1KDUMP.CONF(5) File Formats Manual KDUMP.CONF(5)
2
6 kdump.conf - configuration file for kdump kernel.
7
10 kdump.conf is a configuration file for the kdump kernel crash collec‐
11 tion service.
12 kdump.conf provides post-kexec instructions to the kdump kernel. It is
14 stored in the initrd file managed by the kdump service. If you change
15 this file and do not want to restart before it takes effect, restart
16 the kdump service to rebuild to initrd.
17 For most configurations, you can simply review the examples provided in
19 the stock /etc/kdump.conf.
20 NOTE: kdump.conf only affects the behavior of the initramfs. Please
22 read the kdump operational flow section of kexec-kdump-howto.txt in the
23 docs to better understand how this configuration file affects the
24 behavior of kdump.
25
28 path <path>
29 Append path to the filesystem device which you are dumping to.
30 Ignored for raw device dumps. If unset, will default to
31 /var/crash.
32 core_collector <command> <options>
34 This allows you to specify the command to copy the vmcore. You
35 could use the dump filtering program makedumpfile, the default
36 one, to retrieve your core, which on some arches can drastically
37 reduce core file size. See /usr/sbin/makedumpfile --help for a
38 list of options. Note that the -i and -g options are not needed
39 here, as the initrd will automatically be populated with a con‐
40 fig file appropriate for the running kernel.
41 If you use other command as a core_collector, probably you also
43 need to provide its full path in extra_bins. Note, for ssh dump,
44 scp should be used instead of cp.
45 raw <partition>
47 Will write /proc/vmcore into raw <partition>.
48 nfs <nfs mount> | nfs4 <nfs mount> | net <nfs mount>
50 Will mount fs and copy /proc/vmcore to
51 <mnt>/<path>/%HOST-%DATE/, supports DNS. Note that a fqdn should
52 be used as the server name in the mount point. By default, NFSv3
53 is used, nfs4 will use NFSv4. "net" is deprecated by "nfs"
54 ssh <user@server> | net <user@server>
56 Will copy /proc/vmcore to <user@server>:<path>/%HOST-%DATE/ via
57 SSH, supports DNS. If makedumpfile is the core_collector, it is
58 piped to an "ssh" shell, otherwise use the specified core_col‐
59 lector like scp. NOTE: make sure user has necessary write per‐
60 missions on server and that a fqdn is used as the server name.
61 "net" is deprecated by "ssh"
62 <fs type> <partition>
64 Will mount -t <fs type> <partition> /mnt and copy /proc/vmcore
65 to /mnt/<path>/127.0.0.1-%DATE/. Currently, the supported fs
66 types are: ext4, ext3, ext2, btrfs and xfs, other fs types like
67 "auto" are not supported. NOTE: <partition> can be a device
68 node, label or uuid.
69 disk_timeout <seconds>
71 Number of seconds to wait for disks to appear prior to continue
72 to save dump. By default kdump waits 180 seconds for the disks
73 to show up it needs. This can be useful in some cases if disk
74 never shows up (Either because disk was removed or because kdump
75 is waiting on wrong disk).
76 link_delay <seconds>
79 Some network cards take a long time to initialize, and some
80 spanning tree enabled networks do not transmit user traffic for
81 long periods after a link state changes. This optional parame‐
82 ter defines a wait period after a link is activated in which the
83 initramfs will wait before attempting to transmit user data.
84 kdump_post <binary | script>
86 This directive allows you to run a specified executable just
87 after the memory dump process terminates. The exit status from
88 the dump process is fed to the kdump_post executable, which can
89 be used to trigger different actions for success or failure.
90 Note that scripts written for use with this directive must use
92 the /bin/hush interpreter
93 kdump_pre <binary | script>
95 Works just like the kdump_post directive, but instead of running
96 after the dump process, runs immediately before. Exit status of
97 this binary is interpreted as follows:
98 0 - continue with dump process as usual
100 non 0 - reboot/halt the system
102 Note that scripts written for this directive must use the
104 /bin/hush interpreter
105 extra_bins <binaries | shell scripts>
107 This directive allows you to specify additional binaries or
108 shell scripts you'd like to include in your kdump initrd. Gener‐
109 ally only useful in conjunction with a kdump_post binary or
110 script that relies on other binaries or scripts.
111 extra_modules <module(s)>
113 This directive allows you to specify extra kernel modules that
114 you want to be loaded in the kdump initrd, typically used to set
115 up access to non-boot-path dump targets that might otherwise not
116 be accessible in the kdump environment. Multiple modules can be
117 listed, separated by a space, and any dependent modules will
118 automatically be included. Module name should be specified
119 without ".ko" suffix.
120 options <module> <option list>
122 This directive allows you to specify options to apply to modules
123 in the initramfs. This directive overrides options specified in
124 /etc/modprobe.conf. Module name should be specified without
125 ".ko" suffix.
126 blacklist <module>
128 The blacklist keyword indicates that all of that particular mod‐
129 ules are to be ignored in the initramfs. General terminology for
130 blacklist has been that module is present in initramfs but it is
131 not actually loaded in kernel. This directive can be specified
132 multiple times or as a space separated list. Module name should
133 be specified without ".ko" suffix.
134 sshkey <path>
136 Specifies the path of the ssh key you want to use when do ssh
137 dump, the default value is /root/.ssh/kdump_id_rsa.
138 default <reboot | halt | poweroff | shell | mount_root_run_init>
140 Action to perform in case dumping to intended target fails. If
141 no default action is specified, "reboot" is assumed default.
142 reboot: If the default action is reboot simply reboot the system
144 (this is what most people will want, as it returns the system to
145 a nominal state).
146 halt: bring the system to a halt, requiring manual reset.
148 poweroff: The system will be powered down.
150 shell: If the default action is shell, then drop to an hush ses‐
152 sion inside the initramfs from where you can manually perform
153 additional recovery actions. Exiting this shell reboots the
154 system.
155 mount_root_run_init: Mount root filesystem and run init. Kdump
157 initscript will try to save dump to root filesystem in
158 /var/crash dir. This will likely require a lot more memory to be
159 reserved for kdump kernel.
160 debug_mem_level <0-3>
162 Turns on debug/verbose output of kdump scripts regarding
163 free/used memory at various points of execution. Higher level
164 means more debugging output.
165 0 - no output
167 1 - partial /proc/meminfo
169 2 - /proc/meminfo
171 3 - /proc/meminfo + /proc/slabinfo
173 force_rebuild <0 | 1>
175 By default, kdump initrd only will be rebuilt when necessary.
176 Specify 1 here to force rebuilding kdump initrd every time when
177 kdump service starts.
178 fence_kdump_args <arg(s)>
180 Command line arguments for fence_kdump_send (it can contain all
181 valid arguments except hosts to send notification to).
182 fence_kdump_nodes <node(s)>
184 List of cluster node(s) separated by space to send fence_kdump
185 notification to (this option is mandatory to enable
186 fence_kdump).
187
190 see /etc/kdump.conf
191
194 kexec(8) mkdumprd(8)
195kexec-tools 07/23/2008 KDUMP.CONF(5)