Hi Kostas,
Even better ! The script exits and then leaves behind ten or so
MySQL processes that prevent you from re-running without error.
If you kill these, the config ends with:
2117] Listening on port 9002
done
Starting glite-lb-interlogd ... done
gLite LB Local Logger has been started [ OK ]
gLite LB Interlogger has been started [ OK ]
Starting the gLite WMS Log Monitor...
Starting LogMonitor... [ OK ]
The gLite WMS Log Monitor has been started [ OK ]
Starting the gLite WMS Workload Manager...
starting workload manager... ok
The gLite WMS Workload Manager has been started [ OK ]
Starting the gLite WMS Network Server...
Starting NetworkServer... [ OK ]
[ERROR] Could not start the gLite WMS Network Server
[ERROR] Please verify and re-run the script
BUT:
[root@areyousure glite_wmslb_install]# tail -f
/var/log/glite/networkserver_events.log
25 May, 16:48:55 -F- " NS::main ": Starting Network Server...
25 May, 16:48:55 -F- " NSR::drop ": Already running in an unprivileged
account...
25 May, 16:48:55 -F- " NSR::daemon": Daemon Started...
25 May, 16:48:55 -F- " NSR::init ": Listening on port: 7772
25 May, 16:48:55 -M- " NSR::init ": Authentication Handshaking Enabling:
Ok.
25 May, 16:48:55 -F- "NS2WM::init ": Initializing NS2WM Proxy...
25 May, 16:48:55 -F- "NS2WM::init ": FileQueue is:
/var/glite/workload_manager/input.fl
25 May, 16:48:55 -F- "NS2WM::init ": NS2WMProxy Initialization Done.
25 May, 16:48:55 -F- " NS::main ": Initialization Done. Running
Server...
25 May, 16:48:55 -F- " NS::main ":
--------------------------------------
Well - is the NS running or isn't it ?!!! I'm having thoughts
about harming myself again...
Cheers,
Barry.
On Thu, 25 May 2006, Kostas Georgiou wrote:
> Date: Thu, 25 May 2006 16:49:22 +0100
> From: Kostas Georgiou <[log in to unmask]>
> Reply-To: LHC Computer Grid - Rollout <[log in to unmask]>
> To: [log in to unmask]
> Subject: Re: [LCG-ROLLOUT] Old bugs in the WMS still there...
>
> On Thu, May 25, 2006 at 04:29:08PM +0100, Barry MacEvoy wrote:
>
> > As always (experienced previously with RC0 and RC2), I get the
> > following problems:
> >
> > 1. MySQL PID warnings
> >
> > 2. Failure to configure the LB correctly with the yaim script.
> >
> > /opt/glite/etc/config/scripts/glite-lb-config.py
> >
> > is the problem.
> >
> > I comment out lines 128 & 129 as suggested by Di Qing
> > as a solution to bug no. 16421 and replace with:
> >
> > #if not os.path.exists('/tmp/mysql.sock'):
> > # os.symlink('/var/lib/mysql/mysql.sock', '/tmp/mysql.sock')
>
> >From the os.path docs
> exists(path)
> Return True if path refers to an existing path.
> Returns False for broken symbolic links.
>
>
> So os.path.exists('/tmp/mysql.sock') returns false if /tmp/mysql.sock
> exists but /var/lib/mysql/mysql.sock doesn't exist. The os.symlink
> code will fail to create the symlink then because /tmp/mysql.sock
> does exist. If you are using a recent version of python you can
> use os.path.lexists...
>
> lexists(path)
> Return True if path refers to an existing path.
> Returns True for broken symbolic links. Equivalent to exists()
> on platforms lacking os.lstat(). New in version 2.4.
>
> Of course after a reboot /tmp/mysql.sock will go away so I don't really
> see the point why would anyone want to create it at configuration time...
>
> Cheers,
> Kostas
>
--------------------------------------------------------------
Dr Barry MacEvoy
High Energy Physics Group
Imperial College London
Blackett Laboratory
Prince Consort Road
LONDON SW7 2BW
England
T: +44 20 7594 7802
F: +44 20 7823 8830
M: 07767 323871
http://www.hep.ph.ic.ac.uk/cms/people/based_at_imperial.html
http://www.hep.ph.ic.ac.uk/e-science/people/macevoy.html
--------------------------------------------------------------
|