[Beowulf] ganglia on CentOS 8?

[David Mathog]

On Wed, 1 Apr 2020, Jonathan Engwall wrote:

> About getex:GNU nano 4.6               /usr/local/etc/gmond.conf            

Values are the same other then gexec.  Swapping that doesn't resolve the 
issue that no hosts are reported.

> Though not exactly what you are doing, bringing up the services:
> #!/bin/bash
> 
> service gmond stop
> service gmetad stop
> nginx
> nginx -s reopen
> nginx -s reload
> service gmetad start
> service gmond start
> service ipfw3 stop
> service ipfw3 start

It is all started as far as systemctl is concerned:

systemctl status gmond.service
● gmond.service - Ganglia Meta Daemon
    Loaded: loaded (/usr/lib/systemd/system/gmond.service; enabled; vendor 
preset: disabled)
    Active: active (running) since Thu 2020-04-02 11:16:14 PDT; 2min 18s 
ago
   Process: 3770 ExecStart=/usr/sbin/gmond (code=exited, status=0/SUCCESS)
  Main PID: 3771 (gmond)
     Tasks: 2 (limit: 26213)
    Memory: 2.0M
    CGroup: /system.slice/gmond.service
            └─3771 /usr/sbin/gmond

Apr 02 11:16:14 poweredge.cluster systemd[1]: Starting Ganglia Meta 
Daemon...
Apr 02 11:16:14 poweredge.cluster systemd[1]: Started Ganglia Meta Daemon.
[root at poweredge ganglia]# systemctl status gmetad.service
● gmetad.service - Ganglia Meta Daemon
    Loaded: loaded (/usr/lib/systemd/system/gmetad.service; enabled; vendor 
preset: disabled)
    Active: active (running) since Thu 2020-04-02 11:12:33 PDT; 6min ago
  Main PID: 3659 (gmetad)
     Tasks: 9 (limit: 26213)
    Memory: 3.0M
    CGroup: /system.slice/gmetad.service
            └─3659 /usr/sbin/gmetad -d 1

Apr 02 11:12:33 poweredge.cluster systemd[1]: Started Ganglia Meta Daemon.
Apr 02 11:12:33 poweredge.cluster gmetad[3659]: Sources are ...
Apr 02 11:12:33 poweredge.cluster gmetad[3659]: Source: [my cluster, step 
15] has 1 sources
Apr 02 11:12:33 poweredge.cluster gmetad[3659]:         127.0.0.1
Apr 02 11:12:33 poweredge.cluster gmetad[3659]: Data thread 
139734408959744 is monitoring [my cluster] data sou>
Apr 02 11:12:33 poweredge.cluster gmetad[3659]:         127.0.0.1

I did just notice though that /var/lib/ganglia/rrds is NOT populated on 
the CentOS 8 system, whereas it is on my Ubuntu laptop.  Also there is 
something peculiar going on with the network, because even though iptables 
is not running anywhere the gmond/gmetad ports on the CO8 system are not 
visible from the Ubuntu system, but in the reverse direction they are.

A thread on this problem has been started here:

https://forums.centos.org/viewtopic.php?f=54&t=73893

Regards,

David Mathog
mathog at caltech.edu