PDA

View Full Version : Kernel error spammed in /var/log/messages



gertdesmet
6th February 2007, 12:12 AM
Hello,

I have this error spamming since last night on FC6. It starts about 5 minutes after I log into a session. At the same moment two CD-Rom icons appear on the desktop (Audio and Audio 2) and I can't use my cd-rom drives anymore untill i reboot my computer. What could be the cause of this ? I have an Asus P5B-E motherboard with latest BIOS.

Here is the content of /var/log/messages :



Feb 4 20:46:11 nuxbox kernel: hda: cdrom_pc_intr: The drive appears confused (ireason = 0x01). Trying to recover by ending request.
Feb 4 20:46:23 nuxbox kernel: hda: cdrom_pc_intr: The drive appears confused (ireason = 0x01). Trying to recover by ending request.
Feb 4 20:46:23 nuxbox kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 4 20:46:23 nuxbox kernel: ide: failed opcode was: unknown
Feb 4 20:46:23 nuxbox kernel: hda: drive not ready for command
Feb 4 20:46:25 nuxbox kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 4 20:46:25 nuxbox kernel: ide: failed opcode was: unknown
Feb 4 20:46:25 nuxbox kernel: hda: drive not ready for command
Feb 4 20:46:25 nuxbox kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 4 20:46:25 nuxbox kernel: ide: failed opcode was: unknown
Feb 4 20:46:25 nuxbox kernel: hda: drive not ready for command
Feb 4 20:46:25 nuxbox kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 4 20:46:25 nuxbox kernel: ide: failed opcode was: unknown
Feb 4 20:46:25 nuxbox kernel: hda: drive not ready for command

...

Feb 5 23:45:03 nuxbox kernel: ide: failed opcode was: unknown
Feb 5 23:45:03 nuxbox kernel: hdb: drive not ready for command
Feb 5 23:45:03 nuxbox kernel: hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 5 23:45:03 nuxbox kernel: ide: failed opcode was: unknown
Feb 5 23:45:03 nuxbox kernel: hdb: drive not ready for command
Feb 5 23:45:03 nuxbox kernel: hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 5 23:45:03 nuxbox kernel: ide: failed opcode was: unknown
Feb 5 23:45:03 nuxbox kernel: hdb: drive not ready for command
Feb 5 23:45:03 nuxbox kernel: hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 5 23:45:03 nuxbox kernel: ide: failed opcode was: unknown
Feb 5 23:45:03 nuxbox kernel: hdb: drive not ready for command
Feb 5 23:45:03 nuxbox kernel: hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
Feb 5 23:45:03 nuxbox kernel: ide: failed opcode was: unknown

gertdesmet
6th February 2007, 09:36 AM
I tried to boot on Kernel 2.6.18 and the problem doesnt appear.

I also found those lines in the log :


Feb 6 00:28:26 nuxbox kernel: md: Autodetecting RAID arrays.
Feb 6 00:28:26 nuxbox kernel: md: autorun ...
Feb 6 00:28:26 nuxbox kernel: md: ... autorun DONE.
Feb 6 00:28:26 nuxbox kernel: device-mapper: multipath: version 1.0.5 loaded
Feb 6 00:28:26 nuxbox kernel: hda: command error: status=0x51 { DriveReady SeekComplete Error }
Feb 6 00:28:26 nuxbox kernel: hda: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
Feb 6 00:28:26 nuxbox kernel: ide: failed opcode was: unknown
Feb 6 00:28:26 nuxbox kernel: ATAPI device hda:
Feb 6 00:28:26 nuxbox kernel: Error: Illegal request -- (Sense key=0x05)
Feb 6 00:28:26 nuxbox kernel: Illegal mode for this track or incompatible medium -- (asc=0x64, ascq=0x00)
Feb 6 00:28:26 nuxbox kernel: The failed "Read 10" packet command was:
Feb 6 00:28:26 nuxbox kernel: "28 00 00 00 00 00 00 00 08 00 00 00 00 00 00 00 "
Feb 6 00:28:26 nuxbox kernel: end_request: I/O error, dev hda, sector 0
Feb 6 00:28:26 nuxbox kernel: Buffer I/O error on device hda, logical block 0
Feb 6 00:28:26 nuxbox kernel: Buffer I/O error on device hda, logical block 1
Feb 6 00:28:26 nuxbox kernel: Buffer I/O error on device hda, logical block 2
Feb 6 00:28:26 nuxbox kernel: Buffer I/O error on device hda, logical block 3
Feb 6 00:28:26 nuxbox kernel: hda: command error: status=0x51 { DriveReady SeekComplete Error }
Feb 6 00:28:26 nuxbox kernel: hda: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
Feb 6 00:28:26 nuxbox kernel: ide: failed opcode was: unknown
Feb 6 00:28:26 nuxbox kernel: ATAPI device hda:
Feb 6 00:28:26 nuxbox kernel: Error: Illegal request -- (Sense key=0x05)
Feb 6 00:28:26 nuxbox kernel: Illegal mode for this track or incompatible medium -- (asc=0x64, ascq=0x00)
Feb 6 00:28:26 nuxbox kernel: The failed "Read 10" packet command was:
Feb 6 00:28:26 nuxbox kernel: "28 00 00 00 00 00 00 00 02 00 00 00 00 00 00 00 "
Feb 6 00:28:26 nuxbox kernel: end_request: I/O error, dev hda, sector 0
Feb 6 00:28:26 nuxbox kernel: Buffer I/O error on device hda, logical block 0
Feb 6 00:28:26 nuxbox kernel: hda: command error: status=0x51 { DriveReady SeekComplete Error }
Feb 6 00:28:26 nuxbox kernel: hda: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
Feb 6 00:28:26 nuxbox kernel: ide: failed opcode was: unknown
Feb 6 00:28:26 nuxbox kernel: ATAPI device hda:
Feb 6 00:28:26 nuxbox kernel: Error: Illegal request -- (Sense key=0x05)
Feb 6 00:28:26 nuxbox kernel: Illegal mode for this track or incompatible medium -- (asc=0x64, ascq=0x00)
Feb 6 00:28:26 nuxbox kernel: The failed "Read 10" packet command was:
Feb 6 00:28:26 nuxbox kernel: "28 00 00 04 96 80 00 00 02 00 00 00 00 00 00 00 "
Feb 6 00:28:26 nuxbox kernel: end_request: I/O error, dev hda, sector 1202688
Feb 6 00:28:26 nuxbox kernel: hda: command error: status=0x51 { DriveReady SeekComplete Error }
Feb 6 00:28:26 nuxbox kernel: hda: command error: error=0x54 { AbortedCommand LastFailedSense=0x05 }
Feb 6 00:28:26 nuxbox kernel: ide: failed opcode was: unknown
Feb 6 00:28:26 nuxbox kernel: ATAPI device hda:
Feb 6 00:28:26 nuxbox kernel: Error: Illegal request -- (Sense key=0x05)
Feb 6 00:28:26 nuxbox kernel: Illegal mode for this track or incompatible medium -- (asc=0x64, ascq=0x00)
Feb 6 00:28:26 nuxbox kernel: The failed "Read 10" packet command was:
Feb 6 00:28:26 nuxbox kernel: "28 00 00 00 00 00 00 00 02 00 00 00 00 00 00 00 "
Feb 6 00:28:26 nuxbox kernel: end_request: I/O error, dev hda, sector 0
Feb 6 00:28:26 nuxbox kernel: EXT3 FS on dm-0, internal journal
Feb 6 00:28:26 nuxbox kernel: kjournald starting. Commit interval 5 seconds
Feb 6 00:28:26 nuxbox kernel: EXT3 FS on sdb1, internal journal
Feb 6 00:28:26 nuxbox kernel: EXT3-fs: mounted filesystem with ordered data mode.
Feb 6 00:28:26 nuxbox kernel: Adding 2031608k swap on /dev/VolGroup00/LogVol01. Priority:-1 extents:1 across:2031608k

Hopefully someone could help me. Is it worth searching for a solution or should I just reinstall FC6 ?

mchnz
6th February 2007, 10:21 AM
Bad CD in the CDROM drive?

gertdesmet
6th February 2007, 11:49 AM
Hi and thanks for your reply. Unfortunately there is no cd's in the drives.

I managed to get rid of this message by adding the nodma option to the kernel when booting in grub. However I dont know what this option exactly does and I dont understand why this error is spamming the syslog when I dont use the nodma option. I would appreciate if someone could enlighten me.

Thanks in advance :)

Edit : Actually the nodma option didnt solve it. I just looked in /var/log/messages and the error is spamming again. It just took more time to start. I am getting desperate :(

wintermute000
6th February 2007, 12:29 PM
Yep ditto problem here - also I noticed it f-cks up playback of audio CD (leading me to trash some perfectly good audio CDs... grrr), I get a harsh buzzing noise.

However it BURNS fine (audio, data, movie DVDs....) and reads /rips fine as well?>!?>!?!

I think problem is with my mobo controllers, i had to use noprobe kernel line and change BIOS settings to get it to recognise my SATA drives properly and enable full performance (otherwise it was stuck on slow-as-molasses hdparm settings)

gertdesmet
6th February 2007, 12:40 PM
I found out that the problem comes from the JMicron controller on my ASUS P5B-E motherboard.

I found this thread while googling : http://www.linuxquestions.org/questions/showthread.php?t=514483

I have read some stuff about changing the controller mode to AHCI in the BIOS which I did, but the problem still comes back. This is getting too technical for me. :confused:

xen
6th February 2007, 01:36 PM
Yeah this problem is something to do with the JMicron SATA controller support in the 2.6.19-1 kernel. I've tried settting the mode to AHCI in the BIOS but it can't seem to find any of my partitions/mount points, only way to get around this is revert to using the IDE mode.

I'm also experiencing the same behavior with my DVD-ROM drive and also getting those messages!

wintermute000
6th February 2007, 02:14 PM
Yeah its a real downside to 'nix, using really new hardware often causes grief - I too had to muck around with AHCI settings etc. to get decent HDD performance. And sad to say its still not up to XP speeds :(

Xen, have you tried the noprobe kernel boot options, seemed to work for me (forgot how I figured that one out otherwise I'd put you onto the stuff I found which gave me the hint)

e.g. for me, the two HDDs on one of my mobo controllers (hda and hdc) were performing like molasses whilst the one on my other controller was fine. So I did "hda=noprobe hdc=noprobe" in my kernel line + turned on ACHI mode in BIOS, and finally they came up as SATA drives tweakable via hdparm.

Still, rather be a linux junkie than a M$ shill or apple fanboy ;)

xen
6th February 2007, 02:20 PM
I've tried booting with AHCI mode and also in conjunction with the noprobe parameters, but using AHCI it seems to be unable to find my partitions no matter what. It knows the hard drive is there and GRUB loads, but it can't mount the root partition.

It's very similar to these errors here: http://www.nabble.com/Kernel-2.6.19,-jmicron-controller-anche-Core-2-Duo-t3097311.html

It seems that the new kernel has newly named modules and perhaps this has not updated smoothly.

Any ideas?

gertdesmet
6th February 2007, 02:56 PM
Thanks for your replies.

On my side the harddrives are ok. They are both connected on sata.

I guess my most important question is : Is this bad for the system or the hardware ? I mean I can live without my cdrom drives untill this get fixed but I worry about the /var/log/messages getting huge and my brand new hardware ^^.

I 've been reading other forums with people with the same problem and they reverted to Kernel 2.6.18. However in my case I want to use my graphic card to play games but if i revert to 2.6.18 and try to reinstall the kmod-nvidia, yum forces me to install kernel 2.6.19 as a dependency. Whats best ? Staying with 2.6.19 and ignore the messages and the cd/dvd drives madness ? Or reverting to 2.6.18 and try to install the nvidia driver another way ?

Thanks guys.

xen
6th February 2007, 04:04 PM
Thanks for your replies.

On my side the harddrives are ok. They are both connected on sata.

I guess my most important question is : Is this bad for the system or the hardware ? I mean I can live without my cdrom drives untill this get fixed but I worry about the /var/log/messages getting huge and my brand new hardware ^^.

I 've been reading other forums with people with the same problem and they reverted to Kernel 2.6.18. However in my case I want to use my graphic card to play games but if i revert to 2.6.18 and try to reinstall the kmod-nvidia, yum forces me to install kernel 2.6.19 as a dependency. Whats best ? Staying with 2.6.19 and ignore the messages and the cd/dvd drives madness ? Or reverting to 2.6.18 and try to install the nvidia driver another way ?

Thanks guys.

I'm in exactly the same boat as you with the kmod-nvidia problem. Your hardware is fine though, no need to worry about that. I've been living with this problem for over 2 weeks now in hope of a kernel update.

I noticed yesterday that the new stable kernel is 2.6.20 on kernel.org, and so I hope that this will find it's way into FC6 such as the 2.6.19 kernel did, hopefully this might fix our problems?

gertdesmet
6th February 2007, 04:24 PM
Ok cool.

I hope the next kernel update will fix this issue.

Thanks again.

wintermute000
15th February 2007, 12:55 PM
has anyone had any luck with this yet?

I'm finding that even mounted DVDs start stuffing up after a few days (of total inactivity...). Aargh

xen
15th February 2007, 01:24 PM
I have just tried the new Fedora kernel update but the problem apparently still exists. The kernel changelog did mention something about the JMicron controllers, I still feel that it's something to do with the module names changing and the correct ones are not being loaded because of this.

wintermute000
15th February 2007, 09:31 PM
the new udpdate is just security and still part of teh 2.6.19 kernel, 2.6.20 isn't out yet via yum

gertdesmet
21st February 2007, 05:25 PM
I have to check my log files, but since the latest kernel update, I didnt experience any problems with my ide drives.

wintermute000
1st March 2007, 01:07 PM
nope, the kernel update didn't fix mine. Did it work for you gertdesmet???

if anything it seems to have gotten worse :(

after ~2 years of learning fedora and 'nix, this is probably the most annoying issue I've encountered so far.... coz its the only real issue left I haven't found a fix for :)

gertdesmet
2nd March 2007, 03:17 PM
Actually it seemed better for a couple of days, but the problem is back and I still have the spamming messages as well as the mounted CD icons on my desktop. :(

wintermute000
12th March 2007, 04:16 AM
Updated to kernel 2.6.19-1.2911.6.5.fc6, problems still there

AARGH (bangs head on table in frustration)

does anyone know if you can do something similar to the 'all-generic-ide' boot optione xcept only on one specific slot? (i.e. in my case, hde). Problem is my SATA drives are connected to the same controller and they work fine in SATA (with AHCI mode enabled in bios) so I can't just issue the all-generic-ide option on the entire controller.

wintermute000
15th March 2007, 12:12 PM
I think I've fixed it - turn off DMA to your optical drive!!!

The way I did it was run hdparm in my /etc/rc.local

xen
15th March 2007, 12:36 PM
Won't this botch up DVD playback?

wintermute000
15th March 2007, 12:54 PM
hmm mixed results

- didn't recognize one DVD, but played the other one OK even without DMA.
- those messages are still there, but just occur much less frequently and don't seem to 'hang' the PC like they used to

Truth be told, since it reads data discs OK I'm not fussed, I watch DVDs on my media centre :)

xen
15th March 2007, 01:39 PM
I've just just tried the new 2.6.20-1 Fedora kernel update and the problem still exists - despite numerous references to JMicron fixes in the changelog

http://www.kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.20

Search for 'JMicron' in this document.. perhaps there is something else we need to do?

xen
15th March 2007, 01:47 PM
I just looked in the source of the Fedora kernel update (2.6.20-1.2925.fc6-i686) in the following directory: /usr/src/kernels/2.6.20-1.2925.fc6-i686/drivers/ide/pci where I thought that the JMicron module source would be, only this directory is empty except for a Makefile.

Inside the Makefile the JMicron module is mentioned however: obj-$(CONFIG_BLK_DEV_JMICRON) += jmicron.o

Could it be that it's just not being compiled?

xen
15th March 2007, 01:55 PM
I've done a "modprobe -l" and the kernel only has generic ide support compiled in. Is there an easy way to recompile the current kernel from the headers, with the same configuration but including the JMicron modules?

Thanks in advance,
xen

wintermute000
16th March 2007, 04:23 AM
is that a release kernel or just testing?

unfortunately my knowledge of compiling is limited to ./configure, make, make install ;)

wintermute000
22nd April 2007, 06:50 AM
MInor update: turning off HALDAEMON stops the mess.

Unfortuantley it also stops auto mounting :(

that's about as far as I've gotten....
xen have you had any luck?

wintermute000
28th April 2007, 09:36 AM
Anyone had any luck??

lmo
28th April 2007, 03:47 PM
I had this problem on kernel > 2.6.18 too http://forums.fedoraforum.org/forum/showthread.php?t=151658
My first attempt to fix the problem was to stop the haldaemon service. That worked mostly ok for me, except that without haldaemon, you cannot shutdown from the gui login.

My last solution was to put the cdrom on a different controller cable.

wintermute000
5th May 2007, 11:00 AM
Well there's no choice for P965 chipset users as there is only 1 IDE slot that runs through the bloody Jmicron controller.

Since the 2.6.20-1.2944.fc6, fedora doesn't even detect an optical drive, ditto for 2.6.20-1.2948.fc6..... no idea where to start looking.... I attach a bootup log in case anybody can help!

ganniterix
9th May 2007, 03:24 PM
I have a similar issue with this controller that I have not been able to resolve also

http://forums.fedoraforum.org/forum/showthread.php?t=150975

My problem is a bit more complex in the sense that I am using all the SATA ports (6), and when booting with a kernel 2.6.20, I leaves out some of the drives. If anyone is interested just have a look at my post, maybe someone might get new ideas!

wintermute000
9th May 2007, 11:50 PM
a.) I found out that the 'non detect' issue was caused by the 'noprobe' kernel options I was using. noprobe was required for earlier kernels to properly configure AHCI (otherwise my hard drives were coming up in IDE mode, and running SLOW). I removed noprobe and it detects everything OK. But I still had above errors.

b.) I also get tje DMA invalid errors, but it doesnt' seem to do anything.

c.) I purchased a PATA-SATA converter for my optical drive, which solved my errors. Unfortunately I note you are already using all your SATA drives. NOte I don't have anything plugged into the Jmicron controller (ie the 2 purple ports on the left of the main column of 4 SATA ports).

wt6g
7th August 2007, 06:23 AM
I've just noticed a problem today with F7 and the Execute Disable Bit. If that bit is disabled the system runs fine but if it is enabled, meaning the feature is enabled in the motherboard bios, then haldaemon hogs about 50% of my CPU continuously. Turn it off and everything returns to normal. Not sure if this is a kernel issue or a hal daemon issue, but it's repeatable and probably causes other invisible problems.

I only caught it because I routinely use system monitor and noticed the cpu time running above 0 at idle when I'd made a few changes to BIOS. If you don't run cpu monitor and/or dont check TOP often, you might never notice this and my guess is whatever is going on is going to bother any realtime process. :eek:

/Len