I have run a command below at different time each at around 5 minute lapse at both node. The result is compiled in the attachment. Also, a cpu utilization graph both node from HPE IMC today for your reference.
1. show clock
2.
Original Message:
Sent: Jan 17, 2021 12:03 PM
From: Davide Poletto
Subject: ArubaCX: High CPU utilization
Hi! what's the output of show system resource-utilization command executed on both VSX nodes at various times?
I'm observing similar results (either on the VSX Secondary and VSX Primary, after about 60 days from upgrade to 10.05.0021 for our two Aruba 8320...thus on a scenario not exactly comparable with your one)...but I never noticed anything alarming from the ArubaOS-CX standpoint on both VSX nodes.
The CPU usage was 30/35% on average exactly when I checked with the top cpu command but then dropped to lower levels:
vsx-2# top
cpu memory
vsx-2# top cpu
top - 18:39:31 up 58 days, 7:43, 1 user, load average: 1.18, 1.53, 1.53
Tasks: 241 total, 2 running, 170 sleeping, 0 stopped, 0 zombie
%Cpu(s): 28.6 us, 25.4 sy, 0.0 ni, 44.4 id, 0.0 wa, 0.0 hi, 1.6 si, 0.0 st
KiB Mem : 15964776 total, 9668040 free, 1564156 used, 4732580 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 11832548 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
24836 root 20 0 82244 920 832 R 100.0 0.0 1:02.48 /lib/systemd/systemd-socket-proxyd 169.2+
3591 root 20 0 1087732 318416 94132 S 62.5 2.0 31596:48 /usr/sbin/ops-switchd --no-chdir --pidfi+
13833 admin 20 0 17736 2420 1976 R 37.5 0.0 0:00.14 /usr/bin/top -b -n 2 -c -o %CPU -w 110 -+
and quite the same on VSX Primary (same process same %CPU value).
The point is that...in both cases...the top cpu command, after just a few seconds from its first execution, once re-run changed completely (that's expected) and indeed the /lib/systemd/systemd-socket-proxyd process disappeared entirely...both VSX nodes' CPU usage concurrently dropped at 5-10%.
------------------------------
Davide Poletto
Original Message:
Sent: Jan 17, 2021 11:18 AM
From: Saiful Azlan Anuar
Subject: ArubaCX: High CPU utilization
Dear Sir,
Today we have upgraded 2 unit core switch firmware from FL.10.05.0030 to FL.10.05.0040. Right after the upgrade, i noticed the CPU of primary core spike frequently while the secondary cpu looks normal. You may find the cpu utilization graph by HPE IMC in the attachment.
I have run the top cpu on the primary switch as shown below.
**********************************************************************
ZA-CX6405-CS1# top cpu
top - 23:13:28 up 10:53, 2 users, load average: 5.43, 5.48, 5.46
Tasks: 279 total, 2 running, 223 sleeping, 0 stopped, 0 zombie
%Cpu(s): 18.8 us, 22.9 sy, 0.0 ni, 58.3 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 16030564 total, 11237628 free, 1446556 used, 3346380 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 14077612 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
10836 root 20 0 75700 484 412 R 108.3 0.0 639:51.31 /lib/systemd/systemd-socket-proxyd 169.2+
799 root 20 0 53484 29056 7036 S 25.0 0.2 71:05.95 /usr/sbin/ovsdb-server --remote=ptcp:664+
6788 azlan 20 0 3344 2140 1644 R 25.0 0.0 0:00.06 /usr/bin/top -b -n 2 -c -o %CPU -w 110 -+
1669 root 20 0 647188 28004 11624 S 8.3 0.2 17:45.16 /usr/bin/ndmd --pidfile -vSYSLOG:INFO
1673 root 20 0 276144 50496 7384 S 8.3 0.3 2:40.39 python /usr/bin/hpe_sshd
1 root 20 0 7480 6336 3044 S 0.0 0.0 0:04.46 /sbin/init
2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 [kthreadd]
4 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 [kworker/0:0H]
6 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 [mm_percpu_wq]
7 root 20 0 0 0 0 S 0.0 0.0 0:01.96 [ksoftirqd/0]
8 root 20 0 0 0 0 I 0.0 0.0 0:50.51 [rcu_sched]
9 root 20 0 0 0 0 I 0.0 0.0 0:00.00 [rcu_bh]
10 root rt 0 0 0 0 S 0.0 0.0 0:00.03 [migration/0]
11 root rt 0 0 0 0 S 0.0 0.0 0:00.08 [watchdog/0]
12 root 20 0 0 0 0 S 0.0 0.0 0:00.00 [cpuhp/0]
13 root 20 0 0 0 0 S 0.0 0.0 0:00.00 [cpuhp/1]
14 root rt 0 0 0 0 S 0.0 0.0 0:00.06 [watchdog/1]
15 root rt 0 0 0 0 S 0.0 0.0 0:00.02 [migration/1]
16 root 20 0 0 0 0 S 0.0 0.0 0:02.19 [ksoftirqd/1]
18 root 0 -20 0 0 0 I 0.0 0.0 0:00.00 [kworker/1:0H]
Primary core switch
ZA-CX6405-CS1# sh system
Hostname : ZA-CX6405-CS1
System Description : FL.10.05.0040
System Contact :
System Location :
Vendor : Aruba
Product Name : R0X24A 6405 Chassis
Chassis Serial Nbr : SG9AKM70BW
Base MAC Address : f860f0-612000
ArubaOS-CX Version : FL.10.05.0040
Time Zone : Asia/Kuala_Lumpur
Up Time : 11 hours, 51 minutes
CPU Util (%) : 35
Memory Usage (%) : 12
Secondary Core Switch
ZA-CX6405-CS2# sh system
Hostname : ZA-CX6405-CS2
System Description : FL.10.05.0040
System Contact :
System Location :
Vendor : Aruba
Product Name : R0X24A 6405 Chassis
Chassis Serial Nbr : SG9AKM70L8
Base MAC Address : f860f0-616400
ArubaOS-CX Version : FL.10.05.0040
Time Zone : Asia/Kuala_Lumpur
Up Time : 11 hours, 54 minutes
CPU Util (%) : 0
Memory Usage (%) : 13
*************************************************************************************
May i know the detail of the process that use 108% CPU as shown above. Or could it be some kind of a bug?
Thank you
------------------------------
Saiful Azlan Anuar
------------------------------