Centos7/RHEL7 开启kdump

Kdump是一种基于kexec的Linux内核崩溃捕获机制，简单来说系统启动时会预留一块内存，当系统崩溃调用命令kexec(kdump kernel)在预留的内存中启动kdump内核，

该内核会将此时内存中的所有运行状态和数据信息收集到一个coredump文件中以便后续分析调试。

本文介绍如何在Centos7/RHEL7 开启kdump。

步骤1 安装kexec-tools

[vagrant@localhost ~]$ yum install kexec-tools

步骤2 设置crashkernel预留内存大小

    [vagrant@localhost ~]$ cat /etc/default/grub
    GRUB_TIMEOUT=5
    GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
    GRUB_DEFAULT=saved
    GRUB_DISABLE_SUBMENU=true
    GRUB_TERMINAL_OUTPUT="console"
    GRUB_CMDLINE_LINUX="crashkernel=256M rd.lvm.lv=centos/root rd.lvm.lv=centos/swap rhgb quiet"
    GRUB_DISABLE_RECOVERY="true"

修改crashkernel的大小，我的系统内存是1G，保留了256M，注意预留内存大小，过小会导致生成coredump文件失败（不知道设置多少时，可以尝试每次增加128M）

修改后还需重新生成grub配置文件，重启系统才能生效

    [vagrant@localhost ~]$ grub2-mkconfig -o /boot/grub2/grub.cfg
    [vagrant@localhost ~]$ reboot

步骤3 修改kdump默认配置/etc/kdump.conf

centos7 默认已安装kdump，根据需要修改默认配置

    [vagrant@localhost ~]$ vi /etc/kdump.conf
    #指定coredump文件存储位置
    path /var/crash
    #增加-c参数，代表压缩coredump文件
    core_collector makedumpfile -c -l --message-level 1 -d 31
    #生成coredump后，重启系统，
    default reboot

步骤4 开启kdump服务

    [vagrant@localhost ~]$ systemctl start kdump.service //启动kdump
    [vagrant@localhost ~]$ systemctl enable kdump.service //设置开机启动

步骤5 测试kdump功能

检查kdump是否开启成功

    [vagrant@localhost ~]$ service kdump status
    Redirecting to /bin/systemctl status kdump.service
    ● kdump.service - Crash recovery kernel arming
    Loaded: loaded (/usr/lib/systemd/system/kdump.service; enabled; vendor preset: enabled)
    Active: active (exited) since Mon 2017-12-18 09:12:56 UTC; 43min ago
    Process: 913 ExecStart=/usr/bin/kdumpctl start (code=exited, status=0/SUCCESS)
    Main PID: 913 (code=exited, status=0/SUCCESS)
    CGroup: /system.slice/kdump.service
    [vagrant@localhost ~]$ systemctl is-active kdump.service
    active

手动触发crush

[root@cloud ~]# echo 1 > /proc/sys/kernel/sysrq ; echo c > /proc/sysrq-trigger

如果没有问题，系统会自动重启，重启后可以看到在/var/crash/目录下生成了coredump文件

    [vagrant@localhost ~]$ ls /var/crash/
    127.0.0.1-2017-12-18-08:25:11

步骤6 安装crash，分析coredump文件

[vagrant@localhost ~]$ yum install crash

[vagrant@localhost ~]$ crash /var/crash/127.0.0.1-2017-12-18-08\:25\:11/vmcore /usr/src/kernels/linux-`uname -r`/vmlinux

输入问号”？“，查看crash支持的命令

    crash> ?

    * files mach repeat timer
    alias foreach mod runq tree
    ascii fuser mount search union
    bt gdb net set vm
    btop help p sig vtop
    dev ipcs ps struct waitq
    dis irq pte swap whatis
    eval kmem ptob sym wr
    exit list ptov sys q
    extend log rd task

比如log命令，可以看到系统crash时的打印信息

    [ 470.018926] CPU: 0 PID: 1490 Comm: iwconfig Tainted: GF O 3.10.69 #1
    [ 470.019251] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
    [ 470.019603] task: ffff880039406ae0 ti: ffff880035d1c000 task.ti: ffff880035d1c000
    [ 470.019931] RIP: 0010:[<ffffffff8129d3cd>] [<ffffffff8129d3cd>] memcpy+0xd/0x110
    [ 470.020419] RSP: 0018:ffff880035d1dc38 EFLAGS: 00010206
    [ 470.020697] RAX: ffff88003dba3508 RBX: ffff8800362eb400 RCX: 1ffffffffffffffc
    [ 470.021218] RDX: 0000000000000006 RSI: 000000000000001a RDI: ffff88003dba3508
    [ 470.021735] RBP: ffff880035d1dc88 R08: 0000000000016320 R09: ffff88003dba34c0
    [ 470.022917] R10: 0000000000000000 R11: ffff880035d1d9c6 R12: ffff8800341a3280
    [ 470.023437] R13: ffffc90000226104 R14: ffffffffffffffe2 R15: ffff88003dba34c0
    [ 470.023958] FS: 00007ff662ec1740(0000) GS:ffff88003fc00000(0000) knlGS:0000000000000000
    [ 470.024725] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
    [ 470.025223] CR2: 000000000000001a CR3: 0000000038f6a000 CR4: 00000000000406f0
    [ 470.025775] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
    [ 470.026334] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400