Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.
Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.
Kind-of sounds like I/O contention... what's your disk configuration? Do you run RAID? If so, what kind? What do you observe if you monitor "iostat -x 5" while the problem is occuring?
Art, yes RAID-1
Do the relevant pqact(1) configuration-file entries use the "-flush" or "-close" option?
Steve, yes..except EXEC entries, are they needed there?
How are you dividing out the write tasks with LDM? Each pqact only has 32 pipes, so if you are attempting to fit too many products through those available pipes, there will be lags.Install system monitoring tools like 'dstat' and 'sysstat', monitor the disk IO rates before you do something drastic like reformatting or changing filesystems.
http://ldm01.michiganwxsystem.net/munin/localhost/localhost/index.html
When I have seen problems before, it is almost always a LDM configuration issue and not having enough pqacts running.daryl
hmm.. maybe I should look that over thanks to everyone .. gave me some food for thought.. -Jeff
ldm-users archives: