本文转载于 https://bugs.launchpad.net/ubuntu/+source/dmraid/+bug/310928
Bug Description
Dist: Ubuntu 8.10 Server
Hardware: System Sunfire 2250
root@lolly:~# dmraid -ay
RAID set "isw_hgbhachhf_
RAID set "isw_hgbhachhf_
RAID set "isw_hgbhachhf_
root@lolly:~# dpkg -l | grep dmraid
ii dmraid 1.0.0.rc14-
ii libdmraid1.0.0.rc14 1.0.0.rc14-
root@lolly:~# dmraid -r
/dev/sdb: isw, "isw_hgbhachhf", GROUP, ok, 1953525165 sectors, data@ 0
/dev/sda: isw, "isw_hgbhachhf", GROUP, ok, 1953525165 sectors, data@ 0
root@lolly:~# dmraid -b
/dev/sdb: 1953525168 total, "GTF000PAKD6J8F"
/dev/sda: 1953525168 total, "GTF000PAKG99LF"
root@lolly:~# dmraid -V
dmraid version: 1.0.0.rc14 (2006.11.08) shared
dmraid library version: 1.0.0.rc14 (2006.11.08)
device-mapper version: 4.14.0
root@lolly:~# dmraid -s
*** Group superset isw_hgbhachhf
--> Active Subset
name : isw_hgbhachhf_
size : 1953519616
stride : 128
type : mirror
status : ok
subsets: 0
devs : 2
spares : 0
root@lolly:~# dmraid -rD -d -d -d -v -v -v
WARN: locking /var/lock/
NOTICE: /dev/sdb: asr discovering
NOTICE: /dev/sdb: ddf1 discovering
NOTICE: /dev/sdb: hpt37x discovering
NOTICE: /dev/sdb: hpt45x discovering
NOTICE: /dev/sdb: isw discovering
NOTICE: writing metadata file "sdb_isw.dat"
NOTICE: writing offset to file "sdb_isw.offset"
NOTICE: writing size to file "sdb_isw.size"
NOTICE: /dev/sdb: isw metadata discovered
NOTICE: /dev/sdb: jmicron discovering
NOTICE: /dev/sdb: lsi discovering
NOTICE: /dev/sdb: nvidia discovering
NOTICE: /dev/sdb: pdc discovering
NOTICE: /dev/sdb: sil discovering
NOTICE: /dev/sdb: via discovering
NOTICE: /dev/sda: asr discovering
NOTICE: /dev/sda: ddf1 discovering
NOTICE: /dev/sda: hpt37x discovering
NOTICE: /dev/sda: hpt45x discovering
NOTICE: /dev/sda: isw discovering
NOTICE: writing metadata file "sda_isw.dat"
NOTICE: writing offset to file "sda_isw.offset"
NOTICE: writing size to file "sda_isw.size"
NOTICE: /dev/sda: isw metadata discovered
NOTICE: /dev/sda: jmicron discovering
NOTICE: /dev/sda: lsi discovering
NOTICE: /dev/sda: nvidia discovering
NOTICE: /dev/sda: pdc discovering
NOTICE: /dev/sda: sil discovering
NOTICE: /dev/sda: via discovering
INFO: RAID devices discovered:
/dev/sdb: isw, "isw_hgbhachhf", GROUP, ok, 1953525165 sectors, data@ 0
/dev/sda: isw, "isw_hgbhachhf", GROUP, ok, 1953525165 sectors, data@ 0
WARN: unlocking /var/lock/
Everything looks as it should.
Pull One Drive from the mirror and you get:
Giving up waiting for root device. Common Problems:
- Boot args (cat /proc/cmdline)
- Check rootdelay= (did the system wait long enough?)
- Check root= (did the system wait for the right device?)
- Missing modules (cat /proc/modules; ls /dev)
ALERT! /dev/mapper/
Dropping to shell!
BusyBox v1.10.2 ..
(initramfs) dmraid -ay
ctrl-d
ERROR: isw: unsupported map state 0x2 on /dev/sda for Mirror
ERROR: adding /dev/sda to RAID set "isw_hgbhachhf"
ERROR: removing RAID set "isw_hgbhachhf"
Related branches
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2008-12-26: | #1 |
Dan Eicher (dhe) wrote on 2008-12-29: Re: [Bug 310928] Re: boot fails after a one drive failure in a mirror | #2 |
Will rc15 work on 8.10?
The reason I ask, is I want to run an oracle DB server on this box.
Oracle is very pick as to all the bits and pieces it needs to install.
I've found docs for installing Oracle 11g on Ubuntu 8.10 - but
jaunty is a little close to the bleeding edge.
Thanks,
Dan
Giuseppe Iuculano wrote:
> Hi,
>
> Unfortunately dmraid 1.0.0.rc14 doesn't support intel degraded raid
> disk, you should try rc15 version on jaunty.
>
> Cheers,
>
> Giuseppe.
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2008-12-29: | #3 |
Hi,
Dan Eicher ha scritto:
> Will rc15 work on 8.10?
>
> The reason I ask, is I want to run an oracle DB server on this box.
> Oracle is very pick as to all the bits and pieces it needs to install.
> I've found docs for installing Oracle 11g on Ubuntu 8.10 - but
> jaunty is a little close to the bleeding edge.
I uploaded rc15 on my PPA, you can get it adding in your sources.list the
following lines:
deb http://
deb-src http://
Cheers,
Giuseppe.
Dan Eicher (dhe) wrote on 2009-01-05: | #4 |
I think there might be some issue with the upgrade to rc15 on intrepid.
I added the addition sources as indicated.
Then:
apt-get update
apt-get upgrade dmraid -
One package being held back.
apt-get install dmraid -
dmraidrc15 seem to install fine.
Reboot.
No block device found (printed 4 times).
Messages about searching usb and sda
Then:
Gave up waiting for root device. Common problems:
[insert standard text here]
ALERT! /dev/mapper/
shell!
Giuseppe Iuculano wrote:
[...]
> Hi,
>
> Dan Eicher ha scritto:
>> Will rc15 work on 8.10?
>>
>> The reason I ask, is I want to run an oracle DB server on this box.
>> Oracle is very pick as to all the bits and pieces it needs to install.
>> I've found docs for installing Oracle 11g on Ubuntu 8.10 - but
>> jaunty is a little close to the bleeding edge.
>
> I uploaded rc15 on my PPA, you can get it adding in your sources.list the
> following lines:
>
> deb http://
> deb-src http://
>
> Cheers,
> Giuseppe.
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
Dan Eicher (dhe) wrote on 2009-01-15: | #5 |
(initramfs) ls -la /dev/mapper
crw-rw---- 1 0 0 10, 60 control
drwxr-xr-x 8 0 0 13380 ..
drwxr-xr-x 2 0 0 60 .
Dan H. Eicher wrote:
[...]
> I think there might be some issue with the upgrade to rc15 on intrepid.
>
> I added the addition sources as indicated.
> Then:
> apt-get update
> apt-get upgrade dmraid -
> One package being held back.
> apt-get install dmraid -
> dmraidrc15 seem to install fine.
> Reboot.
> No block device found (printed 4 times).
> Messages about searching usb and sda
> Then:
> Gave up waiting for root device. Common problems:
> [insert standard text here]
>
> ALERT! /dev/mapper/
> shell!
>
>
>
>
> Giuseppe Iuculano wrote:
>> Hi,
>>
>> Dan Eicher ha scritto:
>>> Will rc15 work on 8.10?
>>>
>>> The reason I ask, is I want to run an oracle DB server on this box.
>>> Oracle is very pick as to all the bits and pieces it needs to install.
>>> I've found docs for installing Oracle 11g on Ubuntu 8.10 - but
>>> jaunty is a little close to the bleeding edge.
>>
>> I uploaded rc15 on my PPA, you can get it adding in your sources.list the
>> following lines:
>>
>> deb http://
>> deb-src http://
>>
>> Cheers,
>> Giuseppe.
>>
>
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
Dan Eicher (dhe) wrote on 2009-01-15: | #6 |
(initramfs) dmraid -ay
RAID set "isw_hgbhachhf_
RAID set "isw_hgbhachhf_
RAID set "isw_hgbhachhf_
(initramfs) ls -la /dev/mapper
crw-rw---- 1 0 0 10, 60 control
brw-rw---- 1 0 6 254, 0 isw_hgbhachhf_
brw-rw---- 1 0 6 254, 1 isw_hgbhachhf_
brw-rw---- 1 0 6 254, 2 isw_hgbhachhf_
drwxr-xr-x 8 0 0 13380 ..
drwxr-xr-x 2 0 0 120 .
Giuseppe Iuculano wrote:
> Dan H. Eicher ha scritto:
>> (initramfs) ls -la /dev/mapper
>> crw-rw---- 1 0 0 10, 60 control
>> drwxr-xr-x 8 0 0 13380 ..
>> drwxr-xr-x 2 0 0 60 .
>
>
> Ok, do also dmraid -ay and after again ls -la /dev/mapper
>
>
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-15: | #7 |
Dan Eicher ha scritto:
[...]
> (initramfs) dmraid -ay
> RAID set "isw_hgbhachhf_
> RAID set "isw_hgbhachhf_
> RAID set "isw_hgbhachhf_
>
> (initramfs) ls -la /dev/mapper
> crw-rw---- 1 0 0 10, 60 control
> brw-rw---- 1 0 6 254, 0 isw_hgbhachhf_
> brw-rw---- 1 0 6 254, 1 isw_hgbhachhf_
> brw-rw---- 1 0 6 254, 2 isw_hgbhachhf_
> drwxr-xr-x 8 0 0 13380 ..
> drwxr-xr-x 2 0 0 120 .
>
Ok, so I suspect something goes wrong with the udev rule...
Please, can you paste me the output of:
udevadm info --query=all --name=sda
udevadm info --query=all --name=sdb
dpkg -l | grep dmraid
Cheers,
Giuseppe.
Dan Eicher (dhe) wrote on 2009-01-16: Re: [Fwd: Re: [Bug 310928] Re: boot fails after a one drive failure in a mirror] | #8 |
Sorry for the delay - both NIC's on this motherboard died at the same
time and I had to get a motherboard swap.
:~# udevadm info --query=all --name=sda
P: /devices/
N: sda
S: disk/by-
S: disk/by-
S: disk/by-
S: block/8:0
E: ID_VENDOR=ATA
E: ID_MODEL=
E: ID_REVISION=GKAO
E: ID_SERIAL=
E: ID_SERIAL_
E: ID_TYPE=disk
E: ID_BUS=scsi
E: ID_ATA_
E: ID_PATH=
E: ID_FS_USAGE=raid
E: ID_FS_TYPE=
E: ID_FS_VERSION=
E: ID_FS_UUID=
E: ID_FS_UUID_ENC=
E: ID_FS_LABEL=
E: ID_FS_LABEL_ENC=
E: ID_FS_LABEL_SAFE=
~# udevadm info --query=all --name=sdb
P: /devices/
N: sdb
S: disk/by-
S: disk/by-
S: disk/by-
S: block/8:16
E: ID_VENDOR=ATA
E: ID_MODEL=
E: ID_REVISION=GKAO
E: ID_SERIAL=
E: ID_SERIAL_
E: ID_TYPE=disk
E: ID_BUS=scsi
E: ID_ATA_
E: ID_PATH=
E: ID_FS_USAGE=raid
E: ID_FS_TYPE=
E: ID_FS_VERSION=
E: ID_FS_UUID=
E: ID_FS_UUID_ENC=
E: ID_FS_LABEL=
E: ID_FS_LABEL_ENC=
E: ID_FS_LABEL_SAFE=
~# dpkg -l | grep dmraid
ii dmraid 1.0.0.rc15-
ii libdmraid1.0.0.rc14 1.0.0.rc14-
ii libdmraid1.0.0.rc15 1.0.0.rc15-
Dan H. Eicher wrote:
[...]
>
>
> -------- Original Message --------
> Subject: Re: [Bug 310928] Re: boot fails after a one drive failure in a
> mirror
> Date: Thu, 15 Jan 2009 15:52:15 -0000
> From: Dan Eicher <email address hidden>
> Reply-To: Bug 310928 <email address hidden>
> To: <email address hidden>
> References: <email address hidden>
>
>
> (initramfs) dmraid -ay
> RAID set "isw_hgbhachhf_
> RAID set "isw_hgbhachhf_
> RAID set "isw_hgbhachhf_
>
> (initramfs) ls -la /dev/mapper
> crw-rw---- 1 0 0 10, 60 control
> brw-rw---- 1 0 6 254, 0 isw_hgbhachhf_
> brw-rw---- 1 0 6 254, 1 isw_hgbhachhf_
> brw-rw---- 1 0 6 254, 2 isw_hgbhachhf_
> drwxr-xr-x 8 0 0 13380 ..
> drwxr-xr-x 2 0 0 120 .
>
>
>
> Giuseppe Iuculano wrote:
>> Dan H. Eicher ha scritto:
>>> (in...
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-16: | #9 |
Dan Eicher ha scritto:
> Sorry for the delay - both NIC's on this motherboard died at the same
> time and I had to get a motherboard swap.
I need some debug logs.
Extract the attached .tar.gz, and you have a dmraid-debug directory
cd dmraid-debug
cp -a sbin/* /sbin/
cp -a 85-dmraid.rules /etc/udev/rules.d/
cat dmraid.
update-initramfs -u
Reboot, and you will get /dev/dmraid.log and /dev/dmraid-
Cheers,
Giuseppe.
Dan Eicher (dhe) wrote on 2009-01-19: | #10 |
Giuseppe Iuculano wrote:
[...]
> Dan Eicher ha scritto:
>> Sorry for the delay - both NIC's on this motherboard died at the same
>> time and I had to get a motherboard swap.
>
> I need some debug logs.
> Extract the attached .tar.gz, and you have a dmraid-debug directory
>
> cd dmraid-debug
> cp -a sbin/* /sbin/
> cp -a 85-dmraid.rules /etc/udev/rules.d/
> cat dmraid.
> update-initramfs -u
>
> Reboot, and you will get /dev/dmraid.log and /dev/dmraid-
>
> Cheers,
> Giuseppe.
>
>
> ** Attachment added: "dmraid-
> http://
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-19: | #11 |
Dan Eicher ha scritto:
>> ** Attachment added: "dmraid-
>> http://
Please attach also /dev/dmraid-
Cheers,
Giuseppe.
Dan Eicher (dhe) wrote on 2009-01-20: | #12 |
Giuseppe Iuculano wrote:
> Dan Eicher ha scritto:
>
>>> ** Attachment added: "dmraid-
>>> http://
>
> Please attach also /dev/dmraid-
>
> Cheers,
> Giuseppe.
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
description: | updated |
Changed in dmraid: | |
status: | New → Confirmed |
Dan Eicher (dhe) wrote on 2009-01-23: | #13 |
FYI - Despite upgrading to this newer version - when I reboot, the machine still shows in the BIOS as RAID group is in a status of rebuild.
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-24: | #14 |
From ataraid mailing list:
The handling of group RAID sets has changed in rc15 in order to allow
for eg. isw RAID volumes to be selected, hence this is intended.
Wildcard support needs adding in order to allow for easier name
selection.
Patches welcome ;-)
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-29: | #15 |
Hi,
I've uploaded a snapshot dmraid version on my PPA. Could you try that version please?
For intrepid:
deb http://
deb-src http://
For jaunty:
deb http://
deb-src http://
@Dan: Make sure you have the original /etc/udev/
Cheers,
Giuseppe.
Dan Eicher (dhe) wrote on 2009-01-29: Re: [Bug 310928] Re: dmraid-activate is broken for isw arrays (dmraid rc15) | #16 |
Hi,
I'm very sorry, I had to move this machine into production, so I went
with Software Raid 1.
Dan
Giuseppe Iuculano wrote:
[...]
> Hi,
>
> I've uploaded a snapshot dmraid version on my PPA. Could you try that
> version please?
>
> For intrepid:
> deb http://
> deb-src http://
>
> For jaunty:
> deb http://
> deb-src http://
>
> @Dan: Make sure you have the original /etc/udev/
> before upgrade
>
> Cheers,
> Giuseppe.
>
[...]
--
-------
mailto:
http://
Work Phone : (352) 392-1499 | CISE Department
Work Fax : (352) 392-1220 | CSE 314D
-------
John Ryan (johnryannz) wrote on 2009-01-30: | #17 |
Hi Giuseppe,
I can confirm that your new versions seems to work ok. I had an Intel raid array which the BIOS reported as degraded. In Intrepid 8.10 with rc14 installed, the array made no attempt to repair itself. Installing your version resulted in a clean boot and the array rebuilt itself overnight. The BIOS now report the array as clean.
Thanks!
John
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-30: | #18 |
Great, committed in the git repository and uploaded to Debian experimental.
John Ryan (johnryannz) wrote on 2009-01-30: | #19 |
Hi Giuseppe,
I might have spoken too soon! I just got home from work, turned on my PC and my array no longer exists. The BIOS reports no RAID volumes defined and both my physical disks are listed as offline members. I have no idea whether this problem was caused by dmraid or not. I will try to restore the array from a live CD and let you know how I get on.
John
John Ryan (johnryannz) wrote on 2009-01-30: | #20 |
Well after a fair bit of mucking around, it looks like I have completely lost my array. The BIOS now reports the array status as "Failed", while /dev/sda is listed as a "Non-RAID Disk" and /dev/sdb is a "Member Disk(0)". However, I cannot seem to be able to add /dev/sda back into the array either in the BIOS or with "dmraid -R isw_bbiechciij_
The problem all seemed to stem from the fact that when I first re-added a disk to the array last night (prior to the successful rebuild), the name of the array was changed by the BIOS. After a reboot (or in my case, powering off for a day), I then had two different array names and neither could be started. I had to use "dmraid -x isw_daifjieafd_
Giuseppe Iuculano (giuseppe-iuculano) wrote on 2009-01-30: | #21 |
John Ryan ha scritto:
> The problem all seemed to stem from the fact that when I first re-added
> a disk to the array last night (prior to the successful rebuild), the
> name of the array was changed by the BIOS. After a reboot (or in my
> case, powering off for a day), I then had two different array names and
> neither could be started. I had to use "dmraid -x isw_daifjieafd_
> (from either a live CD or the busybox prompt) to delete the new array,
whit dmraid rc15 you can try dmraid -R to add a new drive into the array, from
man page:
{-R| --rebuild} RAID-set [device-path]
added. For Intel chipset based systems, there are two methods in which a
new drive is added to the system.
1. Using OROM to identify a new drive
rebuild drive.
2. Using dmraid to identify a new drive
as the second parameter.
3. Using hot spare drive
command. Then use the dmraid command to start the rebuild.
> then go into the BIOS and re-add the disk again. Unfortunately, this
> resulted in a new name for the array being generated and so on and so
> forth. This went on until I deleted the array on /dev/sda instead of
> the one on /dev/sdb and as a result ended up stuck with the "Failed"
> array.
>
Anyway I think this is a different issue, intel rebuild is a new features
introduced in rc15 and it is quite experimental.
Cheers,
Giuseppe.
John Ryan (johnryannz) wrote on 2009-01-31: | #22 |
Hi Giuseppe,
I concur that this is most likely unrelated. Unfortunately, after much fiddling around I was forced to abandon my array. None of the above commands worked to add a spare or to rebuild. Fortunately, this is only on a test box anyway. When I rebuild it for production use, I will have to carefully weigh my options between dmraid and mdadm.
Thanks for all your work.
John
PsYcHoK9 (psychok9) wrote on 2009-02-06: | #23 |
Regression bug of dmraid :/
I don't know if i must open a new bug:
I've installed today Jaunty alpha4 and work fine with Ext4 partition but have dmraid_rc15.xxx package that don't work with my Intel ICH10 Raid.
I've typed this and others commands without success:
$ sudo dmraid -ay
Intel etcetc was not activated.
ls /dev/mapper list only "control".
After apt-get purge dmraid, apt-get clean and apt-get autoclean I've installed dmraid_
Plase fix it soon.
My mainboard is P5Q Deluxe.
PsYcHoK9 (psychok9) wrote on 2009-02-10: | #24 |
Last version, dmraid_
Kyle Jones (mutiny32) wrote on 2009-02-10: | #25 |
Fake upstream fix since Launchpad doesn't support Git yet
Kyle Jones (mutiny32) wrote on 2009-02-10: | #26 |
I would hold off, it looks as if a lot of isw issues are being hammered out still.
PsYcHoK9 (psychok9) wrote on 2009-02-11: | #27 |
The problem still with latest dmraid_
I hold my dmraid_1.0.0.rc14! Fortunately was not installed on my main Ubuntu on the raid array (i've tried it on another Ubuntu on single/normal sata disk).
Kyle Jones (mutiny32) wrote on 2009-02-13: | #28 |
Committed fix doesn't fix anything.
Launchpad Janitor (janitor) wrote on 2009-02-13: | #29 |
This bug was fixed in the package dmraid - 1.0.0.rc15-
---------------
dmraid (1.0.0.
* Sync with Debian, introducing a few changes, which will be in Debian's
next revision:
* debian/
before copying each udev rule
* debian/
raid groups and subsets
* debian/
RAID0 activation for all metadata types, and the kernel event handling
setup is present in rc15 anyway (LP: #310928)
* debian/
symbolic link in the install file causes the build to FTBFS
dmraid (1.0.0.rc15-1~exp6) experimental; urgency=low
* [51d68d1] debian/
Metadata is not put in the NORMAL state after dmraid rebuild is
started manually
* [8165a84] debian/
corrupted after rebuilding is started
* [0f588f5] debian/control: Fix Vcs-Browser field to play nice with
cl2vcs
* [d832984] Added libdmraid1.0.0.rc15 symbols file
* [f0e50b8] Added description in 07_isw_
and 08_isw_
* [492e108] debian/
(LP: #310928)
dmraid (1.0.0.rc15-1~exp5) experimental; urgency=low
* [f23f890] debian/patches: Deleted old patch 08_promise-
fixes.dpatch
* [db69b2c] Switch to quilt
* [e40b076] debian/
Ubuntu has the default rules installed into /lib/udev/rules.d, so
refer also to udev rules file in /lib/udev/rules.d
* [fed9520] debian/
for "No RAID disks"
* [c85b78f] debian/
(Closes: #512771), (LP: #320482)
* [8d88132] debian/control: Remove priority and section fields in the
binary package, and fix duplicate-
* [1fd2986] Fixed command-
warning
* [36cd8e5] debian/control: Removed initramfs-tools depends
* [b3eb93f] debian/control: Fixed duplicate-
warning
-- Luke Yelavich <email address hidden> Fri, 13 Feb 2009 12:33:43 +1100
Hi,
Unfortunately dmraid 1.0.0.rc14 doesn't support intel degraded raid disk, you should try rc15 version on jaunty.
Cheers,
Giuseppe.