Skip to content
This repository has been archived by the owner on Jun 29, 2022. It is now read-only.

Accounting errors in jboss.eap.processes logic #435

Open
kdelee opened this issue Nov 1, 2017 · 1 comment
Open

Accounting errors in jboss.eap.processes logic #435

kdelee opened this issue Nov 1, 2017 · 1 comment
Assignees
Labels

Comments

@kdelee
Copy link

kdelee commented Nov 1, 2017

Specify type:

  • Bug

Bug severity (if applicable):

  • High

Description:

Error output in scan found jboss.eap.processes indicates that the logic that ps -A -f e | grep eap will spawn two processes that have eap in them is not reliable.


Bug Report

Version of rho:

[ 53570d8 ]

Expected behavior:

The processes that are searching for eap processes must be correctly accounted for.

Actual behavior:

An error is being produced that indicates a problem with the accounting.
Also, once I started looking into this more, it appears that more processes than expected are being spawned on Fedora, because on the Fedora 26 test machine without any jboss runninng, we get a report of 1 process.

etc_release.release jboss.eap.processes
Red Hat Enterprise Linux Server release 5.9 (Tikanga) Bad result (1 processes) from 'ps -A -f e | grep eap'
Fedora release 26 (Twenty Six) 1 EAP processes found

Steps to reproduce:

I was able to produce the "error" form of this bug by scanning a RHEL 5 machine without JBoss running. (there is a dormant wildfly zip install). The over counting is exhibited by scanning Fedora machine.

Idea: 💡
Maybe instead of a priori assuming X number of processes will be spawned by your search, you could collect the PIDs of processes spawned?

Environment information:

Provide detailed information on the scanning and target systems (if applicable); see example below.

Type Operating System OS Version Python Version Virtualization
Scanning System Fedora 25 3.6 Bare Metal
Target System RHEL 5.9 2.X ESXi
Target System Fedora 26 2.X ESXi
@kdelee kdelee added the bug label Nov 1, 2017
@kdelee
Copy link
Author

kdelee commented Nov 1, 2017

@noahl It is also important to note all this information is from scans using the scanbasic user we have on the test machines. (a non-sudo user)

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

2 participants