cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Paging File monitor

ssamraj
Helper

Hi Team,

 

Is it possible to monitor Paging File in windows server using DT one agent? Interest is , when Page File Name is "_Total" and alert when it's usage breaches threshold of 90%.

 

Many thanks,

Srikanth Samraj

9 REPLIES 9

AntonioSousa
DynaMight Guru
DynaMight Guru

You should be able to get the information that you want at the system level. In the host section, you should select memory and then "Swap usage":

AntonioSousa_0-1648238211643.png

If you want to setup alerts, you should use the metric: builtin:host.mem.swap.used

Antonio Sousa

Rick_Lee
Frequent Guest

Please note, in system admin swap alerting is based on percentage, any chance Dynatrace can make this a calculated metric out of box with percentage value? thanks!

Hi,

Since you have these two metrics:

  • builtin:host.mem.swap.used
  • builtin:host.mem.swap.total

I think you can create your own metric selector event:

 

(builtin:host.mem.swap.used / builtin:host.mem.swap.total) * 100

 

Best regards

❤️ Emacs ❤️ Vim ❤️ Bash ❤️ Perl

Hi Anton. The reason this is not an acceptable solution is because when creating a metric event using this calculation, we get the error "The number of metric dimension values in the last 24 hours (15448) exceeds the limit of 1000. Please refine the filter criteria or use a metric key based query definition." In order to actually set up an alert for our entire environment based on SWAP %, we would need it to be a standalone metric. Please let us know if this is possible.

Hi,

I do not see another way to do it. You can split scope and creating more than one metric, for example, per host groups.

If not, I would suggest to raise a product idea about it.

Best regards

❤️ Emacs ❤️ Vim ❤️ Bash ❤️ Perl

That's great! I will vote for it. 

Rick_Lee
Frequent Guest

I dis-agree this is resolved, any Unix Admin started their training, the first command to learn is vmstat. everything shows from vmstats is in percentage format from the native Linux OS. even RedHat admin book has it on percentage. All other monitoring tools also monitoring via percentages, we are looking for a percentage presentation as base metric so we can alert on this metric. 

When this option is available in remote agent (remote unix monitor), I don't know why it is a challenge to be made available for hosts.

ext:tech.RemoteAgent.swap_used_percent

Featured Posts