[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[McIDAS #XRR-599672]: FW: Please remove port 23 access from shu.cup.edu, and add ports 22 & 112.

Hi Samuel,

re: unable to access shu from the 128.117.156 subnet at Unidata
> I don't know any blocked subnets for shu.cup.edu.  Not for port 112, or
> 23, or any port for that matter.  You might want to check with your network
> guy.

I have been conferring with our system admin (also our network guru) right along
on this.  I know that there is no systematic problem since I routinely login
to user's machines from my workstation which is one of the machines on that
particular subnet.

> Do you want me to check with our guy?

Sure, but it is not critical.

> I'm not sure how much you'd want to use our ADDE server though.  I'm sure you
> can find better servers than ours, but I'll leave 112 open so you can track 
> down
> the problem(if you want)

Thanks.  You leaving port 112 open will not cause any problems for you.  It will
let me kibitz on your ingest/processing of data in a very easy way, so I would
appreciate you leaving it open permanently.

> Thank you for fixing the MySQL problems.  If you see this problem often, you 
> might
> want to make up a HOW-TO webpage with directions on how to install MySQL for 
> McIDAS.
> I did the install that way, because it was the only one I could find for our 
> version
> of RedHat.

Actually, your installation of MySQL was correct.  The problems that I saw 
arose because
you copied MySQL libraries to the ~mcidas/lib directory.

> I have modified the LOCDATA.BAT, did the mcidas -config.  I like how mcidas 
> is working,

Very good.

> it's a very nice product.

McIDAS is very useful for doing manipulations with satellite data.

> I thought we had more RAM, but I thought I committed about 2 Megs of shared 
> memory to IDV.

I believe that you mean that you committed at least 512 MB of RAM to the IDV.  
It really
won't run/run well with less.

> We run a batch job every 15 minutes for IDV to process a GOES-EAST image, to 
> put it on a
> website for a weather display.  That might be something I'd want to tweak.  I 
> want IDV to
> point to our local ADDE server, once we get GOES-East 1km images ingesting.

OK.  There is only one source of GOES-East 1km VIS data in the IDD, the NIMAGE 
datasetream
of images from NOAAPORT.  1 km VIS data is available via ADDE from a variety of
different locations/datasets:

GOESEAST   unidata2.ssec.wisc.edu
GINIEAST   variety of hosts serve this data
EAST       unidata2.ssec.wisc

Take a look at the webpage I maintain on publically accessible ADDE servers to 
see a
more complete list:

http://www.unidata.ucar.edu/software/mcidas/adde_servers.html

> Should I re-define the shared memory in our system by putting a line in our
> /etc/rc.d/rc.local file like the following:
> echo 1048576 > /proc/sys/kernel/shmmax

You would only need to do this if you want to run McIDAS sessions with LOTS of
frames or run LOTS of simultaneous McIDAS sessions.

> Our network guy informed us that our traffic should be routed over I2 now, so
> I'm going to try out our feed from ldm.meteo.psu.edu.

Very good.

> Do you think I should un-comment the CONDUIT entries, and add WMO like you 
> suggest?

I left WMO enabled via the request for the UNIDATA datastream.  'UNIDATA' is a
compound datastream mnemonic that is the union of IDS|DDPLUS, HRS, and UNIWISC.
I strongly suggest that you leave that feed request in place, but you should
switch it to Penn State if you are now able to get data from them over I2.

> I'm going to tell our network guy to close SSH access to shu thru port 22.
> If you still need access, just let me know.

How about you leave an entry for port 22 for machines from the 128.117.140 
subnet
(and your own, of course).  This way, if you have problems we could login in
quickly and not have to wait until you can get hold of your system/network 
admin.

> Otherwise, please log out of LDM when you're done.

I will do so when I get into work (I'm still at home).

> I believe we want CONDUIT, but I'll check with our Meterology people.

CONDUIT was _the_ reason your machine was so I/O bound.  Quite frankly you do 
not
have enough memory in your machine to ingest the CONDUIT data and do the other
things that you are currently doing.

> I'm not much of an expert on this weather data, I'm just a system 
> administrator

Gotcha.  Please talk to Chad about the various datastreams that he is interested
in getting.  If he has questions, please have him submit an inquiry to
address@hidden.

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: XRR-599672
Department: Support McIDAS
Priority: Normal
Status: Closed