perfSONAR User Q&A and Other Discussion

[pete <>]

Hey Antoine, That worked!!  I had to set the default to openjdk-8, when 
I unstalled openjdk-9 it broke things.

However, I reinstalled openjdk-9, and did this:

# sudo update-alternatives --config java
There are 2 choices for the alternative java (providing /usr/bin/java).

  Selection    Path                                            Priority 
  Status
------------------------------------------------------------
* 0            /usr/lib/jvm/java-9-openjdk-amd64/bin/java       1091 
 auto mode
  1            /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java   1081 
  manual mode
  2            /usr/lib/jvm/java-9-openjdk-amd64/bin/java       1091 
  manual mode

Press <enter> to keep the current choice[*], or type selection number: 1


Then cassandra started up.


By the way, I was playing with the CentOS 7 version today, and I noticed 
that firewalld is running by default on the central management bundle 
for CentOS 7.  It is NOT on the debian/ubuntu central management bundle.

Would you be able to let them know that too please?

Thanks again!!!

Pete


On 10/20/2017 04:28 PM, Antoine Delvaux wrote:
> Hello Pete,
>
> I think I've found the reason for this.  I suspect the 
> perfsonar-centralmanagement bundle dependencies have installed 2 versions of 
> openjdk: 8 and 9 and cassandra is trying to run with 9 and failing.
>
> Can you try to either change the default java to be openjdk-8 or even 
> completely remove openjdk-9 from your server and then try starting cassandra 
> again?
>
> This should be possible with something like `apt-get purge 
> openjdk-9-jre-headless` or any other openjdk-9 package you would have 
> installed at the moment.
>
> If you confirm this is the problem, I'll make a new cassandra package 
> available in our repo that would conflict with openjdk-9.
>
> Thanks,
>
> Antoine.
>
>
> > Le 20 oct. 2017 à 14:43, pete 
> > <>
> >  a écrit :
> >
> >
> >
> > I've done some testing on a vanilla system.  I choose only defaults during the 
> > install.  It appears to be the startup script (for cassandra) conversion from 
> > update-rc -> systemd that might be the issue:
> >
> > cassandra doesn't start at least on the debian setup of the central 
> > management bundle.  Here is the error when cassandra is installed:
> > ---
> > update-rc.d: warning: start and stop actions are no longer supported; falling 
> > back to defaults
> > ---
> >
> > update-rc and systemd appear to be confused on whether cassandra is running.
> > Here is what is looks like according to systemd--it thinks it's running:
> >
> > ╚═» systemctl status cassandra
> > ● cassandra.service - LSB: distributed storage system for structured data
> >    Loaded: loaded (/etc/init.d/cassandra; bad; vendor preset: enabled)
> >    Active: active (exited) since Fri 2017-10-20 09:35:07 CDT; 3min 31s ago
> >      Docs: man:systemd-sysv-generator(8)
> >   Process: 1263 ExecStart=/etc/init.d/cassandra start (code=exited, 
> > status=0/SUC
> >     Tasks: 0
> >    Memory: 0B
> >       CPU: 0
> >
> > Oct 20 09:35:06 ubuntu16 systemd[1]: Starting LSB: distributed storage system 
> > fo
> > Oct 20 09:35:07 ubuntu16 systemd[1]: Started LSB: distributed storage system 
> > for
> > Oct 20 09:38:06 ubuntu16 systemd[1]: Started LSB: distributed storage system 
> > for
> >
> >
> > However, It's not running according to ps, and there is nothing running on 
> > it's port.  ie., lsof -i :9160 shows nothing.
> >
> > Has anyone every had a central management bundle on debian work with 
> > cassandra before?
> >
> > Any ideas?  Thanks!
> >
> >
> > Pete
> >
> >
> >
> > On 10/19/2017 09:23 PM, pete wrote:
> > > Hi, I'm running perfsonar 4.0.1.  I have 2 test-points on raspberrypis 
> > > running ubuntu-jessie, and I'm trying to get them to archive to a central 
> > > management host which is on ubuntu 16.04.
> > > I'm not sure where to point the archive to in my mesh config, becauset here 
> > > is no http://mycentralmanagementhost/esmond/perfsonar/archive on the central 
> > > host---as there is on a standard toolkit.
> > > esmond is configured, and I've added ip access as the docs suggest.
> > > eg.,/usr/share/esmond/util/esmond_manage add_user_ip_address example_user 
> > > 10.0.1.1 10.0.2.0/24
> > > I noticed that cassandra is not running, which I thought would be needed for 
> > > esmond to run, but there are no logs in /var/log/cassandra.
> > > It looks like cassandra is set to run via systemctl, but it doesn't start.
> > > I followed the docs, and I set this up in my meshconfig-agent.conf on my 
> > > test-points:
> > > configure_archives 1
> > > use_toolkit 0
> > > Here is the log when I restart meshconfig-agent:
> > > 2017/10/20 02:10:02 (30741) WARN> perfSONARRegularTesting.pm:656 
> > > perfSONAR_PS::MeshConfig::Generators::perfSONARRegularTesting::__build_tests - 
> > > Unable to find measurement archive for test 'Ping Test Between MOREnet Latency 
> > > Hosts'. Proceeding with test but results will not be stored.
> > > Any ideas on what's missing?  Please let me know if you my configs would 
> > > help, but I'm thinking something should be running on the central management 
> > > host by default at /esmond/perfsonar/archive.  And cassandra not running 
> > > seems odd, but I'm not too familiar with esmond/cassandra as they usually run 
> > > well from the initial setup.
> > > Thanks,
> >
> > --
> > Pete Greenwell
> > System Administrator
> > Missouri Research and Education Network [MOREnet]
> > http://www.more.net

--
Pete Greenwell
System Administrator
Missouri Research and Education Network [MOREnet]
http://www.more.net