无法启动kdump

Question

我的系统总是崩溃。因此，我决定启用kdump来查看问题，因为我无法在日志文件上看到可能的错误。

我跟踪了从此站点设置kdump的步骤。我的服务器运行在CentOS 5.8和16 on内存上。下面是我为配置kdump所执行的步骤：

1. Install kexec-tools, `yum install kexec-tools` and follow the installation steps
2. Edit the /boot/grub/grub.conf to configure the kdump memory usage
3. Edit the /etc/kdump.cof to configure the target type to /var/crash/ and core_collector
4. Enable kdump through `chkconfig kdump on`.
5. Reboot the server

当我运行service kdump status时，上面写着Kdump is not operational。我该怎么做才能让kdump开始运作。我错过什么配置了吗？下面是我的/boot/grub/grub.conf和/etc/kdump.conf的内容。

以下是文件/boot/grub/grub.conf的内容

# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE:  You have a /boot partition.  This means that
#          all kernel and initrd paths are relative to /boot/, eg.
#          root (hd0,0)
#          kernel /vmlinuz-version ro root=/dev/sda3
#          initrd /initrd-version.img
#boot=/dev/sda
default=0
timeout=5
splashimage=(hd0,0)/grub/splash.xpm.gz
hiddenmenu
title CentOS (2.6.18-308.el5)
        root (hd0,0)
        kernel /vmlinuz-2.6.18-308.el5 ro root=LABEL=/
crashkernel=128M
        initrd /initrd-2.6.18-308.el5.img

下面是文件/etc/kdump.conf的内容

# Configures where to put the kdump /proc/vmcore files
#
# This file contains a series of commands to perform (in order) when a
# kernel crash has happened and the kdump kernel has been loaded.  Directives in
# this file are only applicable to the kdump initramfs, and have no effect if
# the root filesystem is mounted and the normal init scripts are processed
#
# Currently only one dump target and path may be configured at once
# if the configured dump target fails, the default action will be preformed
# the default action may be configured with the default directive below.  If the
# configured dump target succedes
#
# For filesystem based dump, it's recommended to use UUID and LABEL
# instead of device name in dump target.
#
# See the kdump.conf(5) man page for details of configuration directives

#raw /dev/sda5
#ext3 /dev/sda3
#ext3 LABEL=/boot
#ext3 UUID=03138356-5e61-4ab3-b58e-27507ac41937
#net my.server.com:/export/tmp
#net user@my.server.com
path /var/crash
core_collector makedumpfile -c --message-level 1
#core_collector cp --sparse=always
#link_delay 60
#kdump_post /var/crash/scripts/kdump-post.sh
#extra_bins /usr/bin/lftp
#disk_timeout 30
#extra_modules gfs2
#options modulename options
#default shell
#sshkey /root/.ssh/kdump_id_rsa

我还注意到，我的/boot/grub/grub.conf文件与教程中的示例grub.conf文件不同。它们有两条不同之处：

From tutorial
   kernel /vmlinuz-2.6.32-220.el6.x86_64 ro root=/dev/sda3
   initrd /initramfs-2.6.32-220.el6.x86_64.img

From own conf 
   kernel /vmlinuz-2.6.18-308.el5 ro root=LABEL=/
   initrd /initrd-2.6.18-308.el5.img 

这些行会不会导致kdump不启动？

/var/log/消息的编辑1内容

    Feb 25 02:18:28 61540 kernel: Command line: ro root=LABEL=/ crashkernel=128M
    Feb 25 02:18:28 61540 kernel: BIOS-provided physical RAM map:
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 0000000000010000 - 000000000009a000 (usable)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 000000000009f800 - 00000000000a0000 (reserved)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 0000000000100000 - 00000000cfda0000 (usable)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000cfda0000 - 00000000cfdd1000 (ACPI NVS)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000cfdd1000 - 00000000cfe00000 (ACPI data)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000cfe00000 - 00000000cff00000 (reserved)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 00000000fec00000 - 0000000100000000 (reserved)
    Feb 25 02:18:28 61540 kernel:  BIOS-e820: 0000000100000000 - 000000042f000000 (usable)
    Feb 25 02:18:28 61540 kernel: DMI 2.4 present.
    Feb 25 02:18:28 61540 kernel: No NUMA configuration found
    Feb 25 02:18:28 61540 kernel: Faking a node at 0000000000000000-000000042f000000
    Feb 25 02:18:28 61540 kernel: Bootmem setup node 0 0000000000000000-000000042f000000
    Feb 25 02:18:28 61540 kernel: Memory for crash kernel (0x0 to 0x0) notwithin permissible range
    Feb 25 02:18:28 61540 kernel: disabling kdump
    Feb 25 02:44:39 61540 kdump: No crashkernel parameter was specified or crashkernel memory reservation failed
    Feb 25 02:44:39 61540 kdump: failed to start up

编辑2，还是应该将ro root=LABEL=代码更改为ro root=/dev/sda3 3？

Answer 1

title CentOS (2.6.18-308.el5)
        root (hd0,0)
        kernel /vmlinuz-2.6.18-308.el5 ro root=LABEL=/
crashkernel=128M
        initrd /initrd-2.6.18-308.el5.img

看起来您将crashkernel参数放入一个新行中。这就是Kdump is not operational消息产生的原因。所有内核参数必须放在与kernel相同的行上：

title CentOS (2.6.18-308.el5)
        root (hd0,0)
        kernel /vmlinuz-2.6.18-308.el5 ro root=LABEL=/ crashkernel=128M
        initrd /initrd-2.6.18-308.el5.img

重新启动后，查看一下/var/log/messages，您将看到如下所示：

localhost kdump: kexec: loaded kdump kernel
localhost kdump: started up

以及：

# /etc/init.d/kdump start
Starting kdump:                                            [  OK  ]
# /etc/init.d/kdump status
Kdump is operational

kdump: No crashkernel parameter was specified or crashkernel memory reservation failed
kdump: failed to start up

根据这文档，尝试如下：

crashkernel=128M@16M

Answer 2

你的核太小了。试试256米，384米，甚至512米。消息“crashkernel的内存.不在允许范围内”告诉您这一点。