cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2705
Views
5
Helpful
8
Replies

PnP slow response

jwiget
Level 1
Level 1

Hi Team

We use APIC-EM Version 1.5.1.1054 and PnP Version 1.5.1.35.

APIC-EM installation on a VM with 10 CPU's, 64G Memory.

In the last time we had installed a lot of new miniswitches.

Now, we have over 500 devices connected to APIC-EM. I found a limit that not more than 200 devices should be marked as unclaimed.

Because of that, we moved the devices to state ignored.

But we have still the issue, that it seems that PnP has slow response, if you choose any of the button.

Discovery, Host inventory and so on give normal response.

As soon we go to PnP and choose a button (like Devices, Bulk Import) the response get very slow. And the cpu state under system health change from normal 50% to between 80 - 90 %.

A reboot of APIC-EM helps not really. After reboot the same issue.

Do we have a possibility to improve the situation to get besser response for PNP?

Thanks for your help.

Kind Regards

Jochen

1 Accepted Solution

Accepted Solutions

Hi Adam

It seems that update to version 1.6 has improve the response of PnP.

Jochen

View solution in original post

8 Replies 8

aradford
Cisco Employee
Cisco Employee

Hi Jochen,

What is your disk performance like.

Ssh to the grapevine linux console and try the following command:

dd if=/dev/zero of=/tmp/foo bs=1M count=512 conv=fdatasync

It will give you a performance of the IO on the disk.

Adam

Hello Adam

Sorry for the delay.

This are the result of the command.

[Fri Oct 27 08:55:29 CEST] grapevine@131.152.2.181 (grapevine-root-1) ~

$ dd if=/dev/zero of=/tmp/foo bs=1M count=512 conv=fdatasync

512+0 records in

512+0 records out

536870912 bytes (537 MB) copied, 3.19221 s, 168 MB/s

(grapevine)

Jochen

Hi Adam

An Update to the issue.

The ESX team had moved the datastore of the VM to SD disk and the throughput of the I/O is now average 800 Mb.

But on APIC-EM, the PnP part has still slow performance if you click on any button there.

Regards

Jochen

Hi Adam

It seems that update to version 1.6 has improve the response of PnP.

Jochen

Thanks for update.  I will check with engineering to see if the reason for this.

Adam

Hi Adam

With PNP version 1.6.0.2000039 it worked fine.

After update to  PNP version 1.6.1.105 we have again the slow performance.

A switch  to check an send the configuration take between 40 to 90 min.

Disk performance should be good:

$ dd if=/dev/zero of=/tmp/foo bs=1M count=512 conv=fdatasync

512+0 records in

512+0 records out

536870912 bytes (537 MB) copied, 0.990282 s, 542 MB/s

And the other issue that we have, in the configuration listed ip name-server "ip address".

At PNP in teh configuration file ist this line incuded , but after deploying configuration to the switch this line is missing.

Thanks a lot

Regards

Jochen

That is really strange.

What platform is the configuration being deployed on? What version of IOS?

Adam

We are testing with ws-c3560cg-8pc version 15.2.E7 and 15.2.E8.

And in generell PnP has a slow response if you click on any button there.


Jochen

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:


This community is intended for developer topics around Data Center technology and products. If you are looking for a non-developer topic about Data Center, you might find additional information in the Data Center and Cloud community