over Ethernet) offload driver is now supported.

BZ#728549

The dm-mod and dm-crypt kernel modules were missing from the list of kernel modules which are pre-loaded for the FIPS-140 (Federal Information Processing Standards) check. With this update, these modules have been added to the list. This update also introduces the dracut-fips-aesni subpackage which should be installed if the aesni-intel module is used in FIPS mode.

BZ#702593

Prior to this update, the dump utility passed wrong arguments to the "clone(2)" system call. As a result, dump became unresponsive when executed on the S/390 or IBM System z architecture. This bug has been fixed in this update so that dump now passes correct arguments and no longer hangs.

BZ#691434

Under certain circumstances, the dump utility could have failed to detect holes in files correctly. When a user attempted to restore an erroneous backup using the "restore" command, an error message "Missing blocks at end of [path], assuming hole" could have been displayed. In such case, the backup could have not been restored properly. This bug has been fixed in this update so that dump now handles holes in files as expected.

BZ#658890

Prior to this update, the "dump -w" command did not recognize ext4 file systems as supported. With this update, the bug has been fixed so that "dump -w" now recognizes the ext4 file systems as supported.

BZ#676465

Running the "e2fsck" command on certain corrupted file systems failed to correct all errors during the first run. This occurred when a file had its xattr block cloned as a duplicate, but the block was later removed from the file because the file system did not contain the xattr feature. However, the block was not cleared from the block bitmaps. During the second run, e2fsck found the cloned xattr block as in use, but not owned by any file, and had to repair the block bitmaps. With this update, the processing of duplicate xattr blocks is skipped on non-xattr file systems. All problems are now discovered during the first run.

BZ#679931

On certain devices with very large physical sector size, the mke2fs utility set the block size to be as large as the size of the physical sector. In some cases, the size of the physical sector was larger than the page size. As a consequence, the file system could not be mounted and, in rare cases, the utility could even fail. With this update, the default block size is not set to be larger than the system's page size, even for large physical sector devices.

BZ#683906

Previously, multiple manual pages contained typos. These typos have been corrected with this update.

BZ#713475

This update modifies parameters of the "mke2fs" command to be consistent with the "discard" and "nodiscard" mount options for all system tools (like mount, fsck, or mkfs). The user is now also informed about the ongoing discard process.

BZ#730083

Previously, the libcomm_err libraries were built without the read-only relocation (RELRO) flag. Programs built against these libraries could be vulnerable to various attacks based on overwriting the ELF section of a program. To enhance the security, the e2fsprogs package is now provided with partial RELRO support.

BZ#679892

Previously, the tune2fs tool could not set "barrier=0" as the default option on the ext3 and ext4 file systems. With this update, users are now able to set this option when creating the file system, and do not have to maintain the option in the /etc/fstab file across all of the file systems and servers.

BZ#713468

Previously, raw e2image output files could be extremely large sparse files, which were difficult to copy, archive, and transport. This update adds support for exporting images in the qcow format. Images in this format are small and easy to manipulate.

Security Fixes

CVE-2011-1831

A race condition flaw was found in the way mount.ecryptfs_private checked the permissions of a requested mount point when mounting an encrypted file system. A local attacker could possibly use this flaw to escalate their privileges by mounting over an arbitrary directory.

CVE-2011-1832

A race condition flaw in umount.ecryptfs_private could allow a local attacker to unmount an arbitrary file system.

CVE-2011-1834

It was found that mount.ecryptfs_private did not handle certain errors correctly when updating the mtab (mounted file systems table) file, allowing a local attacker to corrupt the mtab file and possibly unmount an arbitrary file system.

CVE-2011-1835

An insecure temporary file use flaw was found in the ecryptfs-setup-private script. A local attacker could use this script to insert their own key that will subsequently be used by a new user, possibly giving the attacker access to the user's encrypted data if existing file permissions allow access.

CVE-2011-1837

A race condition flaw in mount.ecryptfs_private could allow a local attacker to overwrite arbitrary files.

CVE-2011-3145

A race condition flaw in the way temporary files were accessed in mount.ecryptfs_private could allow a malicious, local user to make arbitrary modifications to the mtab file.

CVE-2011-1833

A race condition flaw was found in the way mount.ecryptfs_private checked the permissions of the directory to mount. A local attacker could use this flaw to mount (and then access) a directory they would otherwise not have access to.

Note

The fix for this issue is incomplete until a kernel-space change is made. Future Red Hat Enterprise Linux 5 and 6 kernel updates will correct this issue.

BZ#253077

If the user resized an ESC window and closed it, the window did not preserve its size when opening it again. If the user wanted the window to be larger, for example, to make it easier to read, the user had to resize the window every single time when it was opened again. A patch has been applied to address this issue and the previous window size is now restored when opening ESC.

BZ#682216

Previously, during the shut down sequence of the escd daemon, the daemon reported a failure of certain instances. ESC terminated unexpectedly with a segmentation fault as a consequence. This update modifies the daemon to exit quietly. As a result, ESC no longer terminates unexpectedly.

BZ#702683

The esc-prefs.js file contains helpful commented settings designed to assist the user in trying rarely used settings if the situation warrants. A number of these settings in the file contained typos. The typos have been corrected with this update.

BZ#704281

Previously, ESC could have terminated with a segmentation fault after the user had inserted a new smart card into the reader. This was due to a bug in the code which helped to bring a pop-up window to the foreground. The code is no longer needed to assure window focus, therefore it is no longer being executed. As a result, ESC no longer terminates in the scenario described.

BZ#730647

Due to multilib problems, the fakeroot command was only built on 64-bit architectures as a workaround, one which prevented the RPM package from being built on other architectures. This update resolves the multilib problems so that fakeroot now builds successfully on all architectures.

BZ#639466

When stopping the fcoe service, the fcoe initscript did not properly clean up after itself as expected (did not remove FCoE devices, kill related processes and unload FCoE drivers). As a consequence, FCoE interfaces were not brought down and FCoE related threads were still running after the fcoe had been stopped. The "service fcoe stop" command is used to ensure safe after-update service restarts on FCoE dependent systems, therefore it cannot be used to remove FCoE devices and unload related kernel modules. Concerning this situation, the initscript has been modified to use the "stop force" command option to completely remove FCoE devices and unload related kernel modules. The fcoe service now should be stopped using the "service fcoe stop force" command.

BZ#732485

When removing a network interface with no fcoe port using the "fcoeadm -d" command, the fcoe port state machine set the removal operation incorrectly to wait without responding to fcoemon. This led to an internal error because fcoemon timed out waiting for the response. To resolve the problem, the code has been modified to return the code for no further action under these circumstances. The "fcoeadm -d" command now works for interfaces without the fcoe port as expected.

BZ#732485

The fcoemon service did not maintain any information about the relative state of a physical network interface and its dependent VLAN interfaces. As a consequence, the fcoe port of the VLAN interface could have been out of sync with the fcoe port of the physical device, resulting in undesired behavior, such as processing link events improperly. To fix this problem, a ready flag has been introduced. This flag is set to false when the physical port is disabled. Link events are now processed correctly for the vlan ports.

BZ#732485

When answering to an FCoE Initialization Protocol (FIP) VLAN Discovery request, some switches encapsulate FIP VLAN Discovery replies in a VLAN 0 tag which is wrapped around the packet's FIP frame header. Previously, when a packet containing such a reply reached a target network interface, some devices did not remove the VLAN tag before they started to process the FIP header. If the VLAN tag was not removed, the length of the processed header was larger than was expected, therefore the FIP parsing logic was not able to parse the FIP header correctly causing a loss of the packet. With this update, the parsing logic has been modified to skip over the VLAN header when necessary, and point to the correct start of the FIP header.

BZ#743689

The timeout for a kernel reply to fcoeadm operations was set to 5 seconds, which was not enough when processing an fcoeadm operation on a system with a large number of FCoE ports while a kernel was under heavy load. As a consequence, the "internal error" message was displayed even though the operation was finished successfully. To prevent this bug, the timeout for the kernel reply was increased to 30 seconds. No error message is now sent when an fcoeadm operation succeeds.

BZ#731166

Due to a change in REST API, the fence_rhevm utility incorrectly reported status "UP" as "RUNNING". Consequently, the "fence_rhevm -o status" command always reported "OFF". This bug has been fixed, and fence_rhevm now reports status correctly.

BZ#718924

The fence_drac5 agent failed to clear its SSH sessions on exit as expected by firmware. Consequently, the fence agent appeared to be still connected to the device, and once the connection limit was reached, further logins to the device were not allowed. This bug has been fixed, and fence_drac5 now clears its SSH sessions properly.

BZ#693428

The "monitor" and "status" commands of the fence_ipmilan agent returned chassis status instead of the fence device status. As a result, when a server chassis was powered off, the fence_ipmilan agent exited with the incorrect result code "2" when passed one of these commands. Now, fence_ipmilan returns the correct result code "0" in the described scenario.

BZ#708052

When a blade server was removed from a blade chassis and was fenced via the fence_bladecenter utility with the "--missing-as-off" option enabled, and was scheduled with the "reboot" action, the fence failed. This bug has been fixed, and fence_bladecenter no longer returns an error if a blade server is missing.

BZ#718196

A list operation on fence_drac5 agents resulted in unexpected termination of fence agents. A patch has been provided to address this issue, and fence_drac5 agents now work correctly in the described scenario.

BZ#718207

When the pyOpenSSL package was not present in the system, when an error occurred, the fence_ilo agent terminated with a generic error message, making it difficult to debug the problem. Now, fence_ilo reports that a dependent package is missing in the described scenario, thus fixing this bug.

BZ#732372

The verbose mode of the fence_ipmilan agent exposed user passwords when the whole command was logged by an IPMI tool. Now, the fence_ipmilan output has been changed, and passwords remain undisclosed in the described scenario.

BZ#738384

During simultaneous unfencing operations performed via the fence_scsi agent, all nodes launched their reservation commands at the same time. Consequently, some of the commands failed. Now, fence_scsi retries to unfence a node until its reservation command succeeds.

BZ#734429

A null dereference was discovered in the fence_kdump agent, when the strchr() function returned the NULL value. With this update, the dereference has been fixed in the code and no longer occurs.

BZ#624673

With this update, the new fence_vmware_soap() function has been provided to enable fencing of VMware guests in ESX environments.

BZ#461948

The fence_kdump utility has been updated to integrate fencing with the kernel dump environment.

BZ#698365

With this update, the RelaxNG schema generation for fence-agents has been updated with the rha:description and rha:name attributes in its output to fence attribute group elements.

BZ#726571

The fence_ipmilan agent has been updated to support the -L option of the ipmilan daemon, thus supporting fencing with user session privileges level.

BZ#719645

Prior to this update, the domain parameter was missing from the metadata. As a consequence, existing configurations utilizing the domain parameter did not function correctly when fencing. This update adds the domain parameter for compatibility. Now, existing configurations work as expected.

BZ#720767

Prior to this update, hash mismatches falsely returned successes for fencing. As a consequence, data corruption could occur in live-hang scenarios. This update corrects the hash handling of mismatches. Now, no more false successes are returned and the data integrity is preserved.

BZ#691200

With this update, the libvirt-qpid plugin now operates using QMF version 2.

BZ#676045, BZ#712992, BZ#712988

Prior to this update, the file utility could have been unable to recognize RPM files for certain supported architectures. This update improves the file type recognition, and the RPM files for all supported architectures are now correctly identified as expected.

BZ#688700

Prior to this update, the file utility did not correctly recognized the IBM System z kernel images. This problem has been corrected so that the IBM System z kernel images are now correctly recognized as expected.

BZ#692098

Prior to this update, the file utility attempted to show information related to core dumps for binary files that were not core dumps. This undesired behavior has been fixed in this update so that information related to core dumps is showed only for core dumps and not for the binary files which are not core dumps.

BZ#675691

Prior to this update, file patterns for LaTeX checked only the first 400 bytes of a file to determine the pattern type. This caused an incorrect pattern type recognition as some files could have contained a larger number of comments at the beginning of the file. Furthermore, file patterns which matched a Python script were tried before the LaTeX patterns and this undesired behavior could have caused an incorrect pattern type recognition as LaTeX files could have included a source code written in Python. With this update, the aforementioned problems have been fixed by increasing the number of first bytes checked for a LaTeX file to 4096 bytes, and by trying the LaTeX patterns before the Python patterns.

BZ#690801

Prior to this update, there were several spelling mistakes contained in the magic(5) manual page. This update corrects the spelling mistakes in the respective manual page.

BZ#716665

Prior to this update, the file utility treated MP3 files as text files, and therefore was unable to recognize the MP3 files. This undesired behavior has been fixed in this update, and the file utility now treats the MP3 files as binary files and is able to properly recognize them.

BZ#620063

Prior to this update, certain locale subdirectories in the /usr/share/locale/ directory did not have any owner set. With this update, this bug has been fixed so that the filesystem package now owns the subdirectories of the following locales: bg_BG (Bulgarian), en_NZ (New Zealand English), fi_FI (Finnish), gl_ES (Galician), lv_LV (Latvian), ms_MY (Malaysian), sr_RS (Serbian), en@shaw (Shavian), zh_CN.GB2312 (Chinese Simplified), sr@ijekavian (Serbian Jekavian), and sr@ijekavianlatin (Serbian Jekavian Latin).

BZ#727277

Prior to this update, the fipscheck library was linked without support for read-only relocations (RELRO) flags. The updated fipscheck packages are now provided with partial RELRO support.

Security Fixes

CVE-2011-2377

A flaw was found in the way Firefox handled malformed JPEG images. A website containing a malicious JPEG image could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-0083, CVE-2011-0085, CVE-2011-2363

Multiple dangling pointer flaws were found in Firefox. A web page containing malicious content could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2364, CVE-2011-2365, CVE-2011-2374, CVE-2011-2375, CVE-2011-2376

Several flaws were found in the processing of malformed web content. A web page containing malicious content could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2371

An integer overflow flaw was found in the way Firefox handled JavaScript Array objects. A website containing malicious JavaScript could cause Firefox to execute that JavaScript with the privileges of the user running Firefox.

CVE-2011-2373

A use-after-free flaw was found in the way Firefox handled malformed JavaScript. A website containing malicious JavaScript could cause Firefox to execute that JavaScript with the privileges of the user running Firefox.

CVE-2011-2362

It was found that Firefox could treat two separate cookies as interchangeable if both were for the same domain name but one of those domain names had a trailing "." character. This violates the same-origin policy and could possibly lead to data being leaked to the wrong domain.

BZ#698313

With previous versions of Firefox on Red Hat Enterprise Linux 5, the "background-repeat" CSS (Cascading Style Sheets) property did not work (such images were not displayed and repeated as expected).

Security Fixes

CVE-2011-2982

Several flaws were found in the processing of malformed web content. A web page containing malicious content could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-0084

A dangling pointer flaw was found in the Firefox Scalable Vector Graphics (SVG) text manipulation routine. A web page containing a malicious SVG image could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2378

A dangling pointer flaw was found in the way Firefox handled a certain Document Object Model (DOM) element. A web page containing malicious content could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2981

A flaw was found in the event management code in Firefox. A website containing malicious JavaScript could cause Firefox to execute that JavaScript with the privileges of the user running Firefox.

CVE-2011-2983

A flaw was found in the way Firefox handled malformed JavaScript. A web page containing malicious JavaScript could cause Firefox to access already freed memory, causing Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2984

It was found that a malicious web page could execute arbitrary code with the privileges of the user running Firefox if the user dropped a tab onto the malicious web page.

Security Fix

BZ#734316

It was found that a Certificate Authority (CA) issued a fraudulent HTTPS certificate. This update renders any HTTPS certificates signed by that CA as untrusted, except for a select few. The now untrusted certificates that were issued before July 1, 2011 can be manually re-enabled and used again at your own risk in Firefox; however, affected certificates issued after this date cannot be re-enabled or used.

Security Fix

BZ#735483

The RHSA-2011:1242 Firefox update rendered HTTPS certificates signed by a certain Certificate Authority (CA) as untrusted, but made an exception for a select few. This update removes that exception, rendering every HTTPS certificate signed by that CA as untrusted.

Security Fixes

CVE-2011-2995

Several flaws were found in the processing of malformed web content. A web page containing malicious content could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

CVE-2011-2372

A flaw was found in the way Firefox processed the "Enter" keypress event. A malicious web page could present a download dialog while the key is pressed, activating the default "Open" action. A remote attacker could exploit this vulnerability by causing the browser to open malicious web content.

CVE-2011-3000

A flaw was found in the way Firefox handled Location headers in redirect responses. Two copies of this header with different values could be a symptom of a CRLF injection attack against a vulnerable server. Firefox now treats two copies of the Location, Content-Length, or Content-Disposition header as an error condition.

CVE-2011-2999

A flaw was found in the way Firefox handled frame objects with certain names. An attacker could use this flaw to cause a plug-in to grant its content access to another site or the local file system, violating the same-origin policy.

CVE-2011-2998

An integer underflow flaw was found in the way Firefox handled large JavaScript regular expressions. A web page containing malicious JavaScript could cause Firefox to access already freed memory, causing Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

Security Fixes

CVE-2011-3647

A flaw was found in the way Firefox handled certain add-ons. A web page containing malicious content could cause an add-on to grant itself full browser privileges, which could lead to arbitrary code execution with the privileges of the user running Firefox.

CVE-2011-3648

A cross-site scripting (XSS) flaw was found in the way Firefox handled certain multibyte character sets. A web page containing malicious content could cause Firefox to run JavaScript code with the permissions of a different website.

CVE-2011-3650

A flaw was found in the way Firefox handled large JavaScript scripts. A web page containing malicious JavaScript could cause Firefox to crash or, potentially, execute arbitrary code with the privileges of the user running Firefox.

BZ#664876

Previously, FirstAidKit's GRUB plug-in incorrectly reported failure if GRUB was installed into the Master Boot Record (MBR). Due to the plug-in being unreliable, it has been removed from the firstaidkit package.

BZ#738563

The firstaidkit-plugin-grub package has been removed from Red Hat Enterprise Linux 6.2. As a consequence, in rare cases, the system upgrade operation may fail with unresolved dependencies if the plug-in has been installed in a previous version of Red Hat Enterprise Linux. To avoid this problem, the firstaidkit-plugin-grub package should be removed before upgrading the system. However, in most cases, the system upgrade completes as expected.

BZ#700283

Previously, the Traditional Chinese translation (zh_TW) of the Forward button on the welcome page was different from the action mentioned in the text, on the same page, referring to this button. This update provides the corrected translation.

BZ#700305

Previously, when running firstboot in Japanese locale and the user attempted to continue without setting up an account, an untranslated warning message appeared. With this update, the message is properly translated into Japanese.

Security Fix

CVE-2011-2964

An input sanitization flaw was found in the foomatic-rip print filter. An attacker could submit a print job with the username, title, or job options set to appear as a command line option that caused the filter to use a specified PostScript printer description (PPD) file, rather than the administrator-set one. This could lead to arbitrary code execution with the privileges of the "lp" user.

Security Fix

CVE-2011-0226

A flaw was found in the way the FreeType font rendering engine processed certain PostScript Type 1 fonts. If a user loaded a specially-crafted font file with an application linked against FreeType, it could cause the application to crash or, possibly, execute arbitrary code with the privileges of the user running the application.

Security Fix

CVE-2011-3256

Multiple input validation flaws were found in the way FreeType processed bitmap font files. If a specially-crafted font file was loaded by an application linked against FreeType, it could cause the application to crash or, potentially, execute arbitrary code with the privileges of the user running the application.