How to interpret DMC iostats reports



We just upgraded to Splunk 6.4.1 this weekend (yay team!). I'm looking into the awesomeness of the DMC, and two reports have me perplexed - the "Maximum IO Bandwidth" and the "Median Disk Usage". I have a fair amount of "red" in those reports. I clicked on the "learn more" link, but all it does it take me to the main Splunk doc page. Can someone explain these reports? I've attached an example.

alt text

alt text

0 Karma

Ultra Champion

Quick lazy thought: did you post this question (or a link to this answers post) on the corresponding dashboard's doc page? At the bottom (footer) of the docs page should be a place to ask this to get them to clarify in the docs the answer to your question.

0 Karma

Ultra Champion

I'm looking at our own 6.4.1 DMC.

I see the following when we look at Resource Usage: Instance

alt text

We see two mount points while in your case we don't see them and simply the disk usage against these file systems.

Looking under Resource Usage: Deployment at the Median Disk Usage and we see the graph you posted. One way is to look at the search it generates -

`dmc_set_index_introspection` search_group=* search_group=* sourcetype=splunk_disk_objects component=Partitions             
| eval mount_point = 'data.mount_point'             
| eval free = if(isnotnull('data.available'), 'data.available', '')             
| eval pct_disk_usage = 1 - free / 'data.capacity'             
| `dmc_set_bin_for_timechart_for_disk_usage`             
| eval server_mount_point = host.":".mount_point             
| stats Median(pct_disk_usage) as pct_disk_usage by server_mount_point _time             
| `dmc_disk_usage_rangemap_and_timechart`
0 Karma
.conf21 CFS Extended through 5/20!

Don't miss your chance
to share your Splunk
wisdom in-person or
virtually at .conf21!

Call for Speakers has
been extended through
Thursday, 5/20!