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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.
Bruce, > I've been working with Justin Weaver on an LDM issue we've been having > for the last couple of weeks or so. Specifically... LDM continues > using more and more memory as data arrives until the server crashes... > almost like LDM never releases any memory. I'm certainly not a linux > guru, however. I've worked hard to eliminate memory-leaks, so this is a surprise. Our 6.9 LDM-s don't exhibit memory-leaks. Are you collecting metrics with a periodic, crontab(1)-based, "ldmadmin addmetrics" command? If so, can you plot them via the command "ldmadmin plotmetrics"? If so, what does the "Space" plot show? > We are attempting to run ldm 6.9.7 in a virtual server environment with > CentOS 5.2. I noticed a few errors at the end of the make.log file. > Not sure if that would help or not. We've tried one older version of > ldm (6.9.6) to see if it would run any better, but had the same results > as above. We felt like this was a configuration issue with either pqact > or ldmd.conf, but using bare-bones configurations didn't help our > cause. Also checked the support archives, but didn't find anything that > might help us. I haven't heard of a memory-leak in the LDM for several years. > Any suggestions on what we might try next? Try to get the "ldmadmin plotmetrics" command working and send me the results. > Thanks and best regards, > > Bruce > > -- > > Bruce Haynie > Electronic Systems Analyst > National Weather Service > Lubbock, TX > address@hidden > Amateur Radio Call Sign: NW5S Regards, Steve Emmerson Ticket Details =================== Ticket ID: TBZ-761980 Department: Support LDM Priority: Normal Status: Closed