This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
>From: Unidata Support <address@hidden> >Organization: UCAR/Unidata >Keywords: 200310251920.h9PJKRgf026027 McIDAS install Karthik, I setup the LDM on s400 to ingest data from the University of Illinois IDD relay node flood.atmos.uiuc.edu. The following feeds are now being ingested: WMO - global observations and NCEP NOAAPORT model output FNEXRAD - NEXRAD Level III composite imagery created by Unidata NNEXRAD - NEXRAD Level III products FSL2 - FSL wind profiler data UNIWISC - Unidata-Wisconsin GOES satellite image sectors and products Yesterday evening I sent a note to the University of Michigan asking them to allow s400 to feed from their IDD node, yin.sprl.umich.edu. Today, I sent a note to SUNY Albany asking them to allow you to feed the NLDN lightning data from their IDD node, gusher.atmos.albany.edu. As soon as we hear back from U Michigan and SUNY Albany, I will uncomment appropriate request lines to the LDM configuration file /cutrim1/ldm/etc/ldmd.conf and then stop and restart the LDM. You can monitor how well the data is being ingested through the Unidata real time statistics pages: Unidata HomePage http://my.unidata.ucar.edu Software http://my.unidata.ucar.edu/content/software IDD http://my.unidata.ucar.edu/content/software/idd Current Operational Status http://my.unidata.ucar.edu/content/software/idd/iddgeneral.html Real Time Statistics http://my.unidata.ucar.edu/content/software/idd/rtstats/index.html From the last link, click on the 'Statistics by Host' link and then scroll down until you find a link for the Western Michigan LDM, s400.wood.wmich.edu. Clicking on that link will take you to a page that shows each feed that s400 is ingesting. There are a variety of links that will take you to plots of things like latency, log(latency), histograms of latency, data volume, number of products, and topology (who the data is coming from). I recommend visiting the pages to get an idea of how much data is being ingested by s400 and how well it is keeping up (through the latency plots). I setup McIDAS-XCD and ldm-mcidas decoders to decode the data coming in on the datastreams above. All decoded data is currently being written to subdirectories of /cutrim1/ldm/data. If Elen wants more data ingested (e.g., the NEXRAD Level II data or other model data, or the NOAAPORT GINI imagery), we may eventually have to use some of the space on /cutrim2. We can worry about that if/when the time comes, however. Along with the ingest and decode of data into McIDAS-compatible formats, I setup scouring of the data files and access to them through the McIDAS ADDE remote server (running from the 'mcadde' account): - the scouring was done by running various scour scripts from 'ldm's crontab - McIDAS ADDE service was setup by configuring the file .mcenv in the ~mcadde directory. Since the ADDE setup script always defaults to use of the HOME directory of the user 'mcidas', and since McIDAS-X v2003 is _not_ installed in ~mcidas, I did the following: - remove the McIDAS installation in ~mcidas. This was old and unsupported (v7.6), so the removal should have no impact - made links for a set of directories that point to the actual McIDAS installation point, /cutrim1/mcidas: cd ~mcidas ls -alt ... lrwxrwxrwx 1 mcidas ldm 20 Dec 24 16:38 help -> /cutrim1/mcidas/help lrwxrwxrwx 1 mcidas ldm 24 Dec 24 15:35 savedata -> /cutrim1/mcidas/savedata lrwxrwxrwx 1 mcidas ldm 19 Dec 24 15:34 inc -> /cutrim1/mcidas/inc lrwxrwxrwx 1 mcidas ldm 19 Dec 24 15:34 lib -> /cutrim1/mcidas/lib lrwxrwxrwx 1 mcidas ldm 19 Dec 24 15:34 man -> /cutrim1/mcidas/man lrwxrwxrwx 1 mcidas ldm 19 Dec 24 15:34 tcl -> /cutrim1/mcidas/tcl lrwxrwxrwx 1 mcidas ldm 21 Dec 24 15:34 admin -> /cutrim1/mcidas/admin lrwxrwxrwx 1 mcidas ldm 20 Dec 24 15:34 data -> /cutrim1/mcidas/data lrwxrwxrwx 1 mcidas ldm 24 Dec 24 15:34 workdata -> /cutrim1/mcidas/workdata lrwxrwxrwx 1 mcidas ldm 19 Dec 24 15:33 bin -> /cutrim1/mcidas/bin ... We should consider reinstalling (_not_ copying) McIDAS under ~mcidas. It would make the McIDAS installation a lot cleaner. If we do this, however, it will require some modifications in the McIDAS-related scripts in /cutrim1/ldm/util (mcscour.sh, batch.k, xcd_run). As far as I can tell, everything is working OK except LDM logging. I will have to touch base with Leonard about this when he gets back from his vacation. Until we can sort through why syslog logging is not working correctly, I have done the following: - modified /cutrim1/ldm/bin/ldmadmin and set the '-l logs/ldmd.log' flag on rpc.ldmd - since the normal log rotation mechanism (ldmadmin newlog) will not work now, I setup rotation of the LDM log files using newlog: # # rotate LDM logs # 0 18 * * * /cutrim1/ldm/bin/newlog logs/ldmd.log 14 I have tested the McIDAS ADDE access to the data being decoded on s400 from the 'mcidas' account on s400 and it works correctly. I don't know if other machines at Western Michigan will be able to do the same, however. If they are not, it means that there is some sort of a firewall setup that is blocking them from accessing the ADDE remote server on s400. A problem like this will take a system/firewall administrator to correct. That's all for today. Cheers, Tom