Red Hat Bugzilla – Bug 53492
ext3 errors on boot
Last modified: 2007-04-18 12:37:00 EDT
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.2) Gecko/20010809 Description of problem: I'm receiving errors on boot relating to the hard disk. I had no problems with ext2 (RH 6.2, 2.2.x) or NTFS (w2k) on the same drive/same box. I also do not receive these errors on another system (Laptop, IBM Travelmate drive) with same kernel. Drive: Maxtor DiamondMax 5400 RPM 13GB EIDE CPU: PIII 800 (overclocked to 920) FSB: 100 running at 115 MHz Note that, though the system is over-clocked, this is reproducable at the standard settings as well. Errors: hda: timeout waiting for DMA ide_dmaproc: chipset supported ide_dma_timeout func only: 14 hda: status error: status=0x58 { DriveReady SeekComplete DataRequest } This happens 4 times during bootup, both with the stock kernel and a recompiled one. This error does not appear to happen at any time after a successful boot. Prior to seeing the message, the system is frozen for a few (10?) seconds, then the HDD makes a clunk sound (like when initializing) and the messages are printed. I don't think this is a hardware problem; I have 6 gigs for ext3 and 6 for NTFS, and no problems arise in Win2k. Also, prior to installing the RH beta, ext2 did not display this behavior. In any case, everything works just fine beyond this; it only happens during the initial bootup, and doesn't seem to cause any other problems. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: I'm not sure how to reproduce this. On this one particular system, it happens at every boot. On another system (partitioned 13 gigs Fat32 and 6 gigs ext3), I see no such errors... Actual Results: See description... Additional info:
Could you try booting with "ide=nodma" on the kernel commandline ? (either edit /boot/grub/grub.conf or type on the lilo commandline)
I have tried your suggestion (ide=nodma), and this machine booted faster than I've ever seen -- though I've never seen the error message before, everything does work much faster with regard to disk access. I suspect the issue might be (I probably should have mentioned this before) that I'm using an ATA/66 drive, on an EIDE controller (Asus P3B-F motherboard, 440BX chipset). Both the drive and the CD-ROM are on the same (primary) controller. I was sure ATA/66 was backward compatible, but perhaps it still mucks with the DMA part (the board does support UDMA...) I don't know if disabling DMA is the solution to my problem, or just trouble-shooting advice, but in either case, unless there is a fix, I'll keep DMA off until I hear otherwise. Thanks for the advice.