Bug 680667

Summary: Can't install fedora on SATA SSD
Product: [Fedora] Fedora Reporter: mark gress <mgress>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED WONTFIX QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: urgent Docs Contact:
Priority: unspecified    
Version: 15CC: dbraunwarth, gansalmon, itamar, jonathan, kernel-maint, madhu.chinakonda, minfrin, reb, vanmeeuwen+fedora
Target Milestone: ---   
Target Release: ---   
Hardware: athlon   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-07-11 17:52:04 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Drive initialization error
none
incorrectly detects existing format
none
Can't mount the drive
none
storage.log
none
anaconda log
none
program log
none
no syslog in temp
none
messages file
none
A slightly blurry image of the final shutdown phase
none
File system errors left uncorrected
none
anaconda log
none
program log
none
storage log
none
messages log file none

Description mark gress 2011-02-26 20:28:08 UTC
Created attachment 481187 [details]
Drive initialization error

Description of problem: Anaconda seems to have difficulties with super talent drives.

Version-Release number of selected component (if applicable):  FC14

How reproducible:  Every time

Steps to Reproduce:
1. Boot off CD
2. Attempt install
  
Actual results:  When going through the installation process, no matter what format exists on the drive it is reported as an unknown or msdos file format, and an initialization warning is given.  See screenshots.


Expected results:


Additional info:

Comment 1 mark gress 2011-02-26 20:31:39 UTC
Created attachment 481188 [details]
incorrectly detects existing format

It's been formatted EXT4, this isn't the first attempted install on this drive.  I can use other distros on these drives.

Comment 2 mark gress 2011-02-26 20:44:33 UTC
Created attachment 481190 [details]
Can't mount the drive

This is what happens at the end of the install.  It's trying to tell me something, I just don't know what it says.

It's a FTM32GL25V drive.

It does the same thing with the install DVD.

Comment 3 Chris Lumens 2011-02-27 03:13:30 UTC
Once you hit the first screenshot, can you grab /tmp/storage.log and attach it to this bug?

Comment 4 mark gress 2011-02-27 04:00:26 UTC
Created attachment 481212 [details]
storage.log

Comment 5 David Lehman 2011-02-28 17:29:54 UTC
Please also collect the following log files and attach them individually to this bug report:

  /tmp/program.log
  /tmp/anaconda.log
  /tmp/syslog

Thanks.

Comment 6 mark gress 2011-03-01 03:18:49 UTC
Created attachment 481494 [details]
anaconda log

I'm at the initialization screen for these.(the first screenshot)

Comment 7 mark gress 2011-03-01 03:20:02 UTC
Created attachment 481496 [details]
program log

Comment 8 mark gress 2011-03-01 03:26:18 UTC
Created attachment 481498 [details]
no syslog in temp

There was no syslog file in the tmp folder.  Only one in the etc folder.  it's entire contents fit on the screenshot but I don't know if this is what you need.

Comment 9 David Lehman 2011-03-01 04:20:46 UTC
(In reply to comment #8)
> Created attachment 481498 [details]
> no syslog in temp
> 
> There was no syslog file in the tmp folder.  Only one in the etc folder.  it's
> entire contents fit on the screenshot but I don't know if this is what you
> need.

Ah, must be a live install -- the file will be /var/log/messages.

Comment 10 mark gress 2011-03-06 02:29:55 UTC
Created attachment 482484 [details]
messages file

Yes, it's a live disk.  I'm at this screen of the installer for the messages file upload:

Error processing drive:

pci-0000:01:0b.0-scsi-0:0:0:0
30534MB
ATA STT_FTM32GL25V

This device may need to be reinitialized.

REINITIALIZING WILL CAUSE ALL DATA TO BE LOST!

Comment 11 mark gress 2011-04-28 08:47:19 UTC
Can't install Fedora 15 Beta either.  Are any files from newer releases helpful?

Comment 12 David Lehman 2011-05-03 15:21:46 UTC
The storage.log you attached shows us finding an msdos disklabel on sda. There's nothing there to indicate a drive initialization message would be shown. Are these files all from the same attempt? If not, please clarify what pieces of information are from what attempts and the details of what was in fact on the drive at the start of each attempt.

Comment 13 mark gress 2011-05-04 20:44:45 UTC
The above files are from multiple attempts, and there would've been ext4 formats present.

As best as I can tell, the driver/config for the chip on my NF7S board is corrupting the format.(the board has a silicon image 3112A chip)

If put the drive into another computer(chip:Sil3512), do an F14 install, it works fine consistently.  If I take the drive out and put it into my computer, it'll boot the first time, but upon reboot or shutdown the next time I attempt to boot it only goes to a blank GRUB screen.  The format is lost according to reboot using liveCD.

Right now I'm using this computer with the 3112A chip, on an SSD.  But with an older distro.  All the hardware in question appears to be in working order depending on where they're connected.

Even if it finds an MSDOS label initially, what about the post install phase?  It won't complete the install successfully.

Comment 14 David Lehman 2011-05-04 21:26:39 UTC
I'm still not clear on what's supposed to be on the disk. Is the entire disk filled with an ext4 filesystem? If so, where is grub installed? Or is the disk partitioned with a single ext4 partition, or what?

If the disk's contents are disappearing when you reboot, I'd say the problem is your disk or the kernel or the chipset or whatever -- certainly not the fedora installer. If you want to keep digging anyway, read on.

On a system that is working as expected, collect some info:

 1. figure out which disk device is your SSD
 2. attach /proc/partitions to this bug report, noting which
    disk is the SSD
 3. attach the output of the following command:

      parted /dev/<SSD> print

      where <SSD> is the device corresponding to your SSD,
      eg: parted /dev/sdb print

Thanks.

Comment 15 mark gress 2011-05-07 15:18:49 UTC
What's supposed to be on the disk is what's on there right now.  I just did another install so I could grab those files you requested.  I usually always do a single partition.

While the drive was in the first computer, i went ahead and updated the kernel via YUM, don't know if it will help or not.  I rebooted several times on the first computer, everything seems fine.

I currently have the drive in the second computer now, my expectation is that when I reboot it the format will be destroyed.  Here's the parted information:

[root@localhost user]# parted /dev/sda print
Model: ATA STT_FTM32GL25V (scsi)
Disk /dev/sda: 32.0GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End     Size    Type     File system  Flags
 1      1049kB  32.0GB  32.0GB  primary  ext4         boot

[root@localhost user]# 

And I would agree with you about the problem not being the installer.  That's only how it appeared at the initial creation of the bug report.  Being as I can do an install on another computer with no problem and it will hold the format if I leave it there, but if I put the drive into this computer and either the driver/kernel/setting is nuking the format.  Now, with the problem being a destruction of the format I don't see how I can get you a shutdownlogwhatever file so you can examine the shutdown phase.

The only way I can see to locate the problem(and send you files) would using the installer as a vehicle, because it won't complete the install.  It gives me a superblock error, but only on this computer.(see my third upload file/image "can't mount the drive")

Comment 16 mark gress 2011-05-07 15:59:47 UTC
Alright, I did a 'shutdown' instead of a 'restart' this time.  I'm almost happy about this.  It won't shut itself down.  I took a picture of what it did so that I can show you.  This is the error message I got:

[ 2033.649344] EXT4-fs error (device sda1): ext4_dx_find_entry: inode #799: (commm s01halt) bad entry in directory: inode out of bounds - block=11070offset=0(8192), inode=2099902464, rec_len=4096, name_len=225
/etc/rc0.d/S01halt: line 174: /sbin/fstab-decode: No such file or directory
/etc/rc0.d/S01halt: line 174: /sbin/fstab-decode: No such file or directory
Halting system...
[ 2033.649344] EXT4-fs error (device sda1): ext4_dx_find_entry: inode #799: (commm s01halt) bad entry in directory: inode out of bounds - block=11070offset=0(8192), inode=2099902464, rec_len=4096, name_len=225
/etc/rc0.d/S01halt: line207: /sbin/halt: No such file or directory
/etc/rc0.d/S01hatl: line207: exec: /sbin/halt: cannot execute: No such file or directory

So the filesystem structure must be being destroyed before it can execute the halt command.  Normally my computers completely power down without incedent.

Comment 17 mark gress 2011-05-07 16:02:29 UTC
Created attachment 497542 [details]
A slightly blurry image of the final shutdown phase

I zoomed in, but that made some of the blurs a little worse.

Comment 18 David Lehman 2011-05-18 18:12:04 UTC
To clear up one of the misunderstandings in this report: If the disk contains partitions, then the disk's formatting is a disklabel. The individual partitions may contain different formats, but that is another matter. The most prevalent type of disklabel is an msdos disklabel. So when you see messages about an msdos disklabel (or partition table) on the disk, it suggests that things are as they should be.


More to the point, I still cannot tell exactly what the behavior is that you are reporting. If you can boot from grub on your ssd, then you do no installation or modification to that disk and the next time you try to boot it fails, something is writing to your disk. My guess would be your system's firmware/BIOS. The other possibility is that your ssd is somehow incompatible with your system's firmware or chipset or similar.

Comment 19 mark gress 2011-06-12 13:17:30 UTC
I apologize about the misunderstandings.  I've never seen anything like this so it's been hard to describe.

But I did find something, the drive is usable by any distro I want(including fedora) as long as I have a swap partition.

I just can't do the install on this computer.  If I do that, it doesn't work.  But the error message at install time has changed, I'll upload a new image to show you what I mean.

I have to do an install on a different computer, but when I have a swap partition, it works consistently in this computer.  I can reboot, and it reboots fine.  I can shutdown, and the next time I start up it works.  It doesn't destroy the format and information.

Comment 20 mark gress 2011-06-12 13:22:25 UTC
Created attachment 504308 [details]
File system errors left uncorrected

Unrecoverable Error

Ext4 filesystem check failure on /dev/sda1:

File system errors left uncorrected.

Errors like this usually mean there is a problem with the filesystem that will require user interaction to repair.  Before restarting installation, reboot to rescue mode or another system that allows you to repair the filesystem interactively.  Restart installation after you have corrected the problems on the filesystem.

(The swap file is on /dev/sda2 - don't know if that matters)

Comment 21 David Lehman 2011-06-13 13:44:17 UTC
If you can collect the logs from this last failure I can try to see what's going on. To collect the logs, switch to tty2 and copy the following files to another computer using scp, then attach the files individually to this bug report:

  /tmp/anaconda.log
  /tmp/storage.log
  /tmp/program.log
  /var/log/messages

Comment 22 mark gress 2011-06-15 15:02:09 UTC
Created attachment 504887 [details]
anaconda log

post-failed install.(see 504308, above)

Comment 23 mark gress 2011-06-15 15:02:47 UTC
Created attachment 504888 [details]
program log

Comment 24 mark gress 2011-06-15 15:03:23 UTC
Created attachment 504889 [details]
storage log

Comment 25 mark gress 2011-06-15 15:09:40 UTC
Created attachment 504891 [details]
messages log file

Comment 26 David Lehman 2011-06-15 15:13:47 UTC
Something is wrong with your hardware or else the kernel. There is nothing to be done in anaconda. Here are some errors that typically indicate hardware problems from the system log:


Jun 15 10:50:58 localhost kernel: [  389.825627] SGI XFS with ACLs, security attributes, large block/inode numbers, no debug enabled
Jun 15 10:50:58 localhost kernel: [  389.831683] SGI XFS Quota Management subsystem
Jun 15 10:51:11 localhost kernel: [  403.315174] ata6.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
Jun 15 10:51:11 localhost kernel: [  403.315180] ata6.00: BMDMA2 stat 0x650001
Jun 15 10:51:11 localhost kernel: [  403.315186] ata6.00: failed command: DATA SET MANAGEMENT
Jun 15 10:51:11 localhost kernel: [  403.315195] ata6.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
Jun 15 10:51:11 localhost kernel: [  403.315197]          res 51/04:01:00:00:00/00:00:00:00:00/a0 Emask 0x1 (device error)
Jun 15 10:51:11 localhost kernel: [  403.315201] ata6.00: status: { DRDY ERR }
Jun 15 10:51:11 localhost kernel: [  403.315204] ata6.00: error: { ABRT }
Jun 15 10:51:11 localhost kernel: [  403.315215] ata6.00: device reported invalid CHS sector 0
Jun 15 10:51:11 localhost kernel: [  403.315228] sd 5:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Jun 15 10:51:11 localhost kernel: [  403.315233] sd 5:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
Jun 15 10:51:11 localhost kernel: [  403.315239] Descriptor sense data with sense descriptors (in hex):
Jun 15 10:51:11 localhost kernel: [  403.315242]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
Jun 15 10:51:11 localhost kernel: [  403.315252]         00 00 00 00 
Jun 15 10:51:11 localhost kernel: [  403.315256] sd 5:0:0:0: [sda]  Add. Sense: No additional sense information
Jun 15 10:51:11 localhost kernel: [  403.315261] sd 5:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 00 00 00 3f 00 3f ff c0 00 00
Jun 15 10:51:11 localhost kernel: [  403.315274] end_request: I/O error, dev sda, sector 63
Jun 15 10:51:15 localhost kernel: [  407.436037] ata6.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
Jun 15 10:51:15 localhost kernel: [  407.436044] ata6.00: BMDMA2 stat 0x650001
Jun 15 10:51:15 localhost kernel: [  407.436049] ata6.00: failed command: DATA SET MANAGEMENT
Jun 15 10:51:15 localhost kernel: [  407.436058] ata6.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
Jun 15 10:51:15 localhost kernel: [  407.436060]          res 51/04:01:00:00:00/00:00:00:00:00/a0 Emask 0x1 (device error)
Jun 15 10:51:15 localhost kernel: [  407.436064] ata6.00: status: { DRDY ERR }
Jun 15 10:51:15 localhost kernel: [  407.436067] ata6.00: error: { ABRT }
Jun 15 10:51:15 localhost kernel: [  407.436078] ata6.00: device reported invalid CHS sector 0
Jun 15 10:51:15 localhost kernel: [  407.436089] sd 5:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Jun 15 10:51:15 localhost kernel: [  407.436094] sd 5:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
Jun 15 10:51:15 localhost kernel: [  407.436100] Descriptor sense data with sense descriptors (in hex):
Jun 15 10:51:15 localhost kernel: [  407.436103]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
Jun 15 10:51:15 localhost kernel: [  407.436112]         00 00 00 00 
Jun 15 10:51:15 localhost kernel: [  407.436117] sd 5:0:0:0: [sda]  Add. Sense: No additional sense information
Jun 15 10:51:15 localhost kernel: [  407.436122] sd 5:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 03 91 5d 55 00 28 b0 d8 00 00
Jun 15 10:51:15 localhost kernel: [  407.436135] end_request: I/O error, dev sda, sector 59858261

Comment 27 mark gress 2011-06-15 22:50:00 UTC
Ok bad Kernel or bad hardware.  I know something is wrong either way.

Now, if I can set up the format structure differently and get a different result, then it has to be something in the kernel, right?

Comment 28 Chuck Ebbert 2011-06-25 19:07:39 UTC
(In reply to comment #26)
> Jun 15 10:51:11 localhost kernel: [  403.315186] ata6.00: failed command: DATA
> SET MANAGEMENT

It's trying to do TRIM and failing horribly. Either the drive is defective or it's just incompatible with certain controllers.

Comment 29 MartinKG 2011-07-08 09:46:03 UTC
I have nearly the same issue, I'am running alread Fedora 15 on my system.
But when run disc intensive actions, i get the following error messages in
dmesg.

[ 1343.431787] ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 1343.431792] ata7.00: irq_stat 0x40000001
[ 1343.431796] ata7.00: failed command: DATA SET MANAGEMENT
[ 1343.431803] ata7.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
[ 1343.431805]          res 51/04:00:01:00:00/00:00:00:00:00/e0 Emask 0x1 (device error)
[ 1343.431808] ata7.00: status: { DRDY ERR }
[ 1343.431811] ata7.00: error: { ABRT }
[ 1343.431825] sd 6:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1343.431829] sd 6:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
[ 1343.431834] Descriptor sense data with sense descriptors (in hex):
[ 1343.431836]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
[ 1343.431846]         00 00 00 00 
[ 1343.431850] sd 6:0:0:0: [sda]  Add. Sense: No additional sense information
[ 1343.431854] sd 6:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 07 fd 8c b0 00 00 06 40 00 00
[ 1343.431866] end_request: I/O error, dev sda, sector 134057136
[ 1343.501734] ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 1343.501739] ata7.00: irq_stat 0x40000001
[ 1343.501742] ata7.00: failed command: DATA SET MANAGEMENT
[ 1343.501749] ata7.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
[ 1343.501751]          res 51/04:00:01:00:00/00:00:00:00:00/e0 Emask 0x1 (device error)
[ 1343.501754] ata7.00: status: { DRDY ERR }
[ 1343.501756] ata7.00: error: { ABRT }
[ 1343.501770] sd 6:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1343.501774] sd 6:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
[ 1343.501779] Descriptor sense data with sense descriptors (in hex):
[ 1343.501782]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
[ 1343.501791]         00 00 00 00 
[ 1343.501795] sd 6:0:0:0: [sda]  Add. Sense: No additional sense information
[ 1343.501799] sd 6:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 07 de 8d 20 00 00 00 20 00 00
[ 1343.501811] end_request: I/O error, dev sda, sector 132025632
[ 1355.330393] ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 1355.330398] ata7.00: irq_stat 0x40000001
[ 1355.330402] ata7.00: failed command: DATA SET MANAGEMENT
[ 1355.330409] ata7.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
[ 1355.330411]          res 51/04:00:01:00:00/00:00:00:00:00/e0 Emask 0x1 (device error)
[ 1355.330414] ata7.00: status: { DRDY ERR }
[ 1355.330416] ata7.00: error: { ABRT }
[ 1355.330431] sd 6:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1355.330435] sd 6:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
[ 1355.330440] Descriptor sense data with sense descriptors (in hex):
[ 1355.330443]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
[ 1355.330452]         00 00 00 00 
[ 1355.330456] sd 6:0:0:0: [sda]  Add. Sense: No additional sense information
[ 1355.330460] sd 6:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 06 ea 1d 90 00 00 01 88 00 00
[ 1355.330472] end_request: I/O error, dev sda, sector 116006288
[ 1356.659112] ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 1356.659117] ata7.00: irq_stat 0x40000001
[ 1356.659120] ata7.00: failed command: DATA SET MANAGEMENT
[ 1356.659127] ata7.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
[ 1356.659129]          res 51/04:00:01:00:00/00:00:00:00:00/e0 Emask 0x1 (device error)
[ 1356.659132] ata7.00: status: { DRDY ERR }
[ 1356.659134] ata7.00: error: { ABRT }
[ 1356.659148] sd 6:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1356.659152] sd 6:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
[ 1356.659157] Descriptor sense data with sense descriptors (in hex):
[ 1356.659160]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
[ 1356.659169]         00 00 00 00 
[ 1356.659172] sd 6:0:0:0: [sda]  Add. Sense: No additional sense information
[ 1356.659177] sd 6:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 06 a2 a7 e0 00 00 00 58 00 00
[ 1356.659189] end_request: I/O error, dev sda, sector 111323104
[ 1358.177660] ata7.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
[ 1358.177665] ata7.00: irq_stat 0x40000001
[ 1358.177669] ata7.00: failed command: DATA SET MANAGEMENT
[ 1358.177676] ata7.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
[ 1358.177678]          res 51/04:00:01:00:00/00:00:00:00:00/e0 Emask 0x1 (device error)
[ 1358.177681] ata7.00: status: { DRDY ERR }
[ 1358.177683] ata7.00: error: { ABRT }
[ 1358.177698] sd 6:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1358.177703] sd 6:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
[ 1358.177708] Descriptor sense data with sense descriptors (in hex):
[ 1358.177710]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
[ 1358.177719]         00 00 00 00 
[ 1358.177723] sd 6:0:0:0: [sda]  Add. Sense: No additional sense information
[ 1358.177728] sd 6:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 0d 62 19 50 00 00 00 28 00 00
[ 1358.177740] end_request: I/O error, dev sda, sector 2245

The SSD was already tested on both SATA6 port on the Motherboard and the SATA-Cable
was changed 3 times.

My System:

Processor: i5 3,3 GHz
Memory: G.Skill ECO DIMM Kit 8GB PC3L-12800U CL8-8-8-24 (DDR3L-1600) 1.35V
Power Supply: Seasonic X-Series Fanless X-400FL 400W passiv
Motherboard: ASUS P8P67 Firmware 1850
disc1: SSD Vertex 3 120 GB Firmware 2.08
disc2: Seagate 2TB ST32000641AS
Grafic card: Asus Geforce ENGT520 Silent 1024MB GDDR3 PCIe passiv
TV-Card: Technotrend S2-6400 PCIe
OS: Fedora 15 Linux
Kernelversion: 2.6.38.8-32.fc15.x86_64

any idea ?

Comment 30 Josh Boyer 2011-09-26 14:24:43 UTC
Is this still happening with the latest 2.6.40.x kernels in F15?

Comment 31 mark gress 2011-11-22 21:19:56 UTC
I had a chance to try today with 2.6.41.1, still does it.

But it will shutdown now.(see comment 16)  So something has changed.  But with the drive being wiped, there's no files I can upload to you to compare.

Is there a way to disable TRIM at boot time or something?  A GRUB command or something?

Comment 32 MartinKG 2011-11-24 18:34:06 UTC
first check if your drive supports TRIM with:

hdparm -I /dev/sdc | grep TRIM

You can disable TRIM, you have to delete the option discard
in your /etc/fstab file.

Comment 33 Graham Leggett 2012-04-15 11:47:11 UTC
Having similar problems after an upgrade to FC17 on an SSD disk. The system now refuses to start without going into emergency mode first, forcing an fsck shows all the disks as clean.

The "discard" option is not present in the /etc/fstab, but TRIM is supported by the device:

[root@samantha ~]# hdparm -I /dev/sda | grep TRIM
	   *	Data Set Management TRIM supported (limit unknown)
	   *	Deterministic read data after TRIM

As a result, not sure how to turn TRIM off.

The error looks like this:

Apr 15 09:53:23 samantha kernel: [    8.501067] EXT4-fs (sda1): mounted filesystem with ordered data mode. Opts: (null)
Apr 15 09:53:23 samantha kernel: [    8.667398] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
Apr 15 09:53:23 samantha kernel: [    8.667412] ata1.00: BMDMA stat 0x25
Apr 15 09:53:23 samantha kernel: [    8.667426] ata1.00: failed command: DATA SET MANAGEMENT
Apr 15 09:53:23 samantha kernel: [    8.667453] ata1.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
Apr 15 09:53:23 samantha kernel: [    8.667459]          res 51/04:01:00:00:00/04:00:00:00:00/a0 Emask 0x1 (device error)
Apr 15 09:53:23 samantha kernel: [    8.667473] ata1.00: status: { DRDY ERR }
Apr 15 09:53:23 samantha kernel: [    8.667482] ata1.00: error: { ABRT }
Apr 15 09:53:23 samantha kernel: [    8.667503] ata1.00: device reported invalid CHS sector 0
Apr 15 09:53:23 samantha kernel: [    8.667526] sd 1:0:0:0: [sda]  Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
Apr 15 09:53:23 samantha kernel: [    8.667540] sd 1:0:0:0: [sda]  Sense Key : Aborted Command [current] [descriptor]
Apr 15 09:53:23 samantha kernel: [    8.667555] Descriptor sense data with sense descriptors (in hex):
Apr 15 09:53:23 samantha kernel: [    8.667564]         72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00 
Apr 15 09:53:23 samantha kernel: [    8.667593]         00 00 00 00 
Apr 15 09:53:23 samantha kernel: [    8.667607] sd 1:0:0:0: [sda]  Add. Sense: No additional sense information
Apr 15 09:53:23 samantha kernel: [    8.667622] sd 1:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 01 1f 81 c7 00 3a ff f8 00 00
Apr 15 09:53:23 samantha kernel: [    8.667660] end_request: I/O error, dev sda, sector 18842055

Comment 34 Josh Boyer 2012-07-11 17:52:04 UTC
Fedora 15 has reached it's end of life as of June 26, 2012.  As a result, we will not be fixing any remaining bugs found in Fedora 15.

In the event that you have upgraded to a newer release and the bug you reported is still present, please reopen the bug and set the version field to the newest release you have encountered the issue with.  Before doing so, please ensure you are testing the latest kernel update in that release and attach any new and relevant information you may have gathered.

Thank you for taking the time to file a report.  We hope newer versions of Fedora suit your needs.