starlingx/docs
Code Issues Proposed changes

Additional details on file system sizing for core dump settings (r8, dsR8)

Browse Source 
Add details and example to log fs prereqs

Closes-bug: 2052620

Change-Id: I2881fd4f5a85deab5b3382cdddbe98aaa265531e
Signed-off-by: Ron Stone <ronald.stone@windriver.com>
This commit is contained in:
Ron Stone 2024-02-07 11:48:34 +00:00
parent be17ea7643
commit 2f0404ddc8
1 changed files with 21 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
doc/source/system_configuration/kubernetes/change-the-default-coredump-configuration-51ff4ce0c9ae.rst
View File

@ -38,7 +38,8 @@ The editable parameters and their defaults are as follows:
Minimum: 1G
Maximum values for each configurable coredump parameter depend on system capacity.
Maximum values for each configurable coredump parameter depends on system
capacity.
The parameters accept integer/float values followed by a letter representing
the unit of measurement.
@ -72,11 +73,28 @@ Ensure that you have sufficient storage available on the host's ``log``
filesystem. See :ref:`resizing-filesystems-on-a-host` for more information
about adjusting it's size.
The *log* (``/var/log``) filesystem size is defaulted for standard platform
operation. If there is an expected usage increase due to the application core
dump files, it should be increased proportionally to the core dump size and the
expected number of core dumps that need to be retained. The core dump storage
limits can be explicitly configured using the system ``max_use`` parameter
settings that aligns to the core dump storage and retention requirements.
For example:
Filesystem Setting:
``8G (default) + 10G (core dump size x retention count - 5G x 2) = 18G``
Core Dump Setting:
``max_use = 10G (core dump size x retention count - 5G x 2)``
.. rubric:: |eg|
When you configure a parameter, it will be replicated to the ``coredump.conf`` file of all
existing nodes (controllers, workers, storages).
When you configure a parameter, it will be replicated to the ``coredump.conf``
file of all existing nodes (controllers, workers, storages).
* To add a coredump service parameter: