Bug 1913782

Summary: pmlogger: some metrics get logged twice
Product: Red Hat Enterprise Linux 8 Reporter: Andreas Gerstmayr <agerstmayr>
Component: pcpAssignee: Andreas Gerstmayr <agerstmayr>
Status: CLOSED ERRATA QA Contact: Jan Kurik <jkurik>
Severity: unspecified Docs Contact: Apurva Bhide <abhide>
Priority: unspecified    
Version: 8.3CC: agerstmayr, jkurik, mgoodwin, nathans, patrickm
Target Milestone: rcKeywords: Bugfix, Triaged, ZStream
Target Release: 8.0Flags: pm-rhel: mirror+
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: pcp-5.2.3-1.el8 Doc Type: No Doc Update
Doc Text:
Story Points: ---
Clone Of:
: 1917775 1917777 (view as bug list) Environment:
Last Closed: 2021-05-18 15:19:49 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1917775, 1917777    

Description Andreas Gerstmayr 2021-01-07 15:40:37 UTC
Description of problem:
Some metrics get logged twice every X seconds. This results in higher storage space requirements for pmlogger archives and for the redis database (if configured).

Version-Release number of selected component (if applicable):
pcp-5.1.1-3

How reproducible:
always

Steps to Reproduce:
1. dnf install -y pcp-zeroconf redis
2. systemctl start redis pmproxy
3. wait for some time to get logged metrics (few minutes)

Actual results:
Metrics are logged twice every 10 seconds:

[root@localhost ~]# pmseries "kernel.all.pswitch[count:10]"
300cd8fd0d9e258e68f309d9bbaad7e78392a2b5
    [Thu Jan  7 10:34:19.656194000 2021] 203594
    [Thu Jan  7 10:34:19.644994000 2021] 203579
    [Thu Jan  7 10:34:09.738189000 2021] 201359
    [Thu Jan  7 10:34:09.644312000 2021] 195104
    [Thu Jan  7 10:33:59.732399000 2021] 187307
    [Thu Jan  7 10:33:59.729557000 2021] 187295

[root@localhost ~]# pmval -U /var/log/pcp/pmlogger/host-10-0-136-73/20210107.10.33.0 kernel.all.pswitch
metric:    kernel.all.pswitch
archive:   /var/log/pcp/pmlogger/host-10-0-136-73/20210107.10.33.0
host:      host-10-0-136-73
start:     Thu Jan  7 10:33:49 2021
end:       Thu Jan  7 10:35:19 2021
semantics: cumulative counter (converting to rate)
units:     count (converting to count / sec)
samples:   all
10:33:49.844            5.683E+04
10:33:50.148  Archive logging suspended
10:33:59.732            4222.    
10:33:59.733  Archive logging suspended
10:34:09.738            6.663E+04
10:34:19.644             224.1   
10:34:19.656            1339.    
10:34:29.643             275.0   
10:34:29.652            1381.    
10:34:39.643             188.4   
10:34:39.651            1257.    
10:34:49.644             179.3   
10:34:49.652            1601.    
10:34:59.643             254.7   
10:34:59.652            1511.    
10:35:09.643             209.0   
10:35:09.650            1806.    
10:35:19.643             191.5   
10:35:19.651            1715.    

Expected results:
Metrics are logged once every X seconds.

Additional info:
This bug doesn't affect all metrics (for example mem.vmstat.pgfault is not affected by this bug in the test environment).

This issue is fixed in upstream since PCP 5.2.1. Requesting to backport this fix for RHEL 8.3.

Comment 8 errata-xmlrpc 2021-05-18 15:19:49 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (pcp bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2021:1754