此内容没有您所选择的语言版本。

2. Release Notes for x86_64


  • This release includes WBEMSMT, a suite of web-based applications that provides a user-friendly management interface for Samba and DNS. For more information about WBEMSMT, refer to http://sblim.wiki.sourceforge.net/.
  • Upgrading pm-utils from a Red Hat Enterprise Linux 5.1 Beta version of pm-utils will fail, resulting in the following error:
    error: unpacking of archive failed on file /etc/pm/sleep.d: cpio: rename
    
    To prevent this from occurring, delete the /etc/pm/sleep.d/ directory prior to upgrading. If /etc/pm/sleep.d contains any files, you can move those files to /etc/pm/hooks/.
  • Hardware testing for the Mellanox MT25204 has revealed that an internal error occurs under certain high-load conditions. When the ib_mthca driver reports a catastrophic error on this hardware, it is usually related to an insufficient completion queue depth relative to the number of outstanding work requests generated by the user application.
    Although the driver will reset the hardware and recover from such an event, all existing connections are lost at the time of the error. This generally results in a segmentation fault in the user application. Further, if opensm is running at the time the error occurs, then it will have to be manually restarted in order to resume proper operation.
  • Driver Update Disks now support Red Hat's Driver Update Program RPM-based packaging. If a driver disk uses the newer format, it is possible to include RPM packaged drivers that will be preserved across system updates.
    Please note that driver RPMs are copied only for the default kernel variant that is in use on the installed system. For example, installing a driver RPM on a system running the virtualized kernel will install the driver only for the virtualized kernel. The driver RPM will not be installed for any other installed kernel variant in the system.
    As such, on a system that has multiple kernel variants installed, you will need to boot the system on each kernel variant and install the driver RPM. For example, if your system has both bare-metal and virtualized kernels installed, boot your system using the bare-metal kernel and install the driver RPM. Then, reboot the system into the virtualized kernel and install the driver RPM again.
  • During the lifetime of dom0, you cannot create guests (i.e. xm create) more than 32,750 times. For example, if you have guests rebooting in a loop, dom0 will fail to boot any guest after rebooting guests a total of 32,750 times.
    If this event occurs, restart dom0
  • Using the divider command-line argument may not work properly in this architecture. This issue will be addressed in a future update of Red Hat Enterprise Linux 5.
  • The Red Hat Enterprise Linux 5.1 NFS server now supports referral exports. These exports are based on extensions to the NFSv4 protocol. Any NFS clients that do not support these extensions (namely, Red Hat Enterprise Linux releases prior to 5.1) will not be able to access these exports.
    As such, if an NFS client does not support these exports, any attempt to access these exports may fail with an I/O error. In some cases, depending on the client implementation, the failure may be more severe, including the possibility of a system crash.
    It is important that you take precautions to ensure that NFS referral exports are not accessed by clients that do not support them.
  • GFS2 is an incremental advancement of GFS. This update applies several significant improvements that require a change to the on-disk file system format. GFS file systems can be converted to GFS2 using the utility gfs2_convert, which updates the metadata of a GFS file system accordingly.
    While much improved since its introduction in Red Hat Enterprise Linux 5, GFS2 remains a Technology Preview. The release notes included in the distribution incorrectly states that GFS2 is fully supported. Nevertheless, benchmark tests indicate faster performance on the following:
    • heavy usage in a single directory and faster directory scans (Postmark benchmark)
    • synchronous I/O operations (fstest benchmark test indicates improved performance for messaging applications like TIBCO)
    • cached reads, as there is no longer any locking overhead
    • direct I/O to preallocated files
    • NFS file handle lookups
    • df, as allocation information is now cached
    In addition, GFS2 also features the following changes:
    • journals are now plain (though hidden) files instead of metadata. Journals can now be dynamically added as additional servers mount a file system.
    • quotas are now enabled and disabled by the mount option quota=<on|off|account>
    • quiesce is no longer needed on a cluster to replay journals for failure recovery
    • nanosecond timestamps are now supported
    • similar to ext3, GFS2 now supports the data=ordered mode
    • attribute settings lsattr() and chattr() are now supported via standard ioctl()
    • file system sizes above 16TB are now supported
    • GFS2 is a standard file system, and can be used in non-clustered configurations
  • Installing Red Hat Enterprise Linux 5.1 on HP BL860c blade systems may hang during the IP information request stage. This issue manifests when you have to select OK twice on the Configure TCP/IP screen.
    If this occurs, reboot and perform the installation with Ethernet autonegotiation disabled. To do this, use the parameter ethtool="autoneg=off" when booting from the installation media. Doing so does not affect the final installed system.
  • The nohide export option is required on referral exports (i.e. exports that specify a referral server). This is because referral exports need to "cross over" a bound mount point. The nohide export option is required for such a "cross over" to be successful.
    For more information on bound mounts, refer to man exports 5.
  • This update includes the lvm2 event monitoring daemon. If you are already using lvm2 mirroring, perform the following steps to ensure that all monitoring functions are upgraded properly:
    1. Deactivate all mirrored lvm2 logical volumes before updating. To do this, use the command lvchange -a n <volume group or mirrored volume>.
    2. Stop the old lvm2 event daemon using killall -HUP dmeventd.
    3. Perform the upgrade of all related RPM packages, namely device-mapper and lvm2.
    4. Reactivate all mirrored volumes again using lvchange -a y <volume group or mirrored volume>.
  • Rapid Virtualization Indexing (RVI) is now supported on 64-bit, 32-bit, and 32-bit PAE kernels. However, RVI can only translate 32-bit guest virtual addresses on the 32-bit PAE hypervisor.
    As such, if a guest is running a PAE kernel with more than 3840MB of RAM, a wrong address translation error will occur. This can crash the guest.
    It is recommended that you use the 64-bit kernel if you intend to run guests with more than 4GB of physical RAM under RVI.
  • Running 16 cores or more using AMD Rev F processors may result in system resets when performing fully-virtualized guest installations.
  • Installing the systemtap-runtime package will result in a transaction check error if the systemtap package is already installed. Further, upgrading Red Hat Enterprise Linux 5 to 5.1 will also fail if the systemtap package is already installed.
    As such, remove the systemtap package using the command rpm -e systemtap-0.5.12-1.e15 before installing systemtap-runtime or performing an upgrade.
  • When setting up NFSROOT, BOOTPROTO must be set as BOOTPROTO=dhcp in /etc/sysconfig/network-scripts/ifcfg-eth0.
    If your environment requires a different setting for BOOTPROTO, then temporarily set BOOTPROTO=dhcp in /etc/sysconfig/network-scripts/ifcfg-eth0 before initially creating the initrd. You can reset the original value of BOOTPROTO after the initrd is created.
  • Crash dumping through kexec and kdump may not function reliably with HP Smart Array controllers. Note that these controllers use the cciss driver.
    A solution to this problem, which is likely to involve a firmware update to the controller, is being investigated.
  • Creating a guest more than 1,000 times will cause dom0 to reboot suddenly. It is recommended that you pre-empt this by rebooting dom0 before the domain ID of any guest reaches 1,000.
  • The QLogic iSCSI Expansion Card for the IBM Bladecenter provides both ethernet and iSCSI functions. Some parts on the card are shared by both functions. However, the current qla3xxx and qla4xxx drivers support ethernet and iSCSI functions individually. Both drivers do not support the use of ethernet and iSCSI functions simultaneously.
    As such, using both ethernet and iSCSI functions simultaneously may hang the device. This could result in data loss and filesystem corruption on iSCSI devices, or network disruptions on other connected ethernet devices.
  • When using virt-manager to add disks to an existing guest, duplicate entries may be created in the guest's /etc/xen/<domain name> configuration file. These duplicate entries will prevent the guest from booting.
    As such, you should remove these duplicate entries.
  • Repeatedly migrating a guest between two hosts may cause one host to panic. If a host is rebooted after migrating a guest out of the system and before migrating the same guest back, the panic will not occur.
  • sysreport is being deprecated in favor of sos. To install sos, run yum install sos. This command installs sos and removes sysreport. It is recommended that you update any existing kickstart files to reflect this.
    After installing sos, use the command sosreport to invoke it. Using the command sysreport generates a warning that sysreport is now deprecated; continuing will invoke sosreport.
    If you need to use the sysreport tool specifically, use the command sysreport.legacy to invoke it.
    For more information about sosreport, refer to man sosreport and sosreport --help.
Red Hat logoGithubRedditYoutubeTwitter

学习

尝试、购买和销售

社区

关于红帽文档

通过我们的产品和服务,以及可以信赖的内容,帮助红帽用户创新并实现他们的目标。

让开源更具包容性

红帽致力于替换我们的代码、文档和 Web 属性中存在问题的语言。欲了解更多详情,请参阅红帽博客.

關於紅帽

我们提供强化的解决方案,使企业能够更轻松地跨平台和环境(从核心数据中心到网络边缘)工作。

© 2024 Red Hat, Inc.