Hi all,

just wondering why my zlm servers are so slow to provide updates to themselves. Running "rug up -y -N" on a zlm secondary server that is registered to itself and has a few dozend patches to download and apply brings cpu usage up to 100% for about half an hour until I get something like:

Download failed: (https://duesseldorfzlm01.duesseldorf...zenworks-pkgmg t/repo/getBundleFile/063f70c5cb480ea76cdc8c5f0530cdf8/1/sles-9-i586/libexif-0.5. 12-118.10.i586.patch.rpm) The request timed out

Running the same command again will pick up where it left off and after about another half an hour I'll get the same error again, just with a different rpm to fail on download. Running top thru all this looks like the following most of the time:

4815 root 18 0 433m 68m 7872 S 99.9 6.8 7:22.03 /usr/sbin/ndsd
5012 zenworks 16 0 847m 430m 21m S 82.0 42.5 5:32.53 jsvc.exec -jvm server -user zenworks -home /opt/novell/zenworks/lib/java
4431 root 34 19 217m 127m 9400 R 42.4 12.6 2:25.98 zmd /opt/novell/zenworks/lib/zmd/zmd.exe --sleep-resume
4990 zenworks 15 0 683m 21m 8560 S 8.4 2.2 1:30.62 /opt/novell/zenworks/lib/java/jre/bin/java -Djava.net.preferIPv4Stack=tru
5029 zenworks 15 0 549m 90m 18m S 5.9 8.9 1:09.18 /opt/novell/zenworks/lib/SunJava2-1.4/jre/bin/java -server -Xms256m -Xmx2
7797 root 15 0 22508 8976 4468 S 2.0 0.9 0:01.81 /opt/novell/zenworks/bin/mono /opt/novell/zenworks/lib/rug/rug.exe --debu

Now why is ndsd taking up so much cpu time here? Any ideas?

thanks, Lothar

btw, this on a single CPU sles 9 with 1GB RAM (vmware esx 2.5 running on hp proliant hardware)