All Apps and Add-ons

Splunk for Unix and Linux TA cpu.sh output wrong for Micro-partitioned AIX systems

beeman07
Engager

This isn't so much a question as it is a bug report.

cpu.sh outputs correct information on non-Micro-partitioned AIX servers; but on Micro-partitioned systems with SMT, the sar command adds a new column that invalidates the cpu.sh output.

Column headings from "sar -P ALL 1 1" output on a stand-alone server. This is what cpu.sh expects to see:

11:12:04 cpu %usr %sys %wio %idle physc

Column headings from "sar -P ALL 1 1" on a micro-partitioned system with SMT enabled. Notice the new column:

11:12:02 cpu %usr %sys %wio %idle physc %entc

This difference is addressed in the sar manpage:

Beginning with AIX 5.3, the sar command reports utilization metrics
physc and %entc which are related to Micro-Partitioning and
simultaneous multi-threading environments. These metrics will only be
displayed on Micro-Partitioning and simultaneous multi-threading
environments. physc indicates the number of physical processors
consumed by the partition (in case of system wide utilization) or
logical CPU (if the -P flag is specified) and %entc indicates the
percentage of the allocated entitled capacity (in case of system wide
utilization) or granted entitled capacity (if the -P flag is
specified). When the partition runs in capped mode, the partition
cannot get more capacity than it is allocated. In uncapped mode, the
partition can get more capacity than it is actually allocated. This is
called granted entitled capacity. If the -P flag is specified and there
is unused capacity, sar prints the unused capacity as separate CPU with
cpu id U. Note: The sar command only reports on local activities.

Can this issue get fixed in the official release of the Splunk for Unix and Linux TA app?

(Is there a different way in which I should report this issue?)

araitz
Splunk Employee
Splunk Employee

I created issue SPL-53293 for this. We might need your help in verifying it, as I don't think we have a micropartitioned AIX system to test with.

0 Karma

araitz
Splunk Employee
Splunk Employee

Thanks for the report, I will add this to this list of things that we will consider fixing in the next release of the app.

0 Karma

dwaddle
SplunkTrust
SplunkTrust

If you have enterprise support, you can open a support case on this. That is probably a good idea if for no other reason than to get the bug officially filed and triaged.

According to araitz, this app should eventually wind up on github - you could provide a patch then if you have one. See http://blogs.splunk.com/2011/11/07/splunk-for-unix-and-linux-an-update-and-an-introduction/ -- but I just checked and it's not there yet.

araitz
Splunk Employee
Splunk Employee

Yeah, someone was supposed to put the app on github but never followed through 😕 I'll keep pushing.

Get Updates on the Splunk Community!

Splunk Answers Content Calendar, June Edition

Get ready for this week’s post dedicated to Splunk Dashboards! We're celebrating the power of community by ...

What You Read The Most: Splunk Lantern’s Most Popular Articles!

Splunk Lantern is a Splunk customer success center that provides advice from Splunk experts on valuable data ...

See your relevant APM services, dashboards, and alerts in one place with the updated ...

As a Splunk Observability user, you have a lot of data you have to manage, prioritize, and troubleshoot on a ...