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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Hi Bei, Sorry for taking so long to get back to you... re: > Hi, I think my previous installation is wrong, so I strictly follow > the instruction of LDM homepage. And this time the permission and > ownership of the files which you mentioned in the last mail are > correct now. OK. re: > This time I add the request line in ldmd.conf and change the path in > /etc/rsyslog.conf. Very good. I see from the next comment that logging is still not working: re: > After ldmadmin start, and ldmadmin watch, I successfully have seen > those dataflow again. But still no data is saved and logfile is empty. This typically means that either the entry in the syslogd configuration file (e.g., /etc/rsyslog.conf on your system) is not correctly configured, or the system logging daemon, syslogd, is not running. Questions: - does the file /home/ldm/var/logs/ldmd.log exist on your system? - if yes, what are its read/write permissions? re: > Further, once I ldmadmin stop, and resume ldm by ldmadmin start & > watch, I can't see those dataflow. This typically means that there is something wrong with the system clock on the upstream or receiving machines. Since we know that the clock is correct on the upstream machine for your feed request, idd.unidata.ucar.edu, I have to ask: - is your system clock correct? I.e., are you maintaining the time on your machine by running the 'ntpd' server? re: > I didn't add any task in pqact, and the contents in pqact.conf are all > comments. What I have done is just install, and activate like the > instruction of LDM homepage. OK, this is why no products are appearing in your file system. The LDM provides two capabilities that are semi-independently configured: - moving products from point to point - doing something with products that are received on the local machine The second item is configured by: - running 'pqact' from the LDM configuration file ~ldm/etc/ldmd.conf - configuring one or more entries in a pattern-action file The default invocation for 'pqact' is what is seen in the ldmd.conf file you sent: EXEC "pqact" This line says to run 'pqact' upon LDM startup. The default for this action is to send all products received to the implied/default pattern-action file ~ldm/etc/pqact.conf. So, in order to process the data you are receiving, you have to add one or more properly formatted actions in ~ldm/etc/pqact.conf. Try adding the following action to ~ldm/etc/pqact.conf: EXP ([12][0-9][0-9][0-9])\.([0-3][0-9][0-9])(bfr|ion|wet)Prf(.*\.nc) FILE -close data/Cosmic/\1\2/\3Prf/\3Prf\4 Notes: - some (not all) whitespace in pattern-action files are tabs, not spaces The example action above annotated to show where tabs are required is: EXP<tab>([12][0-9][0-9][0-9])\.([0-3][0-9][0-9])(bfr|ion|wet)Prf(.*\.nc) <tab>FILE<tab>-close <tab>data/Cosmic/\1\2/\3Prf/\3Prf\4 Given a Cosmic product with product ID: pub_out/cosmicrt/level2/ionPrf/2011.056/ionPrf_C001.2011.056.17.02.G12_0001.0001_nc The output file created would be: ~ldm/data/Cosmic/2011056/ionPrf/ionPrf_C001.2011.056.17.02.G12_0001.0001_nc NOTE: - after you make the change to ~ldm/etc/pqact.conf, you need to verify that your modifications/additions are OK: <as 'ldm'> ldmadmin pqactcheck If there are no errors, you can activate the changes in two different ways: <as 'ldm'> ldmadmin restart - OR - ldmadmin pqactHUP If 'ldmadmin pqactcheck' indicated that there was a problem in pqact.conf, you need to re-edit the file and fix whatever problems exist and then rerun 'ldmadmin pqactcheck'. Do this until there are no errors, and then run 'ldmadmin pqactHUP'. > I attached the three conf files to you. OK, please check to make sure that the system logging daemon is running: ps -eaf | grep rsyslogd If is not, you can start it as 'root': chkconfig --levels 345 rsyslogd on /etc/init.d/rsyslog start re: > Thanks! > I think things can be figured out soon! Me too. re: > Sorry for any stupid questions because actually I am not very familiar > with linux. No worries. We'll get this going soon! 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: PGJ-802823 Department: Support IDD Priority: Urgent Status: Closed