Differences

This shows you the differences between two versions of the page.

--- linux_wiki:kernel_tuning [2015/04/28 22:24]
billdozor [Page Cache] added default values for dirty_ratios
+++ linux_wiki:kernel_tuning [2019/05/25 23:50] (current)
@@ Line 6: / Line 6: @@
 **Checklist**
-  * Any distro, but many examples are done via Red Hat based systems
+  * Distro(s): Any
 ----
-===== Hugepages =====
+====== Hugepages ======
 Problem: Unable to access a VM's console, and it shows out of memory errors on boot.
@@ Line 28: / Line 28: @@
 ----
-===== Swappiness =====
+==== HugePage Bug ====
+On some kernels, khugepaged can start running at 100% CPU utilization. This is typically seen on systems that have process/memory intensive processes.
+Other symptoms include:
+  * Normal commands hang (w, uptime, ps)
+  * Processes showing above 100% CPU (ie sometimes up to 1,300%)
+Details
+  * Known Affected kernels
+    * CentOS 6: 2.6.32-431.el6.x86_64
+  * Reference bug report: https://bugzilla.redhat.com/show_bug.cgi?id=879801
+=== Workaround Fix ===
+Disable hugepage defragmenting:
+<code bash>
+echo never > /sys/kernel/mm/redhat_transparent_hugepage/defrag
+</code>
+Add new cron entry
+<code bash>
+vim /etc/cron.d/hugepage_defrag
+# Disable kernel huge page defrag due to bugzilla bug: 879801
+@reboot root /bin/echo never > /sys/kernel/mm/transparent_hugepage/defrag
+</code>
+=== Permanent Fix ===
+Ultimately, the permanent fix is to update the kernel to a newer version.
+TODO: Will add confirmed kernel versions that this bug is fixed in.
+----
+====== Swappiness ======
 Swappiness controls how likely the kernel is to move processes out of memory and onto swap disk space.
 The setting is from 0 to 100.
-  * 0 = Avoid swapping for as long as possible
+  * 0 = Very aggressively avoid swapping for as long as possible
+    * High risk of OOM killing from memory and I/O pressure
   * 10 = Red Hat recommended for Oracle databases
   * **60 = Linux default**
@@ Line 56: / Line 93: @@
 ----
-===== Page Cache =====
+====== Page Cache ======
 Tuning how often cached pages in memory are flushed to disk.
@@ Line 69: / Line 106: @@
 vm.dirty_background_ratio = 3
 </code>
-Explanation: Start flushing pages to disk asynchronously when pagecache is equal to 3% of memory.
+  * Explanation: Start flushing pages to disk asynchronously when pagecache is equal to 3% of memory.
-Default value = 10
+  * Default value = 10
 Foreground (sync) pagecache flushing:
@@ Line 76: / Line 113: @@
 vm.dirty_ratio = 15
 </code>
-Explanation: Sync flush when pagecache is 15% of total memory. This blocks ANY process from using I/O when this happens. If this process takes longer than 120 seconds, kernel panics can happen.
+  * Explanation: Sync flush when pagecache is 15% of total memory. This blocks ANY process from using I/O when this happens. If this process takes longer than 120 seconds, kernel panics can happen.
-Default value = 20
+  * Default value = 20
 ----
-===== File Descriptors =====
+====== File Descriptors ======
 Viewing and setting the system wide file descriptors.
@@ Line 108: / Line 145: @@
 ----
+====== Kernel Panic ======
+Reboot a system after a 10 seconds of kernel panic
+<code bash>
+kernel.panic = 10
+</code>
+----