[McIDAS #VSV-153140]: Mcidas xcd

Hi Heather,

re:
> Thank you for getting back to me.

No worries.

re:
> I made some modifications to my /etc/hosts file and
> disabled selinux on my ingestor, hoping that that would fix the problem.  No 
> luck,

Disabling SELINUS was to allow the LDM to use 'rsyslog' for its logging.  It
has nothing to do with you not being able to read from the Novra.  I see from a
comment further down in your email that disabling SELINUX worked for the
LDM logging problem you reported.  It is time to scratch this one from the list
and move on to the next problem...

re:
> my
> new ingestor still cannot see my Novra receiver.  I have attached files that 
> contain the
> output of ifconfig -a and tcpdump.  You will see output for both my old 
> receiver
> (noaapnew) which still works fine and my new receiver (npingest).  I am at a 
> loss.  I am
> going to email the information to the guy in charge of the network we are 
> working on.
> Hopefully he could be of some help too?

The output included in the file tcpdump.output.newingestor that you attached 
shows that
the output from the Novra _is_ making it to the Ethernet interface on your new 
machine.
This tells me that the most likely cause for not being able to read the UDP 
stream
coming from the Novra is a block in the firewall on the local machine (not on
the general network).  If the local firewall is, in fact, the problem, you need
to set it up to allow _all_ traffic on the interface (Ethernet port) that the
Novra is connected to.  The following is a snippit from /etc/sysconfig/iptables
on our Linux NOAAPort ingest machine (which is running Fedora 16); I include it
as a guideline for what needs to be done on your machine:

-A INPUT -i eth1 -j ACCEPT

This entry allows all traffic on eth1 which is where our Novra is connected.

NB: after making changes to /etc/sysconfig/iptables (as 'root' !!), one must
restart 'iptables' for the change to take effect:

<as 'root'>

service iptables restart

re:
> Here is the output when I try to use cmcs:
> [root@npingest ~]# ./cmcs -ip 192.168.0.138 -pw Novra-S2 -shsat
> Unable to communicate with receiver at IP address 192.168.0.138
> .Error code: -1
> .
> Login unsuccessful.

The output in the file ifconfig.output you attached shows that the
machine's eth1 interface is configured to be on the same subnet
as the Novra:

eth1 -> 192.168.0.10
S300 -> 192.168.0.138

Good!  The fact that this is setup OK, and you are not able to
read data from the interface indicates that the problem is one
of two things:

1) firewall block of eth1 traffic

2) routing not setup for the traffic from the Novra

We addressed 1) above.  The easiest way to insure that 2) is not
the problem is to:

- create (as 'root') the file /etc/sysconfig/static-routes with
  the following content:

any net 224.0.0.0 netmask 240.0.0.0 gw 192.168.1.6 dev eth1

  The read/write/execute settings for 'static-routes' on our
  Linux NOAAPort ingester are '-rw-r--r--'.

  After creating 'static-routes', the network needs to be restarted
  so that the setting will be in use:

<as 'root'>
service network restart

After making these changes, it is likely that the LDM on the machine
that is trying to ingest NOAAPort needs to be restarted:

<as 'ldm'>

ldmadmin restart

Check to see if you are receiving data after the restart:

ldmadmin watch

re:
> As for my decoding machine...
> I disabled selinux and am now getting log files that are being written to!  
> Victory!

Very good.  That was the entire point of disabling SELINUX.

re:
> Unfortunately I am still getting McIDAS errors.  Now in both my XCD_START.LOG 
> and in the
> log files.  I re-installed mcidas using make all and make install.all.  Here 
> are my
> errors:
> 
> From XCD_START.LOG:
> startxcd.k: Done
> xcd_run exiting from MONITOR action at 14258.115455
> Starting DDS at 14258.115456
> problem with file CIRCUIT.DAT
> may need to be initialized with circuit command
> ingetext.k: Done
> 
> From ldmd.log
> Sep 15 07:54:56 npxcd pqact[2842] ERROR: pbuf_flush(): fd=7, 
> cmd=(/usr/local/ldm/decoders/xcd_run HRS): Broken pipe
> Sep 15 07:54:56 npxcd pqact[2842] ERROR: pipe_put(): write error: pid=2909, 
> cmd=(/usr/local/ldm/decoders/xcd_run HRS)
> Sep 15 07:54:56 npxcd pqact[2842] ERROR: [filel.c:305] Deleting failed PIPE 
> entry: pid=2909, cmd="/usr/local/ldm/decoders/xcd_run HRS"
> Sep 15 07:54:56 npxcd pqact[2842] ERROR: pipe_prodput: trying again:    49520 
> 20140915112857.255     HDS 196521568  JUTX02 KNES 151127
> 
> Even after my new mcidas install I do not have the following files anywhere 
> on disk:
> 
> CIRCUIT.DAT
> ingebin.k
> ingetext.k
> 
> These files are all present on my old decoding machine, which still works 
> fine.
> I have attached these log files as well.

Instead of trying to copy files from your old machine to your new one (OK if you
really know how things work; a bad idea if you are new at setting up data 
ingestion
and decoding), you should do the following:

<login as 'mcidas'>

cd $MCDATA
batch.k XCD.BAT

This will create files needed for XCD decoding (e.g., CIRCUIT.DAT, etc.).

Better yet would be to run the script 'mcxconfig' and answer the questions
you will be asked.  'mcxconfig' will lead you through the process of setting
up XCD decoding on a machine, and, in particular, will ask where you want
XCD decoding to occur (i.e., which directory) and then try to write needed
files in that directory.  If you do not have write permission in the directory,
you will need to drop back and fix the read/write/execute permissions on
the directory and try again from the beginning (meaning rerun 'mcxconfig').
'mcxconfig' will also create LDM pattern-action files, a file that contains
the line(s) you need to include in your ~ldm/etc/ldmd.conf file, and a file
of crontab entries you will need to install for your 'ldm' user.

re:
> Thanks again!!

No worries.

Cheers,

Tom
--
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: VSV-153140
Department: Support McIDAS
Priority: Normal
Status: Closed