[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Question about memory usage



The requested memory is the only sure thing you would get. If it differ from your memory limit, you should know that this limit is optional and not guaranteed.

Try setting your requested memory to 1gb

If you want openshit scheduler and/or oom to priorize that container, consider setting your Pod limits equals to it's requests. Make sure to set both cpu+memory values.
Alternatievly, don't set requests, only limits.
This would give your pod the "guaranteed" QoSClass, less likely to get evicted (see oc describe pod xxx)

Regards.

On Fri, Jun 7, 2019, 2:40 PM Ricardo Mendes <maverickws outlook pt> wrote:

Ola Ferry!

 

Thank you for your quick reply! I actually forgot to mention I had been over that setting actually. I tried different values but currently its left to this:

 

cacti

Image: polinux/cacti 77b07bd 429.4 MiB

Ports: 80/TCP, 81/TCP, 443/TCP

Mount: cacti-volume  /data read-write

Memory: 250 MiB to 1500 MiB

 

As I’ve worked with cacti installs with hundreds of sources and still the memory footprint was relatively low (around 1GB) I would assume this settings would suffice.

So just to make sure, master’s memory DOES NOT influence this, right? (despite being 70% of 16GB)

 

Best regards,

Ricardo Mendes

 

 

From: Ferry Manders <ferry manders npo nl>
Date: Friday, 7 June 2019 at 13:33
To: Ricardo Mendes <maverickws outlook pt>
Cc: OpenShift Users List <users lists openshift redhat com>
Subject: Re: Question about memory usage

 

Ola,

You probably have a default limit set on your installation or project, or you have a limit set in your deployment that is 512MB memory (which is 536,870,912 bytes)
You could set the memory limit higher on your deployment, please take a look at the following doc for more information and examples : https://docs.okd.io/latest/dev_guide/compute_resources.html

With Regards,
--
Ferry Manders


Ricardo Mendes wrote on 07/06/2019 14:15:

Hi all,

 

I am fairly new to OpenShift and I’ve been fiddling around this past week.

Yesterday I came across the following issue:

While deploying an image of Cacti (Open-Source monitoring and RRD) I got an error of exhausted memory.

 

Fatal error: Allowed memory size of 536870912 bytes exhausted (tried to allocate 20480 bytes) in /data/www/cacti/lib/database.php on line 252

 

My servers memory usage is:

Master: 70% in use

Infra: 35% in use

Compute/node: 20%

 

I can also confirm the POD is running on the node, and Cacti has a low memory footprint actually, so, do I have to further increase the memory available on the masters??

 

Thanks!

Best regards,

Ricardo Mendes




_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users



_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]