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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: Gilbert Sebenste <address@hidden> >Organization: NIU >Keywords: 200406150354.i5F3sOtK011267 LDM Fedora Core Linux Hi Gilbert, re: When did you upgrade to the *.2190.nptlsmp kernel? >Last week, in utter deperation, when it was still doing it under 2188. OK, this is important information. >I am wondering if what Gerry said is correct. Take a look at my >"partitions". Those are the default, and I'm realizing that may be my >mistake. What Gerry was really talking about was keeping the LDM queue off of a RAID partition. I have been working with Gerry to get his machine running well under FC1 with a RAID. My experince with RAID under Linux has not been a good one. After my experience in trying to use a hardware RAID card (Promise FasTrak TX2000) I switched to use of softare RAID setup by the FC1 installation program. When Gerry experienced performance problems on his dual Xeon 2.4 Ghz box, I suggested that he try a software RAID AND move his LDM queue off of his RAID and onto his system disk (this is how I have setup two machines). After he got his /etc/raidtab entries tuned, his machine started running nicely. -- Aside: the FC1 Linux SMP kernels recognize and use the hyperthreading offered by the Xeon processors. Gerry's dual CPU machine looks and performs like a quad processor machine. His load averages are at or below 1 for the great majority of the time even though he is ingesting ALL data available in the IDD AND decoding all CONDUIT and other data into GEMPAK format AND filing all NEXRAD Level II data! Neither Gerry or the UPCs machines have crashed with buss errors or any other reason. >Take a look at my partitions on weather, which are localized. As far as I can tell, weather3 does not have a RAID: cat /etc/fstab LABEL=/ / ext3 defaults 1 1 LABEL=/boot /boot ext3 defaults 1 2 none /dev/pts devpts gid=5,mode=620 0 0 none /proc proc defaults 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hda3 swap swap defaults 0 0 /dev/fd0 /mnt/floppy auto noauto,owner,kudzu 0 0 /dev/cdrom /mnt/cdrom udf,iso9660 noauto,owner,kudzu,ro 0 0 df -k Filesystem 1K-blocks Used Available Use% Mounted on /dev/hda2 238172636 43623996 182450092 20% / /dev/hda1 202220 25097 166683 14% /boot none 774692 0 774692 0% /dev/shm Looks to me like this has 1 250 GB hard disk. Am I missing anything? >Yep, no problems there, AND I have a 400 MB queue. My experience says that the queue size should not be a problem. Again, I am running under FC1 with LDM queues ranging from 1 to 4 GB. If making the queue small helps, it may well be the case that you have some sort of a problem with memory (the LDM queue is a memory mapped file). What memory do you have installed in weatehr3? >I was going to install >FC2 after they get some crucial bugs out (like cups not working for my >printer, Nvidia driver mega-problems). We tried building the LDM with large queue support (> 2 GB) on a test FC2 machine here yesterday. The test failed miserably. I strongly recommend that you not upgrade to FC2 at the moment! >I suspect it isn't this new kernel. It's my partitions. I am not sure how this would be the case, since I do not see you using a RAID. Cheers, Tom -- NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publically available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.