After booting the with the latest 2.6.26.6-79 kernel on Fedora 9, my system became extremely slow. To the point that Gnome menus would take 20-30 seconds just to drop down. Loading applications as simple as gedit would take a minute or two. In looking into this problem, I opened a terminal and started top. It showed my system at nearly 100% wait time on the CPU. Then I started poking around the logs (/var/log/messages), and I kept seeing the following errors, repeat over and over again:
Nov 4 03:08:54 localhost kernel: ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0 Nov 4 03:08:54 localhost kernel: ata1.00: BMDMA stat 0x24 Nov 4 03:08:54 localhost kernel: ata1.00: cmd c8/00:a0:09:9a:3d/00:00:00:00:00/e0 tag 0 dma 81920 in Nov 4 03:08:54 localhost kernel: res 51/40:00:18:9a:3d/00:00:00:00:00/e0 Emask 0x9 (media error) Nov 4 03:08:54 localhost kernel: ata1.00: status: { DRDY ERR } Nov 4 03:08:54 localhost kernel: ata1.00: error: { UNC } Nov 4 03:10:48 localhost kernel: ata1.00: configured for UDMA/100 Nov 4 03:10:48 localhost kernel: ata1: EH complete So, I thought I had my hard disk failing, and I was very concerned. I found some posts where others were having the same problem, but no one really thought it was an issue. All the other posts where for RAID setups though, on servers, so they may not see the interactivity issues that I was seeing. So, as an experiment, I booted the older 2.6.26.5-45 kernel, and guess what? No ata error messages in /var/log, and the responsiveness of my system has returned to normal. This clearly looks like a bug in the ata code somewhere to me. Any ideas?
-- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines