Re: Strange quota entries on Sparc64 2.2.16-RAID

[Gustav]

On Sun, 9 Jul 2000, Jakub Jelinek wrote:

> It has to be patched, so that it reads/writes on sparc64 the same on-disk
> format as does the kernel (ie. although the utilities may be 32bit, it has
> to write several entries using long long).
> Red Hat 6.x quota should work correctly, if you are using some other
> distribution, check if quota is patched for sparc64.

I'm using quota-2.00pre3-2, as shipped with RedHat 6.2. This
shouldn't be a problem if what you state is correct.

Thanks,
-- 
Gustav   http://users.dhp.com/~gustav/pgp.html 
Key fingerprint = 84 88 E8 E1 1F 30 55 1E  8B D8 65 0E C2 09 46 E8
  pub  1024/AB22EFB9 1999/09/01 Gustav <[EMAIL PROTECTED]>

Re: Strange quota entries on Sparc64 2.2.16-RAID

[Jakub Jelinek]

On Sat, Jul 08, 2000 at 11:42:14PM -0400, Gustav wrote:
>   I've started to notice strange entries in my quota tables on the
> one of my RAID-1 partitions on a sparc64 2.2.16-RAID box. Basically,
> the following entries are appearing, more every few days:

What quota package are you using?
It has to be patched, so that it reads/writes on sparc64 the same on-disk
format as does the kernel (ie. although the utilities may be 32bit, it has
to write several entries using long long).
Red Hat 6.x quota should work correctly, if you are using some other
distribution, check if quota is patched for sparc64.

Jakub

Strange quota entries on Sparc64 2.2.16-RAID

[Gustav]

I've started to notice strange entries in my quota tables on the
one of my RAID-1 partitions on a sparc64 2.2.16-RAID box. Basically,
the following entries are appearing, more every few days:

[root@durden /root]# repquota /usr
Block limits   File limits
Userusedsofthard  graceused  soft  hard  grace
root  --  593660   0   0  37291 0 0
news  --  92   0   0  2 0 0
1249  --   1   0   0  0 0 0
1257  --   0   0   0  1 0 0
1259  --   1   0   0  0 0 0
1262  --   0   0   0  1 0 0
1264  --   1   0   0  0 0 0
1267  --   0   0   0  1 0 0
1269  --   1   0   0  0 0 0

There are not, and have never been users with these UIDs.
Creating a user with one of these UIDs does not result in a user with
an above quota entry. There are no files owned by any of these users,
according to find. My question is if this may be a sign of corruption in
the software RAID, or some other sort of file system anomaly. Has nayone
seen anything like this before?

Just to confirm, I am using a 2.2.16 kernel patched with
raid-2.2.16-A0, raidtools-0.90-6 on a sparc64 platform.

--
Gustav   http://users.dhp.com/~gustav/pgp.html
Key fingerprint = 84 88 E8 E1 1F 30 55 1E  8B D8 65 0E C2 09 46 E8
  pub  1024/AB22EFB9 1999/09/01 Gustav <[EMAIL PROTECTED]>

Re: linux software raid

[Darron Froese]

On Sat, 8 Jul 2000, Bill Jackson wrote:

> BINGO!! you hit it on the head...in /var/log/messages, I have lines
> containing md but they show 0.36.6 as the version, not 0.90.0.  so, Ill get
> the patches from the location you sent me, but I would like instructions on
> how to install the patch...
> >
> > 

Download the patch and copy it to the kernel source directory at
/usr/src/linux.

#patch -p1 < patch-name-here

should do the trick and patch the kernel source with the updated RAID.

Then recompile the kernel (making sure that you select all the RAID
options) and install it.

That should do the trick.
--
Darron
[EMAIL PROTECTED]

Re: linux software raid

[Bill Jackson]

BINGO!! you hit it on the head...in /var/log/messages, I have lines containing
md but they show 0.36.6 as the version, not 0.90.0.  so, Ill get the
patches from the location you sent me, but I would like instructions on
how to install the patch...
 
Darron Froese wrote:
On Sat, 8 Jul 2000, Bill Jackson wrote:
> I am trying to use mkraid to setup a software raid array.  I
have 3 * 4G
> drives with a two partitions each.  the first partition is a
64 byte
> block for an Apple Partition Map, the rest is a unix partition. 
I have
> setup the /etc/raidtab that is attached and when I try mkraid /dev/md0,
> it sees each of the disks/partitions it should be seeing, then aborts
> and tells me to look in syslog and /proc/mdstat for potential clues. 
I
> have attached my raidtab file and my proc/mdstat file and hope someone
> can point me in the right direction.
Bill,
It looks like you're using linux on a Macintosh (YellowDogLinux or
LinuxPPC) and if so, you have to make sure that your raidtools are
in sync
with the version of RAID that's in your kernel.
I had *tons* of problems setting up a RAID on my YDL box the first time
because they were not synched. The userspace RAID tools were for the
new
improved RAID code while the kernel was compiled with the old RAID
code.
For whatever reason the raid creation would fail and I would be told
to
check syslog and mdstat for clues but none could be found.
Take a look at your dmesg (or look at your /var/log/messages at the
time
of your last reboot) to see if you can see a line that looks like
this:
kernel: md driver 0.90.0 MAX_MD_DEVS=256, MAX_REAL=12
If it says that, then you're running the new improved RAID 0.90 code
- if
it doesn't say that then you're running the older RAID code.
To get the new RAID code installed you'll have to download the 2.2.16
kernel source and patch it with mingo's RAID patches - they're located
here:

Let me know if you need any help or more info.
--
Darron
[EMAIL PROTECTED]

-- 
UNIX *is* user friendly.  It is just very picky about who it's friends are.
 

Re: linux software raid

[Bill Jackson]

sorry about that.  I am using linux kernel 2.2.15, raidtools 0.90.0 and no raid
patch.  I thought raid was in the kernel?  i configured the kernel with raid
support.  i still need to apply a patch?


Mike Black wrote:

> You don't say what version of anything you are running.
> What version of Linux?
> What version of raidtools?
> What RAID patch?
> My guess would be that you're using raidtool-0.90 and a stock Linux kernel
> (2.2.16?) without the 0.90 RAID kernel patch
> - Original Message -
> From: "Bill Jackson" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Saturday, July 08, 2000 3:36 PM
> Subject: linux software raid
>
> I am trying to use mkraid to setup a software raid array.  I have 3 * 4G
> drives with a two partitions each.  the first partition is a 64 byte
> block for an Apple Partition Map, the rest is a unix partition.  I have
> setup the /etc/raidtab that is attached and when I try mkraid /dev/md0,
> it sees each of the disks/partitions it should be seeing, then aborts
> and tells me to look in syslog and /proc/mdstat for potential clues.  I
> have attached my raidtab file and my proc/mdstat file and hope someone
> can point me in the right direction.
>
> --
> UNIX *is* user friendly.  It is just very picky about who it's friends
> are.
>
> 
> 
>
> # Sample raid-5 configuration
> raiddev /dev/md0
> raid-level 5
> nr-raid-disks 3
> chunk-size 128
>
> # Parity placement algorithm
> #parity-algorithm left-asymmetric
> # the best one for maximum performance:
> parity-algorithm left-symmetric
> #parity-algorithm right-asymmetric
> #parity-algorithm right-symmetric
>
> # Spare disks for hot reconstruction
> #nr-spare-disks 0
>
> device /dev/sdb2
> raid-disk 0
> device /dev/sdc2
> raid-disk 1
> device /dev/sdd2
> raid-disk 2
>
> 
> 
>
> Personalities : [1 linear] [2 raid0] [3 raid1] [4 raid5]
> read_ahead not set
> md0 : inactive
> md1 : inactive
> md2 : inactive
> md3 : inactive

--
UNIX *is* user friendly.  It is just very picky about who it's friends are.

Failure autodetecting raid0 partitions

[Anders Qvist]

I have a 2.2.11+intl+raid0.90 successfully mounting its ext2 root file
system off /dev/md0, which is autodetected by the kernel. A 2.4-test2
kernel compiled with CONFIG_AUTODETECT_RAID fails to autodetect my 
partitons when I write it to a floppy and boot it. It just says
autodetecting RAID arrays ... autorun DONE.

There is probably something I don't know. I'd be grateful if someone told
me what it was. NB: I'm not on the list.

System info:

PC: i686, 440LX, NCR 53c875 (rev 3), 2x FUJITSU MAB3045SP.

   Device BootStart   EndBlocks   Id  System
/dev/sda1   * 115 65534   83  Linux
/dev/sda21630 65565   82  Linux swap
/dev/sda331   434   17658845  Extended
/dev/sda4   435  1017   2548293   83  Linux
/dev/sda531   434   1765853   fd  Linux raid autodetect

   Device BootStart   EndBlocks   Id  System
/dev/sdb1 115 65534   82  Linux swap
/dev/sdb21630 65565   82  Linux swap
/dev/sdb331  1017   43141775  Extended
/dev/sdb531   434   1765853   fd  Linux raid autodetect
/dev/sdb6   435   674   1049009   83  Linux
/dev/sdb7   675  1017   1499222   83  Linux

(/dev/sda1 contains the kernel and an emergency boot partition)

Working 2.2.11 kernel autodetect message:

Jul  3 17:22:33 vanity kernel: autodetecting RAID arrays
Jul  3 17:22:33 vanity kernel: (read) sda5's sb offset: 1765760 [events: 003c]
Jul  3 17:22:33 vanity kernel: (read) sdb5's sb offset: 1765760 [events: 003c]
Jul  3 17:22:33 vanity kernel: autorun ...
Jul  3 17:22:33 vanity kernel: considering sdb5 ...
Jul  3 17:22:33 vanity kernel:   adding sdb5 ...
Jul  3 17:22:33 vanity kernel:   adding sda5 ...
Jul  3 17:22:33 vanity kernel: created md0
Jul  3 17:22:33 vanity kernel: bind
Jul  3 17:22:33 vanity kernel: bind
Jul  3 17:22:33 vanity kernel: running: 
Jul  3 17:22:33 vanity kernel: now!
Jul  3 17:22:33 vanity kernel: sdb5's event counter: 003c
Jul  3 17:22:33 vanity kernel: sda5's event counter: 003c
Jul  3 17:22:33 vanity kernel: mask fff0
Jul  3 17:22:33 vanity kernel:  rdev->size: 1765760
Jul  3 17:22:33 vanity kernel:  masked rdev->size: 1765760
Jul  3 17:22:33 vanity kernel:   new md_size: 1765760
Jul  3 17:22:33 vanity kernel:  rdev->size: 1765760
Jul  3 17:22:33 vanity kernel:  masked rdev->size: 1765760
Jul  3 17:22:33 vanity kernel:   new md_size: 3531520
Jul  3 17:22:33 vanity kernel: md0: max total readahead window set to 256k
Jul  3 17:22:33 vanity kernel: md0: 2 data-disks, max readahead per data-disk: 128k
Jul  3 17:22:33 vanity kernel: raid0: looking at sda5
Jul  3 17:22:33 vanity kernel: raid0:   comparing sda5(1765760) with sda5(1765760)
Jul  3 17:22:33 vanity kernel: raid0:   END
Jul  3 17:22:33 vanity kernel: raid0:   ==> UNIQUE
Jul  3 17:22:33 vanity kernel: raid0: 1 zones
Jul  3 17:22:33 vanity kernel: raid0: looking at sdb5
Jul  3 17:22:33 vanity kernel: raid0:   comparing sdb5(1765760) with sda5(1765760)
Jul  3 17:22:33 vanity kernel: raid0:   EQUAL
Jul  3 17:22:33 vanity kernel: raid0: FINAL 1 zones
Jul  3 17:22:33 vanity kernel: zone 0
Jul  3 17:22:33 vanity kernel:  checking sda5 ... contained as device 0
Jul  3 17:22:33 vanity kernel:   (1765760) is smallest!.
Jul  3 17:22:33 vanity kernel:  checking sdb5 ... contained as device 1
Jul  3 17:22:33 vanity kernel:  zone->nb_dev: 2, size: 3531520
Jul  3 17:22:33 vanity kernel: current zone offset: 1765760
Jul  3 17:22:33 vanity kernel: done.
Jul  3 17:22:33 vanity kernel: raid0 : md_size is 3531520 blocks.
Jul  3 17:22:33 vanity kernel: raid0 : conf->smallest->size is 3531520 blocks.
Jul  3 17:22:33 vanity kernel: raid0 : nb_zone is 1.
Jul  3 17:22:33 vanity kernel: raid0 : Allocating 8 bytes for hash.
Jul  3 17:22:33 vanity kernel: md: updating md0 RAID superblock on device
Jul  3 17:22:33 vanity kernel: sdb5 [events: 003d](write) sdb5's sb offset: 1765760
Jul  3 17:22:33 vanity kernel: sda5 [events: 003d](write) sda5's sb offset: 1765760
Jul  3 17:22:33 vanity kernel: .
Jul  3 17:22:33 vanity kernel: ... autorun DONE.

Anders "Quest" Qvist

Re: linux software raid

[Darron Froese]

On Sat, 8 Jul 2000, Bill Jackson wrote:

> I am trying to use mkraid to setup a software raid array.  I have 3 * 4G
> drives with a two partitions each.  the first partition is a 64 byte
> block for an Apple Partition Map, the rest is a unix partition.  I have
> setup the /etc/raidtab that is attached and when I try mkraid /dev/md0,
> it sees each of the disks/partitions it should be seeing, then aborts
> and tells me to look in syslog and /proc/mdstat for potential clues.  I
> have attached my raidtab file and my proc/mdstat file and hope someone
> can point me in the right direction.

Bill,

It looks like you're using linux on a Macintosh (YellowDogLinux or
LinuxPPC) and if so, you have to make sure that your raidtools are in sync
with the version of RAID that's in your kernel.

I had *tons* of problems setting up a RAID on my YDL box the first time
because they were not synched. The userspace RAID tools were for the new
improved RAID code while the kernel was compiled with the old RAID code.
For whatever reason the raid creation would fail and I would be told to
check syslog and mdstat for clues but none could be found.

Take a look at your dmesg (or look at your /var/log/messages at the time
of your last reboot) to see if you can see a line that looks like
this:

kernel: md driver 0.90.0 MAX_MD_DEVS=256, MAX_REAL=12

If it says that, then you're running the new improved RAID 0.90 code - if
it doesn't say that then you're running the older RAID code.

To get the new RAID code installed you'll have to download the 2.2.16
kernel source and patch it with mingo's RAID patches - they're located
here:



Let me know if you need any help or more info.
--
Darron
[EMAIL PROTECTED]

linux software raid

[Bill Jackson]

I am trying to use mkraid to setup a software raid array.  I have 3 * 4G
drives with a two partitions each.  the first partition is a 64 byte
block for an Apple Partition Map, the rest is a unix partition.  I have
setup the /etc/raidtab that is attached and when I try mkraid /dev/md0,
it sees each of the disks/partitions it should be seeing, then aborts
and tells me to look in syslog and /proc/mdstat for potential clues.  I
have attached my raidtab file and my proc/mdstat file and hope someone
can point me in the right direction.

--
UNIX *is* user friendly.  It is just very picky about who it's friends
are.



# Sample raid-5 configuration
raiddev /dev/md0
raid-level  5
nr-raid-disks   3
chunk-size  128

# Parity placement algorithm
#parity-algorithm   left-asymmetric
# the best one for maximum performance:
parity-algorithmleft-symmetric
#parity-algorithm   right-asymmetric
#parity-algorithm   right-symmetric

# Spare disks for hot reconstruction
#nr-spare-disks 0

device  /dev/sdb2
raid-disk   0
device  /dev/sdc2
raid-disk   1
device  /dev/sdd2
raid-disk   2


Personalities : [1 linear] [2 raid0] [3 raid1] [4 raid5]
read_ahead not set
md0 : inactive
md1 : inactive
md2 : inactive
md3 : inactive

Re: Raid1 and Journaling

[Luca Berra]

On Fri, Jul 07, 2000 at 05:07:05PM -0500, Craig Servin wrote:
> 
> Do either ext3 or Reiserfs work with a Raid1?

not on 2.2
L.

-- 
Luca Berra -- [EMAIL PROTECTED]
Communication Media & Services S.r.l.

RE: help: read-ahead not set: what is it???

[Abhishek Khaitan]

can u send your raid configuration file(s)? maybe, I will be able to help
then...

> -Original Message-
> From: Sandro Dentella [SMTP:[EMAIL PROTECTED]]
> Sent: Thursday, July 06, 2000 2:23 AM
> To:   [EMAIL PROTECTED]
> Subject:  help: read-ahead not set: what is it???
> 
> Hi,
> 
>I'm trying to configure raid1 w/ 2 disks /dev/hda7 /dev/hda8 but I get:
> 
>mkraid version 0.36.4
>parsing configuration file
>mkraid: aborted
>   
>cat /proc/mdstat:
>
>personalities: [1 linear] [2 raid0] [3raid 1]
>read_ahead not set<< 
>md0 : inctive
> 
>what does read_ahead mean (I modprobed raid1, is there something else
> I'm
>disregarding?), sorry, but I need a hint very soon (if possible ;-)
> 
> sandro
> *:-)
> 
> 
> 
> 
> 
> -- 
> Sandro Dentella  *:-)
> e-mail: [EMAIL PROTECTED] [EMAIL PROTECTED]