https://bugzilla.redhat.com/show_bug.cgi?id=998735
Bug ID: 998735
Summary: Several pmatop reporting/output oddities
Product: Fedora
Version: rawhide
Component: pcp
Assignee: nathans@xxxxxxxxxx
Reporter: nathans@xxxxxxxxxx
QA Contact: extras-qa@xxxxxxxxxxxxxxxxx
CC: fche@xxxxxxxxxx, mgoodwin@xxxxxxxxxx,
nathans@xxxxxxxxxx, pcp@xxxxxxxxxxx
Created attachment 788212
--> https://bugzilla.redhat.com/attachment.cgi?id=788212&action=edit
Screenshot showing negative per-process cpu and invalid cpu% values
Description of problem:
I've noticed pmatop in current PCP is reporting strangely in a couple of ways:
- incorrectly sorting top cpu burners. this might be a 64bit metric issue,
where the metric is being displayed using 32bits worth? (attached pcp archive
shows this)
- some columns are displayed with negative values (attached screenshot shows
this)
- some columns are displayed with both scientific notation and with K suffix
(see screenshot attached)
- column widths for some columns overflows the allotted screen real-estate
(attached archive shows the problem)
Version-Release number of selected component (if applicable):
pcp dev branch exhibits the problem.
How reproducible:
Every time for me, not so much for scox & fche after chatting. May be related
to long running cpu-burning processes I have locally. An archive is attached
showing the incorrect sorting.
Steps to Reproduce:
1. Run pmatop -h local:
2. Run pmatop -r atop 1
3. Both the above show the issues for me
Actual results:
pmatop displays either the wrong top-cpu-burners (case 2), or sometimes
non-existent processes (seen in case 1 occassionally).
Expected results:
pmatop displays top cpu burners that match up with what top(1) shows.
Additional info:
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug
https://bugzilla.redhat.com/token.cgi?t=m0EpmNcd2B&a=cc_unsubscribe
|