[WIP]:DOCTEAM-2062- kdump#628
Conversation
|
SLES-kdump_en.pdf latest |
|
SLES-kdump_en.pdf |
lvicoun
left a comment
There was a problem hiding this comment.
Hi Amrita,
please see my suggestion. Thanks!
| <step><para>Enable and immediately start the &kdump; service:</para> | ||
| <screen>&prompt.sudo; systemctl enable --now kdump </screen> | ||
| <para>When the systemctl command triggers the main <literal>kdump.service</literal>, it automatically calls a helper service named <literal>kdump-commandline.service</literal> before attempting to load the kernel. | ||
| This helper service verifies if the current kernel command line contains the expected <literal>crashkernel= options</literal>. If they are missing, automatically updates the boot loader configuration to apply the settings on the next boot sequence.</para> |
There was a problem hiding this comment.
| This helper service verifies if the current kernel command line contains the expected <literal>crashkernel= options</literal>. If they are missing, automatically updates the boot loader configuration to apply the settings on the next boot sequence.</para> | |
| This helper service verifies if the current kernel command line contains the expected <literal>crashkernel= options</literal>. If the options are missing, the helper service automatically modifies the boot loader configuration to contain the crashkernel options to be applied on the next boot.</para> |
| <listitem><para><emphasis role="bold">VMCOREINFO:</emphasis>special section that tells analysis tools how the kernel's memory was laid out so they can make sense of the raw data.</para></listitem> | ||
| </itemizedlist> | ||
| </section> | ||
| <section xml:id="kexec-kdump"> |
There was a problem hiding this comment.
I'd move it before the vmcore file section. The vmcore is though very briefly described in the beginning, but the vmcore section mentions kexec that is not defined anywhere before you use it.
| <para>Any modifications made to this configuration file require a manual restart of the &kdump; service to take effect. However, if any changes are made specifically to the <literal>KDUMP_CRASHKERNEL</literal> memory allocation size, a full system reboot is required because these settings alter the system's core boot parameters.</para> | ||
| </section> | ||
| <section xml:id="kdump-transact"> | ||
| <title>Configuring &kdump; for transactional systems </title> |
There was a problem hiding this comment.
| <title>Configuring &kdump; for transactional systems </title> | |
| <title>Configuring &kdump; on immutable systems </title> |
There was a problem hiding this comment.
It is officially called immutable system. And the immutable flavor is not available for SLES 4 SAP, so this needs to be profiled for SLES only.
| </section> | ||
| <section xml:id="kdump-transact"> | ||
| <title>Configuring &kdump; for transactional systems </title> | ||
| <para>On a transactional (read-only) file system, the standard <literal>kdump-commandline.service</literal> is unable to directly modify the boot loader configuration. |
There was a problem hiding this comment.
| <para>On a transactional (read-only) file system, the standard <literal>kdump-commandline.service</literal> is unable to directly modify the boot loader configuration. | |
| <para>On an immutable file system, the standard <literal>kdump-commandline.service</literal> is unable to directly modify the boot loader configuration. |
| To specify custom memory sizes directly on a transactional system, you can append the crashkernel parameter to the command by running:</para> | ||
| <screen>&prompt.sudo; transactional-update setup-kdump [--crashkernel=low,high]</screen> | ||
| <para><emphasis role="bold">Other configurations:</emphasis></para> | ||
| <para>Just as with non-transactional systems, modifying the <filename>/etc/sysconfig/kdump </filename> configuration file on a transactional system requires the &kdump; service to be manually restarted for the changes to take effect. |
There was a problem hiding this comment.
You actually need to reboot the system after you run the tr-up command to boot into the newly created snapshot, unlike in regular SLES. So you need to run the tr-up command and reboot, then there is no need to manually restart the service.
| <listitem><para> Install the <package>openssh-clients</package> package to store over SSH.</para></listitem> | ||
| </itemizedlist> | ||
| </listitem> | ||
| <listitem><para>The <literal>mailx</literal> utility to enable sending notifications when a crash dump is saved. </para></listitem> |
There was a problem hiding this comment.
| <listitem><para>The <literal>mailx</literal> utility to enable sending notifications when a crash dump is saved. </para></listitem> | |
| <para>To enable sending notifications when a crash dump is saved, install the <literal>mailx</literal> package. </para> |
| </itemizedlist> | ||
| </listitem> | ||
| <listitem><para>The <literal>mailx</literal> utility to enable sending notifications when a crash dump is saved. </para></listitem> | ||
| </itemizedlist> |
There was a problem hiding this comment.
| </itemizedlist> |
| Ensure no critical workloads are running and no unsaved data is present on the system. Additionally, ensure to | ||
| sync and unmount file systems:</para> | ||
| <screen>echo s > /proc/sysrq-trigger</screen> | ||
| <screen>echo u > /proc/sysrq-trigger </screen> |
There was a problem hiding this comment.
What each command does? An introduction text for each one is really appreciated.
| <screen>echo u > /proc/sysrq-trigger </screen> | ||
| <para>Then you can simulate a kernel crash:</para> | ||
| <screen>echo c > /proc/sysrq-trigger</screen> | ||
| <para>Verify by checking if there is a new directory created under your <literal>KDUMP_SAVEDIR</literal> which is <filename>/var/crash</filename> by default. This contains the dmesg and |
There was a problem hiding this comment.
| <para>Verify by checking if there is a new directory created under your <literal>KDUMP_SAVEDIR</literal> which is <filename>/var/crash</filename> by default. This contains the dmesg and | |
| <para>Check if there is a new directory created under your <literal>KDUMP_SAVEDIR</literal> which is <filename>/var/crash</filename> by default. This contains the dmesg and |
| <para>During &kdump;, if you see error messages mentioning low memory and invoking the | ||
| Out of Memory (OOM) killer, this is the likely cause. In case, you don't see such messages, trying with increased crash kernel reservation is a good | ||
| first step.</para> | ||
| <para>The recommended ways to rectify this are:</para> |
There was a problem hiding this comment.
I think here it should be steps not ways and then use a procedure not an orderedlist.
PR creator: Description
Describe the overall goals of this pull request.
PR creator: Are there any relevant issues/feature requests?
PR reviewer: Checklist for editorial review
Apart from the usual checks, please double-check also the following: