My Oracle Support Banner

Unusual Low Run Queue ("load average") Number in Some UEK4 Kernels (Doc ID 2403890.1)

Last updated on FEBRUARY 26, 2020

Applies to:

Oracle Linux Cloud Service - Version Oracle Linux 6.0 and later
Linux x86-64

Symptoms

When you run/switch between different UEK4 kernels say "4.1.12-103.3.8.el6uek.x86_64" and "4.1.12-112.14.15.el6uek.x86_64" in same Linux OS, you notice that the lower kernel manifests a very low run queue number.

 

"4.1.12-103.3.8.el6uek.x86_64" with low load avg,
---------------------------------------

top - 09:10:55 up 12 days, 19:36, 0 users, load average: 1.17, 1.46, 1.59
Tasks: 4342 total, 56 running, 4285 sleeping, 0 stopped, 1 zombie
Cpu(s): 32.1%us, 23.0%sy, 0.0%ni, 44.0%id, 0.3%wa, 0.0%hi, 0.6%si, 0.0%st
Mem: 1583612860k total, 836440364k used, 747172496k free, 586296k buffers
Swap: 66322428k total, 0k used, 66322428k free, 12332616k cached

top - 09:11:09 up 12 days, 19:36, 0 users, load average: 1.06, 1.42, 1.58
Tasks: 4351 total, 43 running, 4306 sleeping, 0 stopped, 2 zombie
Cpu(s): 31.7%us, 24.2%sy, 0.0%ni, 43.1%id, 0.3%wa, 0.0%hi, 0.8%si, 0.0%st
Mem: 1583612860k total, 836489432k used, 747123428k free, 586296k buffers
Swap: 66322428k total, 0k used, 66322428k free, 12334888k cached

top - 09:11:22 up 12 days, 19:36, 0 users, load average: 1.28, 1.45, 1.59
Tasks: 4350 total, 59 running, 4290 sleeping, 0 stopped, 1 zombie
Cpu(s): 31.0%us, 24.6%sy, 0.0%ni, 43.5%id, 0.2%wa, 0.0%hi, 0.6%si, 0.0%st
Mem: 1583612860k total, 836478476k used, 747134384k free, 586296k buffers
Swap: 66322428k total, 0k used, 66322428k free, 12337248k cached

 

"4.1.12-112.14.15.el6uek.x86_64"l with higher load avg,
---------------------------------------

top - 12:00:29 up 43 min, 1 user, load average: 59.56, 56.01, 47.39
Tasks: 4360 total, 44 running, 4315 sleeping, 0 stopped, 1 zombie
Cpu(s): 32.5%us, 24.3%sy, 0.0%ni, 42.3%id, 0.3%wa, 0.0%hi, 0.7%si, 0.0%st
Mem: 1583612056k total, 824170420k used, 759441636k free, 142004k buffers
Swap: 66322428k total, 0k used, 66322428k free, 1824712k cached

top - 12:00:39 up 43 min, 1 user, load average: 57.85, 55.77, 47.40
Tasks: 4359 total, 50 running, 4308 sleeping, 0 stopped, 1 zombie
Cpu(s): 31.3%us, 23.0%sy, 0.0%ni, 44.8%id, 0.2%wa, 0.0%hi, 0.7%si, 0.0%st
Mem: 1583612056k total, 824169608k used, 759442448k free, 142052k buffers
Swap: 66322428k total, 0k used, 66322428k free, 1826620k cached

top - 12:00:49 up 43 min, 1 user, load average: 57.03, 55.66, 47.45
Tasks: 4356 total, 70 running, 4285 sleeping, 0 stopped, 1 zombie
Cpu(s): 31.9%us, 25.2%sy, 0.0%ni, 41.7%id, 0.3%wa, 0.0%hi, 0.8%si, 0.0%st
Mem: 1583612056k total, 824174544k used, 759437512k free, 142096k buffers
Swap: 66322428k total, 0k used, 66322428k free, 1828980k cached

 

You might have noticed that the CPU usage, memory usage, even number of running processes ( plus "D" state processes) are almost the same in both cases, it is only "run queue" or "average load" showing a big difference.

 

Changes

You may notice this difference when you are patching system or upgrading to newer UEK4 kernel.

Cause

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


In this Document
Symptoms
Changes
Cause
Solution

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.