Description of problem: Using the x86_64 DVD ISO(FC7T4), I attempted to install. The following errors are the last display on my console after I skip the media check in anaconda. After selecting skip, I see an error message about anaconda being terminated. FC6 X86_64 installs just fine on my machine. This seems to be a new failure in FC7. Version-Release number of selected component (if applicable): Unknown How reproducible: Every time Steps to Reproduce: 1. Attempt to install FC7T4 with a Sony DVD burner, model DRU-800A Actual results: Error reported after selecting 'skip test' of Media Check. Expected results: Install should proceed. Additional info: <6> device-mapper: multipath emc: version 0.0.3 loaded <3> SQUASHFS error: sb_bread failed reading block 0x15c5e <3> SQUASHFS error: Unable to read cache block [5716b1d:16f2] <3> SQUASHFS error: Unable to read inode [5716b1d:16f2] <6> sr 3:0:0:0: SCSI error: return code = 0x08000002 <6> sr0: Current: sense key: Hardware Error <6> Additional sense: Logical unit communication CRC error (Ultra-DMA/32) <4> end_request: I/O error, dev sr0, sector 6424656 <3> SQUASHFS error: sb_bread failed reading block 0x1bd4 <3> SQUASHFS error: Unable to read page, block 6f2cfb, size 2566 <6> sr: 3:0:0:0: SCSI error: return code = 0x08000002 <6> sr0: Current: sense key: Hardware Error <6> Additional sense: Logical unit communications CRC error (Ultra-DMA/32) <4> end_request: I/O error, dev sr0, sector 6424656 <6> sr 3:0:0:0: SCSI error: return code = 0x08000002 <6> sr0: Current: sense key: Hardware Error <6> Additional sense: Logical unit communication CRC error (Ultra-DMA/32) <4> end_request: I/O error, dev sr0, sector 6424656 <3> SQUASHFS error: sb_bread failed reading block 0x1bd4 <3> SQUASHFS error: Unable to read page, block 6f2cfb, size 2566 <3> SQUASHFS error: sb_bread failed reading block 0x1bd4 <3> SQUASHFS error: Unable to read page, block 6f2cfb, size 2566 <3> SQUASHFS error: sb_bread failed reading block 0x1bd4 <3> SQUASHFS error: Unable to read page, block 6f2cfb, size 2566
What controller is the drive attached to?
The Motherboard is an ASUS K8N-DL with the nVidia CK804: [sean@home-desk ~]$ sudo lspci -v -s 00:06.0 00:06.0 IDE interface: nVidia Corporation CK804 IDE (rev f2) (prog-if 8a [Master SecP PriP]) Subsystem: ASUSTeK Computer Inc. Unknown device 8162 Flags: bus master, 66MHz, fast devsel, latency 0 [virtual] Memory at 000001f0 (32-bit, non-prefetchable) [disabled] [size=8] [virtual] Memory at 000003f0 (type 3, non-prefetchable) [disabled] [size=1] [virtual] Memory at 00000170 (32-bit, non-prefetchable) [disabled] [size=8] [virtual] Memory at 00000370 (type 3, non-prefetchable) [disabled] [size=1] I/O ports at e800 [size=16] Capabilities: [44] Power Management version 2
Created attachment 153871 [details] ASUS K8N-DL lspci -v output
Created attachment 153872 [details] lsmod output
First thing I would try is NOT skipping the media check on install.
Well, the sha1sum matches: [sean@home-desk f7-test4-x86_64]$ sha1sum F-6.93-x86_64-DVD.iso 3cf40f1837565261be4690df58632742c9fe30a7 F-6.93-x86_64-DVD.iso [sean@home-desk f7-test4-x86_64]$ cat SHA1SUM 3cf40f1837565261be4690df58632742c9fe30a7 F-6.93-x86_64-DVD.iso 6ce7acb3ac9cb8a2630f6ccfb8e64081d0aa0128 F-6.93-x86_64-rescuecd.iso I'll reboot and run the media check now.
I redownloaded and reburned the x86_64 ISO on different media on a different machine with the same results. The media check does fail on both DVD's with similar errors to the original error condition.
I checked the Sony site for updates to the firmware of the DVD burner and it appears that there is a update for my burner(windows only). My burner has version 04 and the update is to version 06. I'll update my drive today on a windows machine and retest tonight.
I flashed the Sony drive on a windows machine and brought it back to test. It fails in the same way as before. I downloaded FC6 and burned it as well to make sure the burner hadn't just failed suddenly. But the FC6 DVD I just burned works fine. Any other suggestions?
Can we get the messages from the sata_nv driver initialization? It should include detection of the CD drive and setting transfer speed.
Well, I'm pretty sure that this is handled by the pata_amd not sata_nv, but they may be the same thing or I may have no clue what I am talking about. Anyway, I copied down the following after booting off the DVD and the 'test your media' curses screen was on the screen: <6> sda: sda1 <5> sd 0:0:0:0: Attached scsi disk sda <7> pata_amd 0000:00:06.0: version 0x28 <7> PCI: Setting latency timer of device 0000:00:06.0 to 64 <6> ata3: PATA max UDMA/133 cms 0x00000000000101f0 ctl 0x00000000000103f6 bmdma 0x000000000001e800 irq14 <6> ata4: PATA max UDMA/133 cms 0x0000000000010170 ctl 0x0000000000010376 bmdma 0x000000000001e808 irq15 <6> scsi2: pata_amd <7> ata3: port disabled: ignoring. <6> scsi3: pata_amd <6> ata4.00: ATAPI, max UDMA/66 <6> ata4.00: configured for UDMA/66 <5> scsi 3:0:0:0: CD-ROM SONY DVD RW DRU-800A KY06 PQ:0 ANSI:5 <4> sr0: scsi3_mmc drive: 62x/62x writer cd/rw xa/form2 cdda tray <6> Uniform CD-ROM driver Revision: 3.20 <7> sr 3:0:0:0: Attached scsi CD-ROM sr0 <7> ISO 9660 Extensions: Microsoft Joliet Level 3 <4> Unable to load NLS charset utf8 <4> Unable to load NLS charset utf8 <7> ISO 9660 Extensions: RRIP_1991A <4> Unable to identify CD-ROM format. <4> VFS: Can't find an ext2 filesystem on dev loop0
After selecting the 'test your media' button, the following was on the console before I rebooted: <7> ISO 9660 Extensions: Microsoft Joliet Level 3 <4> Unable to load NLS charset utf8 <4> Unable to load NLS charset utf8 <7> ISO 9660 Extensions: RRIP_1991A <6>sr 3:0:0:0: SCSI error: return code = 0x08000002 <6> sr0: Current: sense key: Hardware Error <6> Additional sense: Logical unit communication CRC error (Ultra-DMA/32) <4> end_request: I/O error, dev sr0, sector 72 <3> Buffer I/O error on device sr0, logical block 9 <3> Buffer I/O error on device sr0, logical block 10 <3> Buffer I/O error on device sr0, logical block 11 <3> Buffer I/O error on device sr0, logical block 12 <3> Buffer I/O error on device sr0, logical block 13 <3> Buffer I/O error on device sr0, logical block 14 <3> Buffer I/O error on device sr0, logical block 15 <3> Buffer I/O error on device sr0, logical block 16 <3> Buffer I/O error on device sr0, logical block 17 <3> Buffer I/O error on device sr0, logical block 18
I saw alan cox posting something to linux-scsi in the last couple of days that looks interesting and might be related. I'm not sure if it's relevant, but check out this: http://www.spinics.net/lists/linux-scsi/msg16225.html
What kind of IDE cable is attached to the CD drive (40-wire or 80-wire?)
I think we need this patchset, which was not taken upstream. A much more invasive fix was done for 2.6.22 instead.
http://www.mail-archive.com/linux-ide@vger.kernel.org/msg03501.html
Created attachment 154577 [details] 40-wire cable that was used when ticket was reported
Created attachment 154578 [details] 80-wire cable that fixed the issue reported in this ticket
Created attachment 154579 [details] Side by side comparison of both cables
Well, that was surreal. I honestly didn't know that there was such a thing as a 80-wire IDE cable. But sure enough, wikipedia saves the day. In addition, attaching the 80-wire IDE cable appears to allow me to get anaconda started and all appears to be well. I aborted the install after anaconda successfully spun up and fired up X to begin the installation. I'll continue with my testing, but sheesh! Thanks folks. It seems that either the new SCSI->ATA code in the kernel or the DVD-ROM drive is not respecting the mode of the cable when attached in FC7. FC6 was installed with the 40-wire cable installed and didn't have any issues like this.
Thanks for the report. It should be retrying a few times on the cable error and then changing down speed. I am a bit baffled why this didn't occur.
I noted that the BIOS reports that the speed is ATA66 even with the 40 conductor cable attached. Perhaps the BIOS is reporting a bad value to the driver?
The only vaguely reliable way to check the cable type on Nvidia hardware appears to be via ACPI. A pure ACPI driver is under development that may help here - so yes we know about that.
ACPI cable detect is now upstream and will be in 2.6.23. How much this helps for a given board will be dependant upon the BIOS alas.
Making this a bit more of a catch all for all the variants of the same problem now being sorted in -mm tree ready.
Changing version to '9' as part of upcoming Fedora 9 GA. More information and reason for this action is here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping