I ran it again this morning with top and iotop running. At first I need to erase the 386 version of qt-x11 and two dependencies to allow the update to occur. This went quickly. Then I ran the real update. yum started out at around 500MB and shortly after printing "Running transaction test" it went up to around 600MB. It stayed there for a while. I wasn't doing accurate timing, but probably a couple of minutes. Then it went up to 3GB in what appeared to be a coupe of jumps. There were some short term lockups of the display (possibly due to a flush to disk?) and I really couldn't tell if this made the climn in usage appear to occur in jumps or if there really were a couple of steps involved. After several minutes it got to "Running transaction". Before printing out information about the first update, very roughly 20 minutes went by. CPU usuage was low during this entire time. I/O usage was up and down, with yum doing reads. Once the actual installs started happening there was a mix of significant read and write activity. There were 49 steps to the update including a kernel update (3 install, 21 update, 2 remove). Once it got to doing the actual updates, things went relatively fast (excepting the kernel install, but that has a lot of files). -- fedora-test-list mailing list fedora-test-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
