Re: FAQ update
On Mon, Aug 07, 2000 at 08:47:47AM -0700, Gregory Leblanc wrote: -Original Message- From: James Manning [mailto:[EMAIL PROTECTED]] Sent: Saturday, August 05, 2000 6:08 AM To: Linux Raid list (E-mail) Subject: Re: FAQ update [Luca Berra] The patches for 2.2.14 and later kernels are at http://people.redhat.com/mingo/raid-patches/. Use the right patch for your kernel, these patches haven't worked on other kernel revisions yet. i'd add: dont use netscape to fetch patches from mingo's site, it hurts use lynx/wget/curl/lftp Yes, *please* *please* *please* I need some clarification on this. I couldn't make lynx work, it chopped off long lines or something. wget works, I've never heard of the other two. Why exactly is NetScrape bad? That server load thing sounds fishy to me... Greg ok, i'll clarify NutScrape may not work for the same reason lynx failed for you redhat server says the file is text/plain so both netscape and lynx fail if you view the file and than save it to a local file. If you Shift-click on netscape or press 'd' on lynx it should work. i don't give a damn about the load on redhat http server, but i don't like receiving tons of mails saying that the patch from mingo site fails for them. L. P.S. someone could suggest mingo gzips the blasted patches : -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: FAQ update
On Fri, Aug 04, 2000 at 01:47:23PM -0700, Gregory Leblanc wrote: Here's a new version, with a couple of changes. What other questions get asked all the time? Greg The patches for 2.2.14 and later kernels are at http://people.redhat.com/mingo/raid-patches/. Use the right patch for your kernel, these patches haven't worked on other kernel revisions yet. i'd add: dont use netscape to fetch patches from mingo's site, it hurts use lynx/wget/curl/lftp -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: FAQ
On Thu, Aug 03, 2000 at 01:34:33PM -0400, James Manning wrote: there is no bzip2 standard in gnu tar, so let's be intelligent and avoid the issue by going with the .gz tarball as a recommendation. -z is standard. from the info page from gnu tar 1.13.17: `--bzip2' `-I' This option tells `tar' to read or write archives through `bzip2'. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: raid-2.2.17-A0 cleanup for LVM
On Thu, Aug 03, 2000 at 07:55:18PM +0200, Andrea Arcangeli wrote: On Thu, 3 Aug 2000, G.W. Wettstein wrote: Grab 2.2.15aa1 or 2.2.17pre11aa2, they have completly reliable LVM (I also switch between it and 2.4.x stock without changing anything). It can be used for production. You can find the separate patches that compose that kernel in the ftp site as well. strange i was never able to run lvm on top ov raid with 2.4 the lvm tools just don't guess what an md device is :( L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: raid5 troubles
On Fri, Jul 21, 2000 at 11:17:18AM +0200, Martin Bene wrote: "dangerous" tools. Bzw, has anyone checked what's different in this tools package in comparison to the 19990824 release? yes it raises the max number of devices per superblock!!! -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Problems configuring RAID on RedHAt Linux 6.2
Gosh, you made abitof a mess On Tue, Jul 18, 2000 at 11:59:11AM -0700, Vivek Singh wrote: I installed Linux on the system. Now when the system boots up, I cannot see either of the 36 GB discs. The raidtools utility (ver 0.90) was installed yes, they are already mirrored you see only 1 36G scsi disk when I first booted the system after installation. I have installed the patch, but with some errors (while installing, I was getting the messages that the patch has already been installed, and I forced it to use the -R It WAS already applied, redhat applies patches to kernels they provide Vendor: MegaRAID Model: LD0 RAID0 8677R Rev: D Type: Direct-Access ANSI SCSI revision: 02 this is your system disk (/dev/sda) Detected scsi disk sda at scsi0, channel 1, id 0, lun 0 Vendor: MegaRAID Model: LD1 RAID1 34731R Rev: D Type: Direct-Access ANSI SCSI revision: 02 this is the other two in raid1 (/dev/sdb) please, stick to the redhat kernel (uhm get a newer one from updates.redhat.com) stick to hardware raid since you have itavailable and let software raid alone regards L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Raid1 and Journaling
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: MD_BOOT is _flawed_
On Wed, Jun 14, 2000 at 07:31:28PM +0200, Ingo Molnar wrote: initrd is much more capable than the kernel's command line. The kernel command line is limited to 80 characters - which is awfully few. We do I'd like to comment on this, kernel command line _is_ 256 characters redhat supplied lilo is 80 characters, but this is a bug, patch boot.S or use lilo-21.4.3. regards, Luca P.S. this has nothing to do with the md_boot argument L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: mkraid 2.4.0-test1-ac10 fails
On Thu, Jun 08, 2000 at 01:14:47PM +1000, Neil Brown wrote:
Hmmm. works for me.
hello, i have a similar problem but a different bug
Linux version 2.4.0-test1-ac12 (root@Moskowskaya) (gcc driver version pgcc-2.95.2
19991024 (release) executing gcc version egcs-2.91.66) #1 Sat Jun 10 20:00:52 CEST 2000
+ patch-2.4.0test1-B2P (hammered on it)
+ linux-2.3.99-pre9-reiserfs-3.6.6-patch.gz (applied cleanly)
[vc/2]root@Moskowskaya:~# raidstart /dev/md1
(read) scsi/host1/bus0/target4/lun0/part2's sb offset: 17899456 [events: 0005]
(read) scsi/host1/bus0/target5/lun0/part2's sb offset: 17899456 [events: 0005]
(read) scsi/host1/bus0/target6/lun0/part2's sb offset: 17899456 [events: 0005]
autorun ...
considering scsi/host1/bus0/target6/lun0/part2 ...
adding scsi/host1/bus0/target6/lun0/part2 ...
adding scsi/host1/bus0/target5/lun0/part2 ...
adding scsi/host1/bus0/target4/lun0/part2 ...
created md1
bindscsi/host1/bus0/target4/lun0/part2,1
bindscsi/host1/bus0/target5/lun0/part2,2
bindscsi/host1/bus0/target6/lun0/part2,3
running:
scsi/host1/bus0/target6/lun0/part2scsi/host1/bus0/target5/lun0/part2scsi/host1/bus0/target4/lun0/part2
now!
scsi/host1/bus0/target6/lun0/part2's event counter: 0005
scsi/host1/bus0/target5/lun0/part2's event counter: 0005
scsi/host1/bus0/target4/lun0/part2's event counter: 0005
md1: max total readahead window set to 1024k
md1: 2 data-disks, max readahead per data-disk: 512k
raid5: device scsi/host1/bus0/target6/lun0/part2 operational as raid disk 2
raid5: device scsi/host1/bus0/target5/lun0/part2 operational as raid disk 1
raid5: device scsi/host1/bus0/target4/lun0/part2 operational as raid disk 0
kernel BUG at raid5.c:214!
(actually it looks like it is the BUG() statement at line 213
210 static void __put_free_stripe (raid5_conf_t *conf, struct stripe_head *sh)
211 {
212 if (atomic_read(sh-count) != 0)
213 BUG();
214 CHECK_DEVLOCK();
215 CHECK_SHLOCK(sh);
216 clear_bit(STRIPE_LOCKED, sh-state);
217 sh-free_next = conf-free_sh_list;
218 conf-free_sh_list = sh;
219 atomic_inc(conf-nr_free_sh);
220 }
attached are ksymoops and .config
thanks in advance for any help you can give me
L.
--
Luca Berra -- [EMAIL PROTECTED]
Communication Media Services S.r.l.
ksymoops 2.3.4 on i586 2.4.0-test1-ac12. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.0-test1-ac12/ (default)
-m /boot/System.map-2.4.0-test1-ac12 (specified)
invalid operand:
CPU:0
EIP:0010:[ca9353c8]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010086
eax: 001b ebx: c69d4c00 ecx: 0005 edx: c0263954
esi: c69d5c00 edi: 0006 ebp: ca936150 esp: c4175d60
ds: 0018 es: 0018 ss: 0018
Process raidstart (pid: 2272, stackpage=c4175000)
Stack: ca939735 ca939a5a 00d6 c69d4ed4 c69d5c00 ca93628c c69d5c00 c69d4c00
0cc8 c69d5c00 c4175d8c c69d4c00 0006 007f
ca938b30 c69d5c00 0080 0007 c6f8a900 c5c71c14 c026b0a0 c5c71c00
Call Trace: [ca939735] [ca939a5a] [ca93628c] [ca938b30] [c0131357]
[c0188e48] [c01168ae]
[c0189214] [c023356f] [c0189440] [c01895d0] [c018a44c] [c0137de0]
[c0142aca] [c01567c8]
[c012f55a] [c0137e03] [c013f66f] [c010b334]
Code: 0f 0b 83 c4 0c 8b 83 b4 02 00 00 a8 01 75 19 68 d7 00 00 00
EIP; ca9353c8 [raid5]__put_free_stripe+40/88 =
Trace; ca939735 [raid5].rodata.start+55/d9d
Trace; ca939a5a [raid5].rodata.start+37a/d9d
Trace; ca93628c [raid5]grow_stripes+1b0/1d0
Trace; ca938b30 [raid5]raid5_run+5dc/8fc
Trace; c0131357 __invalidate_buffers+87/b4
Trace; c0188e48 do_md_run+1f4/26c
Trace; c01168ae printk+172/180
Trace; c0189214 autorun_array+70/98
Trace; c023356f devfsd_buf_size+11b9b/13bd8
Trace; c0189440 autorun_devices+204/22c
Trace; c01895d0 autostart_array+168/1a8
Trace; c018a44c md_ioctl+384/8c4
Trace; c0137de0 blkdev_ioctl+0/30
Trace; c0142aca dput+2a/168
Trace; c01567c8 devfs_open+ac/174
Trace; c012f55a dentry_open+9e/114
Trace; c0137e03 blkdev_ioctl+23/30
Trace; c013f66f sys_ioctl+183/1b4
Trace; c010b334 system_call+34/40
Code; ca9353c8 [raid5]__put_free_stripe+40/88
_EIP:
Code; ca9353c8 [raid5]__put_free_stripe+40/88 =
0: 0f 0b ud2a =
Code; ca9353ca [raid5]__put_free_stripe+42/88
2: 83 c4 0c addl $0xc,%esp
Code; ca9353cd [raid5]__put_free_stripe+45/88
5: 8b 83 b4 02 00 00 movl 0x2b4(%ebx),%eax
Code; ca9353d3 [raid5]__put_free_stripe+4b/88
b: a8 01 testb $0x1,%al
Code; ca9353d5 [raid5]__put_free_stripe+4d/88
d: 75 19 jne28 _EIP+0x28 ca9353f0
[raid5]__put_free_stripe+68/88
Code; ca9353d7 [raid5]__put_free_stripe+4f/88
f: 68 d7 00 00 00pushl $0xd7
#
# Automatically generated make config: don't edit
Re: how should I set up swap?
On Fri, Jun 02, 2000 at 08:19:40AM +1000, Neil Brown wrote: Note: In 2.2 all filesystems go through the buffer cache, so they don't have a problem. Only swap doesn't and it doesn't whether you are swapping to a device or to a file. In 2.3, the buffer cache is largely sidelined in favour of the page cache, and this is, in part, why the rebuild code is totally different in 2.3.99pre8+ now does anyone who looked at the 2.3/2.4 code know what exactly is the status for the latest kernels?? Regards Luca -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: drive succesion
if you use the 0.90 raid code with permanent superblocks and autostart the raid code will deal with this gracefully. L. On Thu, Jun 01, 2000 at 09:25:43AM +0200, BORBATH Csaba wrote: Hi! I'am new to RAID.. If you have one SCSI adapter(sd) and four drives(sda1,sdb1,sdc1,sdd1), in a succesion made by SCSI IDs, and suppose after one reboot, the 3rd fail, then the 4th will be the the 3rd? How wil affect the software raid personality 5 this double change? Did miss some documentatin? Cs.% -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: boot redundancy - writing lilo to each drive?
get the latest lilo from metalab, it understands raid L. On Tue, May 30, 2000 at 03:22:29AM -0700, Gavin Clark wrote: Hi, I have a raid level 1 on 2 scsi drives and an IDE drive. I'm trying to set MBR on all my drives so if the first one goes down the system will still boot. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: RAID and new kernels, FYI
On Sat, May 06, 2000 at 03:19:49PM -0400, Edward Schernau wrote: The RAID patches will NOT patch cleanly (nor will much else) on a Redhat-supplied kernel. Make sure you start with a that maybe is beacuse the raid patch are already applied L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Raid error messages
On Thu, Apr 27, 2000 at 05:30:15PM +0200, Hans-G=FCnter?= Lütke Uphues wrote: Hi! md: serializing resync, md0 has overlapping physical with md2! md: serializing resync, md1 has overlapping physical with md2! dont worry it says that md0 md1 and md2 are on the same physical disk(s) so it will resync them one at a time for performance reasons. L. P.S. how about changing the message, this is starting to sound much as the 'lp on fire' issue :) L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Autodetection with BSD disklabel
AFAIR there were a lot of places where add_partition is called with
0 as the last argument, should they all be changed ?
L.
On Mon, Apr 17, 2000 at 12:29:39PM +0200, Norbert Eicker wrote:
So the following patch helped to make autodetection run on my system:
--
--- linux.orig/drivers/block/genhd.cThu Apr 13 13:28:25 2000
+++ linux/drivers/block/genhd.c Mon Apr 17 12:12:19 2000
@@ -729,7 +729,7 @@
if (partition-p_size)
add_partition(hd, current_minor,
first_sector+partition-p_offset,
- partition-p_size, 0);
+ partition-p_size, partition-p_fstype);
current_minor++;
}
printk("\n");
--
--
Luca Berra -- [EMAIL PROTECTED]
Communication Media Services S.r.l.
Re: Copying partition information
On Mon, Apr 03, 2000 at 07:34:32AM +0200, Jakob Østergaard wrote: AFAIK there's no tool that can read one partition table and then re-create it on another device in a clever way. You're right that sfdisk (it is part of util-linux) L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Swapping onto RAID: Debian Style?
pardon me? On Sun, Apr 02, 2000 at 02:03:36AM +0200, Thomas Rottler wrote: Hi! I have an autodetecting RAID1 swap set. The kernel does so all the work for me... Thomas -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: ext2resize
On Wed, Mar 29, 2000 at 03:40:42PM -0800, Peter Andersen wrote: I was looking at LVM and wondering what advantage it would give me. I like the idea of the volume groups and logical volumes but what does LVM give me other than the ability to resize/change volumes? LVM snapshot is a nice feature, you create a snapshot at one point in time then lvm will only write changes to the snapshot device, this enables you to make consistent backups. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: offtopic - lost ext2fs
On Thu, Mar 30, 2000 at 11:59:24AM +0300, Mustafa Bodur wrote: The start of partition is not known exactly. I tried some magic and scrolls like debugfs and e2fsck with alternative superblocks, but i can not get the partition back. i can see the disk content if i access raw, even i got a backup copy of disk with dd. Before i go bloody, do you have any advice to me or any experience to share with me ? thanx in advance, try with something like fixdisktable by Gordon Chaffee, it should restore your partition, Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Raid-Related System Locks
On Tue, Mar 28, 2000 at 03:48:44PM +0200, Bernd Burgstaller wrote: (scsi0:0:0:0) Synchronous at 80.0 Mbyte/sec, offset 31. (scsi0:0:2:0) Synchronous at 20.0 Mbyte/sec, offset 15. (scsi0:0:3:0) Synchronous at 10.0 Mbyte/sec, offset 32. (scsi0:0:13:0) Synchronous at 80.0 Mbyte/sec, offset 31. (scsi0:0:14:0) Synchronous at 80.0 Mbyte/sec, offset 31. argh, put tape and cdrom on a different scsi bus besides that disks should be higer priority devices than othrt stuff L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Swapping onto RAID: Good idea?
On Sat, Mar 25, 2000 at 11:31:12AM +0100, Martin Bene wrote: At 09:52 25.03.00, Mike Bilow wrote: 3. Create a swap file on /dev/md0 (which has been combined from /dev/hda1 and /dev/hdc1). Never tried this, sorry. has the same problem as swapping onto md0 :((( L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: can a md device be partitioned?
On Sat, Mar 11, 2000 at 07:17:31PM -0500, Paramasivam Kartik wrote: Hi, Could anyone tell me if an md device can be partitioned. no, you cannot partition an md device you can: 1) create multiple partitions on your disks then create multiple md devices 2) Use LVM (logical volume manager) over an md device Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: to swap or not to swap . . .(hardware RAID)?
On Sun, Mar 12, 2000 at 02:46:36PM -, Matthew Clark wrote: Is swapping on Hardware RAID safer than software RAID? Matthew Clark. yes, hardware raid should appear as a normal disk to linux L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: to swap or not to swap . . .
On Sat, Mar 11, 2000 at 04:02:34PM +0100, Frank Joerdens wrote: a) decreased performance if physical memory is exhausted (if you don't have enough physical memory) b) decreased system stability if physical memory is exhausted c) I don't really know what happens if physical memory is exhausted (I've never tried it, or found myself in the position, or tried setting up a test bed to provoke such a situation) process will get killed randomly -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: big big problem
On Fri, Mar 03, 2000 at 09:06:30PM +0100, Mathieu ARNOLD wrote: Hi a box i have crashed this morning, the problem is that : how could i tell the kernel to start the raid array and do what it can to recover the datas ? as far as i can tell, the problem comes from : 4hdh1's event counter: 0049 4hdg1's event counter: 0048 4hdf1's event counter: 004a 4hde1's event counter: 004a 4hdb1's event counter: 004a you need to modify your /etc/raidtab, changing ro failed-disk one of hdg1 or hdh1 (hdg1 seems to be the most out-of date, so i'd chose that one). then mkraid (pray, cross fingers, whatever) undo modification of raidtab raidhotadd the disk you marked as failed if this does not work you may need to restore from tape. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: mkraid aborting without any useful information
On Mon, Feb 07, 2000 at 03:13:32PM +, Carl L. Roy wrote: I have just installed Mandrake 6.1 with mkraid version 0.90.0. The Linux kernel that I am running is version 2.2.13-7mdk. When I attempt download the kernel update from the mandrake ftp site L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Argh, more problems with SCSI and RAID
On Wed, Feb 02, 2000 at 12:57:45AM +0900, Rainer Mager wrote: Feb 1 22:48:13 dual kernel: (scsi1:0:0:0) Performing Domain validation. Feb 1 22:48:13 dual kernel: (scsi1:0:0:0) Successfully completed Domain validation. personally i removed the printk from the kernel source, i was getting this error from my dds changer and it was filling my logs during nightly backups. i did not notice any ill effect for this. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: raid 1 and kernel upgrade 2.2.12-2.2.14
On Thu, Jan 27, 2000 at 12:54:36AM +0100, Email wrote: hi im running 2.2.14 with raidpatch raid0145-1999-0824-2.2.11 and raidtools 0.90 it works fine for me this is quite interesting, since it shouldn't work : L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: RAID Swapping
On Thu, Jan 20, 2000 at 06:05:19PM -0800, David Neilson wrote: I am using RH 6.0 (2.2.5-15) with raidtools 0.90-3. On December 26, I noticed a letter from Andy Poling to Jakob Ostergaard regarding swapping with RAID. The letter indicated RAID1 is unsafe for swapping (during resync). Can anyone shed some light on this for me? Are there any workarounds or actions I can take? Thanks in advance for your comments. you can check the list archives for the full thread (it is long) only known workaround is to delay swapon untli the raid partition used for swap has finished resync. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: ReiserFS and RAID1
On Mon, Jan 17, 2000 at 12:57:43PM +0100, KS wrote: Hi there, We are currently running RAID1 on some of our servers. We have thre ext2 filesystem. However we are considering seting up some test server to check how much using ReiserFS would improve the speed. On their pages they say that their filesystem works with RAID1 , but not with RAID5. Another thing thay say that the mirroring sync tools have the bug and that one have to sync the mirror with the FS unmounted. Since we have / on raid1 I don't think we could try ReiserFS. 1) don't use reiserfs for / 2) use an initrd that wait for resync to be complete (slowww) 3) i believe that resync may be safe if the fs is mounted readonly. but i wouldn't swear on it. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Latest ver of raid patches?
On Thu, Jan 06, 2000 at 10:24:40AM +0200, Chris Picton wrote: Hi Is there a raid patch which applies cleanly to a 2.2.14 kernel? I looked at kernel.org, but only see one for 2.2.11, which doesnt patch cleanly you should read the list :)) http://www.redhat.com/~mingo/raid-2.2.14-B1 L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Swap on Raid -- revisited
On Tue, Jan 04, 2000 at 05:35:19PM -0800, Michael wrote:
Could the Raid experts revisit a portion of the discussion about swap
on raid. I understand that the use/non-use of buffer space during
reconsturction vs swap creates a problem for swap on raid, however in
my pea-sized brain it appears that this would be a problem only if a
full partition were committed to swap use. What happens if the swap
is done via a file -- does the problem still exist or is the swap
process forced to use file IO to access the allocated space???
Will this circumvent the buffer/no-buffer problem?
According to Stephen Tweedie the problem happens in both case,
writes to the swap file do bypass the buffer cache in ANY case.
only way to be safe is:
do not use spare drives
swapoff before raidhotadding
replace swapon with something like
while grep -q "^${1#/dev/} .* resync=" /proc/mdstat;do sleep 1;done swapon $1
(needs to be smarter than that, to support swapon -a and swap on file)
regards,
Luca
--
Luca Berra -- [EMAIL PROTECTED]
Communication Media Services S.r.l.
Re: Don't try this at home!
On Sun, Jan 02, 2000 at 12:55:31PM -, Jim Ford wrote: I can't get past a kernel panic on boot. Maybe I should have waited longer before I cut the power to the complete system, to allow Raid to sort itself out. I suppose I might be able to rescue the system with a rescue setup running off a ram disk, but it's simpler to do the lot from scratch again. This experience has made me think - it's no great hardship for me, I'm just a hobbyist fiddling around with configurations that interest me. But I'd think again before jumping into software raid for a serious setup. ok, your only chance is to recreate the disk array. make a conf file specifying as failed-disk the drive you cut the power to, then mkraid, data should be there. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: SV: SV: SV: never kernel than 2.2.11
On Sun, Dec 26, 1999 at 01:38:11PM +0100, Johan Ekenberg wrote: sorry, but swap-on-raid is not stable. raid uses the buffer cache, swap does not result: when your swap partition is resyncing, due to disk failure or unclean shutdown swap gets corruppted. Is this a fact even when using a swapfile? That's what I do. No swap yes, sorry. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Swapping in the HOWTO
On Sun, Dec 26, 1999 at 08:32:13PM +0100, Johan Ekenberg wrote: According to Mr. Luca Berra -- [EMAIL PROTECTED], swapping on RAID is not it is not just according to me, it is Stephen Tweedie ([EMAIL PROTECTED]) who spotted the bug. search the recent list archives for the previous discussion, it was very similar to ours. L. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: Setting up Raid 1
On Wed, Dec 22, 1999 at 12:11:30AM -0500, Wilson G. Hein wrote: Do I need to apply the patch to the rpm'd RH source kernel 2.2.12-20? How NO -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: RAID 0 and RH 6.1
On Mon, Dec 13, 1999 at 08:18:26PM -0700, Chris M wrote: Then I needed to rebuild the kernel to include some other options and this is where things go awry. Now when I boot I get the following around the time it is trying to clean/mount /dev/md0: i s'pose you used the redhat-provided kernel-source rpm /dev/md0 must be a nonpersistenc RAID0 or LINEAR array! snip x x [*] Boot support (linear, striped) looks like that kernel option does not work with new raids those with a superblock. i believe redhat installed with an initrd try disabling that option and create an initrd should work. -- Luca Berra -- [EMAIL PROTECTED] Communication Media Services S.r.l.
Re: HELP-- identical disks but fdisk sees them different
On Thu, Dec 09, 1999 at 11:09:18PM +, Lyndon David wrote: I am just setting up a system and have put in two identical disks as hda and hdc. fdisk sees hda as having 255 heads and 63 sectors but hdc as having 16 heads and 63 sectors ! in addition to what has been said try zeroing the partition table and rebooting L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Linux Raid5 and NFS, how stable an reliable?
On Mon, Dec 06, 1999 at 12:56:29PM -0500, [EMAIL PROTECTED] wrote: I've had very little trouble with the old user-space NFS code, but recently I setup a Red Hat 6.1 system using knfsd and have been having some trouble in a setup where the 6.1 box is the server and several older (2.0.x kernel) Linux systems are using autofs to mount it. you should probably check hjlu's nfs patches to make it work also there are som patches that support NFSv3 that is a great boon. stable, nfs ?!? arg L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: scsi channels (Was: ide hardware raid)
On Wed, Dec 01, 1999 at 04:20:49PM -0800, Alvin Oga wrote: they have scsi-2 raid cards from adaptec... 1-channel($430) and 3-channel($650) raid controllers.. a dumb question... what are the channels used for ??? scsi channels are different scsi busses, you use them for load balancing. btw, i doubt any of these card will work with linux :( L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: errors on boot
On Tue, Nov 16, 1999 at 03:14:42PM -0500, Michel Pelletier wrote: It looks to me like the [2/1] means only one of my two 'low level' partitions below /dev/md2 is working. I don't know what the [U_] means, it means the second partiotion is gone U=ok _=fubar -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
networked raid
hello, i found this on lwn, mebbe someone in this list is interested From: Philipp Reisner [EMAIL PROTECTED] To: Linux-HA mailing list [EMAIL PROTECTED] Subject: drbd (was nmbd) Date: Sat, 6 Nov 1999 12:13:42 +0100 You can find it at http://www.complang.tuwien.ac.at/reisner/drbd/ DRBD == by Philipp Reisner [EMAIL PROTECTED] What is it ? Drbd is a block device which is designed to build high availability clusters. This is done by mirroring a whole block device via ( a dedicated ) network. You could see it as a network raid 1. What is the scope of drbd, what else do I need to build a HA cluster ? -- Drbd takes over the data, writes it to the local disk and sends it to the other host. On the other host, it takes it to the disk there. The other components needed are a cluster membership service, which is supposed to be _heartbeat_, and some kind of application that works on top of a block device. Examples: *) A filesystem fsck. *) A journaling FS. *) A database with recovery capabilities. How does it work ? -- Each device (drbd provides more than one of these devices) has a state, which can be 'primary' or 'secondary'. On the node with the primary device the application is supposed to run and to access the device (/dev/nbX). Every write is sent to the local 'lower level block device' and to the node with the device in 'secondary' state. The secondary device simply writes the data to its lower level block device. Reads are always carried out locally. If the primary node fails, heartbeat is switching the secondary device into primary state and starts the application there. (If you are using it with a non-journaling FS this involves running fsck) If the failed node comes up again, it is a new secondary node and has to synchronise its content to the primary. This, of course, will happen whithout interruption of service in the background. Howto ! --- Let's assume that your two machines are named 'node1' and 'node2', and that you want to use /dev/loop0 as the lower level block device on both, then these commands are needed on node1 to setup drbd: insmod drbd.o drbdsetup /dev/nb0 /dev/loop0 node1 node2 On node2 you need: insmod drbd.o drbdsetup /dev/nb0 /dev/loop0 node2 node1 And finally your cluster membership service needs to tell one of the two to become primary by running: drbdsetup /dev/nb0 p Implementation details -- At first I wanted to use UDP so that I could benefit from UDP's multicasting abilities to implement clusters of more than two nodes. But after I finished the first experimental UDP implementation it turned out that the kernel is only storing up to 64 kb of incoming data on an UDP socket. And with faster networks it happens quite easily that your intel box (timer interrupt @ 100 times per second) is not scheduling your receiving process often enough and you get an enormous amount of lost packets. (Alpha did a *lot* better, there you have 1024 interrupts per second). So for now I am using TCP. Accessing the lower level block device is done with a temporary copy of the buffer_head and a call to ll_rw_block. Thus you should never access the lower level block device directly when you have a drbd running on top of it! I am using /dev/nb0 (and major 43) because of ~linux-2.2.7/drivers/block/ll_rw_blk.c line ~447 : --snip-- /* Loop uses two requests, 1 for loop and 1 for the real device. * Cut max_req in half to avoid running out and deadlocking. */ if ((major == LOOP_MAJOR) || (major == NBD_MAJOR)) max_req = 1; --snap-- It _will_ _deadlock_ if you use another major number than one of these two!! Status of drbd-0.1.tar.gz - This is a proof-it-is-no-vapo-ware-release and not more! It is the first time it is somehow working without immediately crashing my machine (you can "even" unload the module). It is not able to handle blocksized other than 1024 and there is not a single line of code of the sync-a-new-secondary stuff yet. The used port is hardcoded into drbdsetup (7788) for now. I have tested it on an intel box and an alpha machine. Want to try something new?Are you a Linux hacker? Volunteer in testing mergemem! (Get it from http://das.ist.org/mergemem) - Philipp ReisnerPGP: http://der.ist.org/~kde/pgp.asc -- Linux HA Web Site: http://linux-ha.org/ Linux HA HOWTO: http://metalab.unc.edu/pub/Linux/ALPHA/linux-ha/High-Availability-HOWTO.html -- -- Luca Berra -- [EMAIL
Re: raidstop --all doesn't do its job
On Mon, Nov 08, 1999 at 11:33:46AM +, Marc Haber wrote: However, I find that a raidstop --all is necessary. Otherwise, the system shutdown scripts would have to know about all RAID devices to stop them all if autodetection isn't being used. no, the kernel deals with stopping all active raid devices in the shutdown/reboot/poweroff syscalls. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Problems moving from debian to RH with RAID
On Tue, Nov 02, 1999 at 11:49:04PM +, Simon Haddon wrote: Hi, I have a difficult RAID problem. no you are just a bit confused by different raid versions 1) i don't believe you running kernel 2.1.15 2) from your mail i suppose you don't need to save data from your old raid set first you should read the excellent howto at http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/ then my quick recipe use either redhat's kernel or kernel 2.2.13ac2 use the raidtools package from the redhat cd do mkraid with your proposed raidtab (it looks ok) change the partition type to 0xfd for the raid disks if you have the raid1 code compiled as a module (eg. redhat default kernel) do mkinitrd --with raid1 /boot/initrd-2.2.5-15.img 2.2.5-15 and add the line initrd=/boot/initrd-2.2.5-15.img to your lilo.conf Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Patch for kernel 2.2.13?
On Wed, Nov 03, 1999 at 04:58:34PM -, [EMAIL PROTECTED] wrote: Does anyone know when the raid 0.90 patch for kernel 2.2.13 should be released? I've looked at kernel.org but latest there is 2.2.11. look in /pub/linux/kernel/alan/2.2.13ac -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: raid on 2.2.9
On Wed, Nov 03, 1999 at 02:54:05PM -0500, Paramasivam Kartik wrote: Do we need to patch Linux 2.2.9 before we can use the raidtools (like mkraid) to install raid. yes -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
RAID and MANDRAKE
it seems there is an update for mandrake 6.1 to support the new raid code. i dinna check personally, i will from the mandrake updates page: October 29, 1999 kernel This is 2.2.13 final. RAID is now supported on this kernel. 27db8f6709b4c8744bae082a749cf79e kernel-2.2.13-22mdk.i586.rpm 5edb7200d67a7deaa58cf8d66eba6ca8 kernel-2.2.13-22mdk.src.rpm 942782496a12aa5b8d1615ca841d0d5a kernel-doc-2.2.13-22mdk.i586.rpm cdb8e6d494221d0bfde5150d29a7969f kernel-fb-2.2.13-22mdk.i586.rpm 67fe8a11e64ab3721292ac582ca83dce kernel-headers-2.2.13-22mdk.i586.rpm dd1770635a622e5010e1959f461ab4c9 kernel-ibcs-2.2.13-22mdk.i586.rpm 2a2f3001e798473f10537033b4ea2bb4 kernel-pcmcia-cs-2.2.13-22mdk.i586.rpm 35b0450d35d42ba77b5066f1d90f9825 kernel-smp-2.2.13-22mdk.i586.rpm ecdc2001887bccb5ba1543f9f4f41bad kernel-source-2.2.13-22mdk.i586.rpm -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: PROs and CONs of booting RAID?
On Tue, Nov 02, 1999 at 08:09:29AM -0800, Roeland M.J. Meyer wrote: Actually, HP-UX, on HP9000 V-class, has a journaling file system. I've got that only guarantees filesystem metadata integrity, not application data integrity. EMC coming in this afternoon, I'll ask them about the battery backup on the write-back cache. My memory may be fuzzy, but I thought it was an option, on well, my 3430 enclosure has a battery backup unit. you have to wait about 5 mins before it shutdowns completely after you switch it off. Symetrix. I build 99.99+% sites ... that's what I do. well, my job deals with that 0.01-% : Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Help for raid
I have a Sparc 10 with Linux6.0 running. I have four disks da 1.7Gb. Linux 6.0 did i miss something, last time i checked it was around 2.3.20 and that wasn't 10 years ago. My kernel is 2.2.5-15 and raidtools-0.90 If I do mkraid /dev/md0 I receive - handling MD device /dev/md0 analyzing super-block disk 0: /dev/sdb1, 1720345KB raid superblock at 1720256KB mkraid: aborted - Can you help me ? no, unless you post the content of your raidtab and any eventual messages in your syslog. L. P.S. the first that suggests that he patches his kernel will win a "mongolino d'oro" (which is a special italian prize for very smart people) -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: raid5 under linux
On Fri, Oct 29, 1999 at 03:45:06PM +0200, michel wrote:
My version of linux is : Linux Version 2.2.9-27mdk
My version of mkraid is mkraid version 0.90.0
this is becoming a FAQ
see my answer to a previous poster
--
Luca Berra -- [EMAIL PROTECTED]
Communications Media Services S.r.l.
On Wed, Oct 27, 1999 at 03:30:31PM -0500, Bill Carlson wrote:
I'm thinking Florin means the kernel is not compiled with support by
default.
no.
This package includes the tools you need to set up and maintain a software
RAID
device under Linux. It only works with Linux 2.2 kernels and later, or 2.0
kernel specifically patched with newer raid support.
To me that implies a 2.2.x kernel does not need a patch. On Mandrake 6.1,
the required RAID modules were already in place after installation.
this is FALSE
kernel 2.2.x does require a patch.
get either a redhat kernel or a linux-2.2.13ac1
mandrake was unfortunate in that they included the raidtools package
but did not patch the kernel to support raid.
{to see if you have a patched kernel look for the "autodetecting raid"
message during bootup}
--
Luca Berra -- [EMAIL PROTECTED]
Communications Media Services S.r.l.
Re: Please Help - SCSI RAID device very slow
On Thu, Oct 28, 1999 at 12:15:19AM +0100, Alan Cox wrote: Does anyone know how I stop the MegaRAID using the onboard INTEL SCSI Controller? (i960) It should be using my SYMBIOS card...??? An i960 isnt a scsi controller. If you want to use the symbios directly load the ncr53c8xx driver and not the symbios driver. I'not totally sure if that will work but it is at least the right driver. That will basically cut out use of the megaraid totally no way! the megaraid controller hides completely the symbios onboard chip so linux does not see it at all. I know of one case where it is possible to bypass the megaraid for at lease one channel, it is the embedded netraid(remarketed megaraid) on hp netservers LH3, you have to change a setting in the bios. All other cards i tested (netraid 1, netraid3, netraid3si) don't allow this option. Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: raid not auto detecting
On Wed, Oct 27, 1999 at 01:38:05PM +0200, Gordon Booth wrote: stock rh 6.0 already includes raid patches fyi it uses kernel 2.2.5 the correct questions are: have you changed the partition type to 0xfd? have you created an initrd with mkinitrd --with raid1 ... Which kernel version are you using?? Have you patched it with the latest RAID patches? I'm running a stock rh 6.0 install. Where can i look to autodetect and start my raid arrays? -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: mkraid aborts, no info?
On Wed, Oct 27, 1999 at 03:30:31PM -0500, Bill Carlson wrote:
I'm thinking Florin means the kernel is not compiled with support by
default.
no.
This package includes the tools you need to set up and maintain a software
RAID
device under Linux. It only works with Linux 2.2 kernels and later, or 2.0
kernel specifically patched with newer raid support.
To me that implies a 2.2.x kernel does not need a patch. On Mandrake 6.1,
the required RAID modules were already in place after installation.
this is FALSE
kernel 2.2.x does require a patch.
get either a redhat kernel or a linux-2.2.13ac1
mandrake was unfortunate in that they included the raidtools package
but did not patch the kernel to support raid.
{to see if you have a patched kernel look for the "autodetecting raid"
message during bootup}
--
Luca Berra -- [EMAIL PROTECTED]
Communications Media Services S.r.l.
Re: raid problem
On Wed, Oct 27, 1999 at 11:09:46PM +0200, Mathieu ARNOLD wrote: I'm using a redhat 6.1 with a 2.3.21 kernel. well, i've gone back to the 2.2 series, i'm now using a 2.2.13 kernel, and well, it does exactly the same thing. does someone have a clue ? yup. neither 2.3.x nor 2.2.x do support new raid code, but you can find a patch for 2.2.x (while there is no patch for 2.3.x) i'd get ftp://ftp.fr.kernel.org/pub/linux/kernel/people/alan/2.2.13ac/patch-2.2.13ac1.gz other option is: ftp://ftp.fr.kernel.org/pub/linux/daemons/raid/alpha/raid0145-19990824-2.2.11.gz and fix the rejects by hand regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Recovering from a lost disk
On Mon, Oct 25, 1999 at 05:59:37PM +, Marc Haber wrote: Is commandname --help really the only documentation besides the source code? Nope --help is misleading, use the source :))) One more thing: In a spare minute, I straced scsidev and found out that it has a buglet: It obviously doesn't check the return code of Was it rewitten recently? last time i checked it had mor buffer overruns tahn lines of code :((( L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: To whoever maintains this list
On Fri, Oct 22, 1999 at 08:28:39AM -0500, Kenneth Cornetet wrote: I'm an idiot. After digging for ways to make outlook look at obscure mail headers like "X-loop:", I noticed what should have been obvious: Mail to the linux-raid list comes addressed to [EMAIL PROTECTED] I hate it when I make a simple problem complicated! really? :) please use the sender header to sort majordomo mailing lists Sender: [EMAIL PROTECTED] -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: raid howto
On Fri, Oct 22, 1999 at 08:12:30AM +0200, Jakob Østergaard wrote: Perhaps we could add a ``read the howto at http://ostenfeld.dk/~jakob/Software-RAID.HOWTO'' to the bottom of all mails on this list ? AFAIR the linux-smp list does that (with the SMP FAQ). Could the list maintainer give an oppinion on this ? i asked dor the same thing some months ago: !answer L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Problems Creating a Raid 1 Device
On Fri, Oct 15, 1999 at 02:41:08PM -0700, Albert E. Whale wrote: I am running 2.2.12. I have raidtools-0.90-6mdk installed. sigh, it seems that mandrake ships with a raidtool package but not with a raid enabled kernel :( (hint what about adding a distro section to the faq?) get the patches from a kernel.org mirror L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Re: raid problem
On Fri, Oct 15, 1999 at 10:33:58AM +0200, Mathieu Arnold wrote: where can i find the latest patchs ? you cannot. this is the problem raid has not been ported to 2.3 kernels, sorry Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: your mail
On Sun, Oct 03, 1999 at 05:34:16PM -0500, Bo Kersey wrote: If I'm not mistaken, the kernel-2.2.12 available as an rpm from the rawhide section of the redhat ftp site has the RAID patches built in But RAID is a module in this kernel, so you'll have to recompile if you want to do root raid... no you don't just create an initrd with mkinitrd --with raid1 (or whatever) L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Adaptec Raidport
On Mon, Sep 27, 1999 at 02:13:18PM -0400, [EMAIL PROTECTED] wrote: Will the Adaptec Raidport 3 ARO-11030U2 work with Linux? NO -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: IBM Netfinity Hardware Raid?
On Sun, Sep 26, 1999 at 08:38:41PM +1000, John Leach wrote: Dear raiders, Does anyone know if the hardware raid on the IBM Netfinity is supported under Linux? It doesn't appear (or seem to appear) on the Red Hat Certified machines list (various Netfinities are supported but raid is not mentioned with them) TIA John Leach try looking in the last 2.2.13preX or 2.3.18acX patches. /*/ /* ips.c -- driver for the IBM ServeRAID controller */ /* */ /* Written By: Keith Mitchell, IBM Corporation */ /* */ /* Copyright (C) 1999 IBM Corporation*/ L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: RPM'd RAID kernel and raidtools?
On Thu, Sep 16, 1999 at 05:38:21PM -0500, Chris Garrigues wrote: It would be nice if the HOWTO or the non-existent FAQ were to explain how to tell if you have a working RAID kernel. In particular, if you've never seen the output of /proc/mdstat, it's kinda hard to know what it should look like. try `grep autorun_array /boot/System.map' L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: mkraid aborted please help
On Fri, Aug 27, 1999 at 05:52:22PM -0400, esl wrote: I updated my kernel from the stock 2.2.5-15 (RedHat 6.0) to 2.2.11. I did not use modules for raid etc. sotck and redhat are a contaddiction in terms the redhat kernel is heavily patched it also happens to include raid pathches that are not in the stock kernel get the patches from ftp.xx.kernel.org and rebuild. L. -- Luca Berra -- [EMAIL PROTECTED]
Re: Autostart failing.
On Sat, Aug 21, 1999 at 10:24:49AM -0700, C. R. Oldham wrote: I'd like to know the answer to this, too. I changed /etc/raidtab such that the argument to persistent-superblock was '1' and ran 'mkraid --force /dev/md0' (after a backup of course). It worked, and my data was unharmed. The RAID now starts automatically. I did not need to re-run mke2fs to create a new filesystem, and fsck is not complaining about the partition size. If I ever fill the partition is it possible that I will overwrite the persistent superblocks? yes, you are in danger, i suggest you shrink your filesystem. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
To mailing-list administrator
hello, what about adding the line http://ostenfeld.dk/~jakob/Software-RAID.HOWTO/ at the end of every mail that comes from the list? Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Problem seting up RAID-1
On Tue, Aug 17, 1999 at 03:53:10PM -0500, Gerardo Muñoz Martin wrote: cannot determine md version: No such file or directory 1. switch off HTML mode in your mailer :) 2. check that the device nodes /dev/md0 ... exist if not create them: mknod /dev/md0 b 9 0 mknod /dev/md1 b 9 1 . L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Re: Re: Problem with md - md.h
On Sat, Jul 31, 1999 at 08:45:38PM +0200, Sonia de Diego Atance wrote: Hi again Alvin, and thanks for being so patient with me!!!. Hi everybody in the list!!! did you try with the sample raidtab i sent you last week? I have installed kernel 2.2.10 with its patch (with gave me some problems trying to install it, but I think at the end I could do it). I have been also which patch are you speaking of? raid0145-19990724.gz what problems, there should be none? able to install raidtools-0.41 but I can´t concatenate disks in linear mode. The error is the same always. i dont believe the new raid code works with old raidtools do you see anything in syslog? what about using strace mdadd ... if syslog does not work -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Re: Problem with md
On Thu, Jul 29, 1999 at 03:33:14PM +0200, Sonia de Diego Atance wrote: Hi again. I want to mount a linear multiple device with md-0.35 (I do not know where to find 0.36). I have RedHat 6.0 and kernel 2.2.5. I have found in RedHat a package called raidtools with programs as raidadd. Do you know if this is the equivalent to md?. Because this is the explication I find. It worked in Slackware but not in RedHat. redhat packaged their distribution with the new raid tools and kernel patches, you should create an /etc/raidtab file (read the documentation) and use the raidxxx commands. Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Fighting RAID0...
On Wed, Jul 28, 1999 at 06:32:04AM -0700, Brian Macy wrote: If someone can tell me what I need to do to use my RAID0 partitions with the latest kernels I'd really appreciate it (what raidtools, what commands, can I use my 0.50 RAID0 partitions with new tools, kernel patches???). Thank you. I do not use raid 0, i consider it suicidal/i have no use for it except as in part of raid0+1 (don't flame me, i won'tchange idea) but if i were in you i'd take a look at the raid0run tool from the last raidtool-0.90 regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Unofficial patch for kernel-2.2.10
thanks, a lot. I could not have said it better. I agree on every point. Luca On Tue, Jul 13, 1999 at 12:03:34PM +0300, Osma Ahvenlampi wrote: Theron J.Lewis [EMAIL PROTECTED] writes: You can find my hacked patch raid0145-19990421-2.2.10.gz with an explaination on how I generated it at the following URL: This patch is BADLY broken for everything but RAID-0. Resyncing a RAID-1/4/5 array will not work due to a priority bug. The "official" RAID code maintainers are working on a good patch for 2.2.10, but recent 2.2 kernels have filesystem corruption problems unrelated to RAID, so they're not safe to use in any case. DO NOT USE THIS PATCH ON 2.2.10!! In fact, do not use ANY RAID patch on a kernel above 2.2.6 until an official patch exists and the filesystem problems in the vanilla 2.2.x, x6 problems have been resolved. -- Osma Ahvenlampi -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: mkraid problems (v0.90/2.2.10/RedHat)
On Thu, Jul 08, 1999 at 01:00:42PM -0700, Zack Hobson wrote: Hello RAID hackers, RedHat 6.0 w/ 2.2.10 kernel (compiled with RAID-1 support) raidtools 0.90 compiled from distributed source (ie, non-RedHat) raidtools 0.90 does not work with stock kernels either you use old raidtools or you patch the kernel But... kernel patches do not work with kernels 2.2.7 (yes i know you can modify a couple of lines, but even if i do it on my own boxes, i will not suggest anybody to take the risk) besides that 2.2.10 has the infamous fs corruption bug. i'd suggest downgrading to the last 'stable' redhat kernel. Luca P.S. (for Felix) you can build an N-way mirror (n2) for added redundancy, also please quote only relevant parts of a message when replying. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: raid 0.90 on 2.0.37 ?
On Wed, Jul 07, 1999 at 02:11:46PM +0200, Giulio Botto wrote: On Wed, Jul 07, 1999 at 09:12:02AM +0200, paolo furieri thusly shaped the electrons: I am using raid-1 0.90 on 2.2.5-22 redhat kernel, and I experienced a lot of instability problem: services (http, syslog, named ...) stay up for some hours, then they suddenly go down. Then I am waiting for a new raid patch for new kernels. But I have seen in mailing-list that 2.2 kernel has filesystem stability problem. Hmm, I'm missing these announces, but I've got a RedHat 6.0 2.2.5-22 with a 3-way SCSI raid5 with an uptime of about a couple of months, and It's hosting http, ftp, nfs, samba and other services with no problem ... are you sure the problem is the kernel ? there is a reported filesystem corruption problem with late 2.2 kernels it does not happen in every case so this has been very hard to trace. but the problem exists and has been reported by different people in different conditions. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: olderstyle raid -- newstyle raid
On Thu, Jul 01, 1999 at 06:53:53PM -0500, Brian wrote: I have recently upgraded some boxes from RH5.2 to RH6.0 and have run into some problems with getting my arrays to work. The arrays were created under RH5.2 with raidtools-0.50beta10-2. I cannot set "persistant-superblock" to 1, because then it would destroy my data correct? So I leave it to 0. I try mkraid, to upgrade and it fails: [root@sol /tmp]# cat /proc/mdstat Personalities : [raid0] read_ahead not set unused devices: none [root@sol /tmp]# /sbin/mkraid --upgrade /dev/md0 handling MD device /dev/md0 analyzing super-block disk 0: /dev/hdc1, 11255296kB, raid superblock at 11255232kB cannot upgrade magic-less superblock on /dev/hdc1 ... this message means "YOU DO NOT NEED TO UPGRADE" mkraid --upgrade is used only if you have a suberblock, and you don't. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Success!
On Fri, Jun 11, 1999 at 12:43:49AM -0400, Robert (Drew) Norman wrote: The best way to do a copy and make sure it all works is like this: find . -print | cpio -pdmv target directory i use find . -print0 | cpio -0admpuV --sparse /target/dir note that cp -dpR and/or cp -a usually fail to copy hard-links correctly L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Raid on Kernel 2.2.9
On Fri, Jun 11, 1999 at 08:44:14AM -0400, Robert (Drew) Norman wrote: I have raid running with kernel 2.2.9-ac4. I had to go about it in the following manner: untar kernel. Apply ac4 patch. Apply raid0145-19990421-2.2.6 Apply the following patch for the failed patch of raid0145-1999 (I had to manually apply it, it wouldn't install itself for some reason) and now you probably think you are smarter than everyone else out there... wait till you discover your raid will not resync :) applying raid patches to different kernels may break things, you have to test your setup if you do, it is not just a matter of having patch return with status 0. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: RAID-1 Problem
On Fri, Jun 04, 1999 at 05:16:44PM -0400, Paul B. Brown wrote: Boot messages: Starting up RAID devices. Cannot determine md version: operation not supported by device. *** An error occurred during the RAID startup. *** Dropping you to a shell; The system will reboot *** when you leave the shell. remove the junk raid lines from redhat startup scripts /etc/rc.d/rc.sysinit /etc/rc.d/init.d/halt also remove the old raidtools rpm dmesg feedback: autodetected RAID arrays could not import sda5 could not import sdb5 autorun ... ... autorun DONE. this is strange, does it work if you start it manually with raidstart /dev/md5??? L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Raid 1 Setup Using RAIDTools 0.50 BETA 10
On Thu, Jun 03, 1999 at 02:19:22PM -0400, Paul B. Brown wrote: Hello all, Sorry for such a simple request but I can't seem to get mkraid to work. I want to mirror (RAID-1) two 4.2 GB SCSI drives (/dev/sda, /dev/sdb). Distribution: Redhat 5.2 (Linux 2.0.36) Software: RAIDTools 0.50 BETA 10 standard answer to this is: get a fresh kernel either from the kernel SRPMS or kernel.org then get the latest raid patch and raidtools from ftp://ftp.us.kernel.org/pub/linux/daemons/raid/alpha/ patch kernel install raidtools edit files /etc/rc.d/rc.sysinit and /etc/rc.d/init.d/halt and remove any reference to the (now obsolete) raid tools. The Multiple Devices Driver Support is enabled in the kernel with all sub-options enabled as well (linear, RAID-0, RAID-1, RAID-4, and RAID-5). The two SCSI disks are laid out identically: are you trying to upgrade from a single disk install to a raid, this will not be that easy. besides that root raid is even more painful since lilo cannot load the kernel from an md device, you nedd a non mirrored /boot partition, or you can try your luck with GRUB. This assumes that /dev/sda2 (swap) and /dev/sda3 (extended partition) do not need to be mirrored. extended should not be mirrored swap was unstable with old raid, it is stable now, so you can do it I tested with kernel 2.2, not 2.0, (Ingo does it also work with 2.0??) mkraid /etc/raidtab mkraid -c /etc/raidtab mkraid mkraid -f usually you do mkraid /dev/mdX Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: patches for 2.2.9?
On Sat, May 29, 1999 at 01:41:10AM +0100, Paul Jakma wrote: On Tue, 25 May 1999, Luca Berra wrote: On Sat, May 22, 1999 at 04:31:01PM -0400, Fred Reimer wrote: Second, where can I get patches for the alpha raid tools that RedHat chose to put in their version of the kernel that will work under 2.2.9? The DON'T use raid patches with kernel 2.2.8 or later, IT WILL HURT. 2.2.8 had a nasty block device problem which was solved in 2.9. What other issues are there with 2.2.9? (i've just gone and hand merged 2.2.9-ac1, raid0145 and devfs-v99.) dunno about 2.2.9-ac1 but for 2.2.8 and 2.2.9 raid will not resync. # uname -a;uptime;cat /proc/mdstat Linux Moskowskaya 2.2.9 #1 Fri May 21 21:43:10 CEST 1999 i586 unknown 4:46pm up 8 days, 19:00, 9 users, load average: 2.08, 2.38, 2.32 Personalities : [raid1] read_ahead 1024 sectors md1 : active raid1 [dev 07:01][1] [dev 07:00][0] 10112 blocks [2/2] [UU] resync=0% finish=551210.7min unused devices: none L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Swap file on raid
On Sat, May 22, 1999 at 12:32:04PM +0200, Francesco Potorti` wrote: Hi, I remember having seen on this list a message saying that making a swap file on a filesystem residing over a raid device was not possible/reliable (because of a race condition in the kernel?). swap on raid works reliably with the latest raid patches. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: installing root raid non-destructively
On Wed, May 12, 1999 at 02:55:15PM +0200, Louis-David Mitterrand wrote: First, kernel-2.2.7's documentation states that booting on raid1 is not supported yet, but looking at the alpha raidtools-0.90 package it seems possible. Does it mean that if I patch the kernel sources appropriately then booting on raid1 is supported (experimentally)? no, lilo does not know anithing about raid devices, so you have to make a separate (non raid) /boot filesystem to hold the kernel. someone said that GRUB may be able to boot from a raid1 device, looking at the docs it seems easonable, but i haven't tested it yet. Second, is there any way at all that I can implement raid1 without destroying all data on my present disks? From the docs this seems inevitable. yes, two ways actually get martin bene's patches that were posted on the list, build a raid with only one device (your new disk), copy everithing on the new raid (find ... -xdev | cpio -p ), add the old disk to teh array and lo! all done. (srcond method explained at next quesition) Third, (naive questions) if raid1 supports on-the-fly disk "reconstruction" why can't I simply add another identical disk alongside my present one, activate raid1 non-destructively and have disk2 be "reconstructed" as the mirror image of disk1? because linux-raid keeps a 4K raid superblock at the end of the partition so if you already created the filesystem you have no room for the superblock. i have not yet tested this, but you could get resize2fs (you need a partition magic license for that, shrink the filesystem by 4K, dd one disk over the other, then create a new raid. Finally, I read that the way to install root-raid is to use the to-be swap partition to install a minimal Linux system, then create the raid1 partition, then test and configure the kernel, then boot from raid and install the complete system. Is this correct? this is a third way, but it is long, boring and just a pain in the ass if you already installed. btw, fourth way involves using a tape drive. :( L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Swap on raid
On Mon, May 10, 1999 at 07:26:53PM +0200, Dietmar Stein wrote: Hi At work we got much HP-Workstations and -Servers; everyone got a swap-partition which is of same size as physical memory (or even bigger). hp-ux uses swap partitions as a dump device, something i'd love to see on linux systems sooner or later. anyway some swap space may be needed since there are process that allocate tons of virtual memory, and they don't use it. if the machine swaps occasionally i think it is acceptable if i have to use swap i wan't it on a raid device. L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Question about logical drive expansion on hardware raid.
On Thu, Apr 29, 1999 at 06:44:56PM -0400, XxEDGExX wrote: We have a raid box from Raid Inc. It has the capability of doing logical drive expansion. I'm somewhat confused as to how this works in our particular setup. I figured someone on here could enlighten me. As far as Linux is concerned, it see our raid array as one device. I've use this device and partitioned it. usually in hw-raid speak a logical unit is what the raid controller shows to the host computer as a single physical disk. (in your case /dev/sda) i beleave the logical drive expansion capability. means you can add disks to an existing lu without low level reformatting it, it has nothing to do with partitions L. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Raw devices, raid and kernel 2.2.x
On Mon, Apr 19, 1999 at 04:58:20PM +0200, Dietmar Stein wrote: I want to set up a box for testing with raw devices and raid0 under kernel 2.2.3 or 2.2.5 with raidtools 0.90. Can anybody point me to the URL for documentation of raw devices, patches and tools? ftpsearch reports: 1 -r--r--r-- 10.0K 1999 Mar 26 ftp.sunsite.auc.dk /disk1/ftp.uk.linux.org/pub/linux/sct/fs/raw-19990326.tar.gz 2 -rw-r--r-- 10.0K 1999 Mar 26 ftp.gwdg.de /pub/linux/misc/linux.org.uk/sct/fs/raw-19990326.tar.gz 3 -r--r--r-- 10.0K 1999 Mar 27 ftp.tux.org /pub/sites/ftp.uk.linux.org/sct/fs/raw-19990326.tar.gz 4 -r--r--r-- 10.0K 1999 Mar 26 ftp.icm.edu.pl /vol/rzm0/linux-uk/sct/fs/raw-19990326.tar.gz -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: Mirroring root and fsck
On Mon, Apr 19, 1999 at 10:22:40AM -0700, Aaron D. Turner wrote: All the md devices are set to automount via the parition ID. So my NO via the partition ID you set md device to autostart, (tis means the raid device is available to the system) mounting refers to filesystem, and mounting filesystems is handled via init scripts and /etc/fstab, usually decent init scripts do fsck before mounting filesystem. question is, how do I fsck all of the partitions? My system hung on the way down to init level 6 and when it came up it had some partition errors and now I have some minor filesystem issues. I know I have to run fsck against the md devices, but I'm not sure how this works with the RAID automounter. md devices are the same as normal disk device wrt filesystems so you just have to fsck /dev/mdX Regards, Luca -- Luca Berra -- [EMAIL PROTECTED]
Re: Easy Peasy!
On Mon, Apr 19, 1999 at 07:33:54PM +0100, Jim Ford wrote: RAID-0 wasn't so difficult after all - not withstanding all the ghost stories I've been reading in the docs! One thing though. When I mke2fs'd /dev/m0 I had some bad blocks reported. I then ran badblocks with the -w and -o options to read/write check it and get a file of bad blocks, with a similar result. I suspect the bad blocks indicate a failing HD (It's very old!) but I'm willing to give it a try. Are these bad blocks automatically locked out, or do I have to give a command to do so using the generated file (there is a e2fsck to do this)? Regards: Jim Ford 1)either use mke2fs -c to check the filesystem or mke2fs -l bad block file 2) e2fsck -c (i think it was introduced in v1.10) will check for bad blocks 3) consider using your hard disk as a paperweight. -- Luca Berra -- [EMAIL PROTECTED] Communications Media Services S.r.l.
Re: swap over soft-RAID = use a swapfile ?
On Wed, Feb 24, 1999 at 05:48:00PM +0100, Benno Senoner wrote: Hi, I heard that actually soft-raid for Linux can't swap over a /dev/md* device. I don't know how much slower it is to swap do a regular file instead using the raw partition, Benno, if i remember correctly the swap-on-raid problem was a memory starvation problem between the raid code needing memory for computations, so the system swaps out pages to disk, leading to the raid system needing memory, and so on. I don't believe that swapping to a file on an md device can do anithing to better this condition. I believe a solution might be reserving enough pages in kernel memory to handle raid, can this need be calculated Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: RAID1 experiences
Sorry if i am going way off topic here, if this is the case please insult me personally. On Mon, Feb 15, 1999 at 02:31:13PM +, Stephen C. Tweedie wrote: Umm, Ingo Molnar == [EMAIL PROTECTED] I think we can assume that there is somebody working for Red Hat who knows a bit about the current state of Raid. :) I believe everybody knows that RedHat is actively funding the Linux development by employing many skilled developers to do just that, and so many we probably all missed Ingo was among them. However, speaking from the point of view of a kernel developer rather than a Red Hat employee, there are real obstacles to including the new Raid stuff in Red Hat Linux, the main one being compatibility with existing installations using older Raid code. I wouldn't like to be the one trying to make Red Hat upgrades work with the new Raid drivers but without breaking old-style Raid volumes... I really hope backwards compatibility issues won't stop linux development if that was true we would probably be still stuck with libc 4 It is not impossible to upgrade from 0.50 raid superblocks to newest raid superblocks, a different problem might be for drives without any superblock. As a last chance you can always trash backward compatibility and provide people with a working system. I don't believe spending much time to ensure backward compatibility is worth the truble, you will never catch all cases. anyway since everybody on the list asking about raid and RedHat 5.2 has been instructed to trash the raidtools rpm and the rc.sysinit script there should be not many people running the old raid on rh52. but please please please someone find some time to rewrite the initscripts package. Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: Linux 2.2.1/raidtools-19990128-0.90 getting a mkraid: aborted
On Wed, Feb 10, 1999 at 07:10:56PM -0500, Stephen Hurrell wrote: I am using a linux 2.2.1 "stock" kernel from ftp.ca.kernel.org. I tried applying the latest raid0145 patch but it failed in many places so I assumed that it was already "rolled into 2.2.1". Is this correct? FYI: I tried the no, the latest raid patch raid0145-19990128-2.2.0 is not included in kernel 2.2.1 the patch should apply to 2.2.1 with only one reject, which, i believe, can be safely ignored. raid0145-19990128-2_2-1-devfs-v90.gz patch but there were failures applying it. I my patch raid0145-19990128-2.2.1-devfs-v90 is supposed to apply over kernel 2.2.1 with the devfs patch already applied (devfs can be found on www.atnf.csiro.au/~rgooch) but you DO need a very recent libc5 or glibc2.1 Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: RAID1 experiences
On Wed, Feb 10, 1999 at 09:43:12AM -0600, Chris Price wrote: We should not criticize Redhat for putting obsolete code on their distro. A while back, when Redhat followed a more bleeding edge approach, they constantly put beta software on their distro, and received much more criticism for it - hence the resulting obsolete inclusions. i'll answer since i was one of the first to point fingers at redhat for their raid implementation, and i did not do that because they put an old (but working, it is in beta, while all later patches are in alpha state) implementation of raid. But for the careless way they integrated raid into the distributions let's see /etc/rc.d/rc.sysinit they do activate swap check/mount/quotaon root activate modules/pnp/stuff load sound modules (?) raidrun -a (if it fails shell, then reboot) check non root fs (if it fails shell, then reboot) clear locks/tmp junk set clock/serials/modules unconditionally load the st module (doing some weird and useles greps ???) beside some unclean shell programming, they do not raidstop before reboot. So you have two cases in which you'll have to resync if you don't do raidstop after fixing in the shell prompt. 1) you have more than 1 md device and not all of them fail 2) fsck fails then take a look at /etc/rc.d/init.d/halt: kill process umount fs stop raid mount readonly all fs you could not umount halt just pray you did not have a hung process that kept you from unmounting that 100Gb raid partition Instead of pointing fingers at Redhat, I would ask if there is someone with teh Linux-raid community that actively corresponds with redhat to let them know of current status of linux-raid? Ingo etal. seem there was someone from redhat on this mailing list just begore they released 5.2 but they have disappeared since. to be doing a superb job in adding funtionality and fixing bugs quickly, but that does result in a myriad of patches being issued fairly regularily - is it Redhat's responsibility to keep track of linux-raid, or is it our responsibility to inform them of stable releases? I tought it was their responsability. Anyway I won't post any bugfix to a mailbot that answers suggesting you did something wrong and directs you somehwere else for help. Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: FC-AL support
On Wed, Feb 10, 1999 at 11:29:39AM -0500, Chris Mauritz wrote: From [EMAIL PROTECTED] Wed Feb 10 11:27:46 1999 Hi, it might be the wrong list(s) to ask, but I there is still a chance that someone can help me. Does anybody has some experiences with 'Fibre Channel' and know, if there is some support under Linux? Doesn't ICP Vortex make a fibre channel RAID controller that supports linux? I also believe that in Alan Cox patches to 2.2 there is a driver for a Qlogic fc controller. L. -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: 2.2.0pre8 and raid0145-19990108-2.2.0-pre5
On Thu, Jan 21, 1999 at 03:38:37PM +1100, Danny Yee wrote: When I apply raid0145-19990108-2.2.0-pre5 to a vanilla linux-2.2.0-pre8 tree, the patch works with just one error message patching file `linux/fs/buffer.c' Hunk #5 succeeded at 631 (offset -12 lines). Hunk #7 succeeded at 665 (offset -12 lines). Hunk #9 FAILED at 923. it did not work you have to look at linux/fs/buffer.c.rej and apply diffs by hand But I get immediate errors about missing include files when I try compiling. And I notice that linux/include/linux/md.h has been clobbered by the patch... it should not be needed anymore. P.S. Can someone direct me to the devfs patches? http://www.antf.csiro.au/~rgooch/ but it is NOT compatible with raid patches (well, until i find some time to release another raid+devfs patch) Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: Can't mkraid
On Wed, Jan 06, 1999 at 11:32:34PM -0800, Aaron D. Turner wrote: Handeling MD device /dev/md0 analysing super block invalid chunk size (0Kb) mkraid: aborting put a chunk-size line in your raidtab, eg. chunk-size 128 Regards, Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: Newbie RAID problems
On Wed, Jan 06, 1999 at 01:41:07PM +, John P . Looney wrote: Hi, sorry to bother you all about this, but I have a few questions. First: I'm working off the latest snapshot of Raidtools-0.90, and a 2.2pre1 kernel, with RAID-1 built in (not as a module). did you apply the raid patches to the kernel??? In the QuickStart.RAID file, it says that mkraid will work like: raid documentation is outdated, use the source Luke. . /dev/md0: Invalid argument is there anything in the syslog??? -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: RAID1 on / for RH 5.2
On Mon, Jan 04, 1999 at 05:59:33PM -, Jack Gavigan wrote: RAID. The Software RAID Howto is less of a HOWTO than a FAQ for experienced RAIDers... Took me about 20 minutes to figure out that mdadd and raidadd are actually the same thing. :) since the raid software is still in alpha stage the HOWTO/FAQ are inevitably outdated, and they will be unless mingo finds the time to keep a changelog Read the mailing list, someone was so kind to provide an archive of the list but i dun remembre the url My question, I guess, is: Am I right in thinking that it's necessary to create a script in rc2.d which executes raidrun and mount, and another in rc0.d which runs unmount and raidstop, to mount and unmount my raid partition? not with the newest raidtoools you just have to set the partition type to 0xfd The OS I'm using is Redhat 5.2, 2.0.38 kernel recompiled for SMP (2 2.0.38? am i missing something Pentiums, 320Mb RAM) with RAID support included at compile time. Version of raidtools in use is raidtools-0.50beta10 (off the RedHat CD). Setup is a bah, junk the raidtools supplied with redhat, junk also every reference to raidadd/raidstop in the /etc/rc.d... scripts and use the newest raidtools instead. RANT why the did they have to include an outdated alpha version of raidtools in their distro/RANT for building a root raid best thing is to install on disk 1 build a raid-with-only-one-disk on thisk 2 copy all files from disk 1 to raid, raidhotadd disk1 (leave /boot out of the raid otherwise lilo will complain). Regards. Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
halp in appliyng raid patches to 2.2.0pre4
hello i was trying to apply raid0145-19981215 to 2.2.0pre4 but int try_to_free_buffer(struct buffer_head * bh, struct buffer_head ** bhp) become int try_to_free_buffers(struct page * page_map) and buffer_head does not contain a reference to the page anymore linux/include/linux/fs.h does not define buffer_page anymore i am stuck any hints someone, Ingo??? -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: lilo and raid bootpartitions
a couple of notes the problem with lilo is not the root device, but the device where the kernel is stored, so if kernel is /boot/vmlinux and /boot is (say /dev/sda1) you don't need changing real-root-dev in initrd, just tell lilo: root=/dev/md0 and it wil work OK. i tried to hack lilo to support booting from a raid1 device, but it isn't so easy, lilo tries to determine the geometry of the md device and this is a no-no, probably i should try adding an option to lilo.conf like md0=/dev/sda1,/dev/sdb1 and then do things twice in the lilo code, but i am sooo lazy :((( L. -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: Help::ckraid RAID1 device on RH5.2
On Thu, Nov 12, 1998 at 07:14:57PM +0800, Enbo Zhang wrote: I use two SCSI partition(each 258M) as a raid1 device md1, and mount as /var on my RH5.2 box. get a fresh kernel, the latest raid patches from ftp.kernel.org and the latest raid tools also. set all raid partitions to type 0xfd and WIPE ALL REFERENCES to raidadd and raidstop from /etc/rc.d/init.d/halt /etc/rc.d/rc.sysinit (they must have smoked some very good floppyes when they wrote those) regards, Luca -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: your mail
On Thu, Nov 12, 1998 at 05:32:30PM -0700, Rob Vanderhoff wrote: Hello. Is there any way to add and remove scsi devices from an active md device on-the-fly? For example: I wish to have a raid1 device set up for if you using the raidtools 0.90 try : raidhotremove /dev/mdX /dev/sdY mount -o ro /dev/sdY /mnt;tar cvf /dev/stZ /mnt;umount /mnt (or dump 0f /dev/stZ /dev/sdY) raidhotadd /dev/mdX /dev/sdY L. -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
Re: Linear/ext2-volume question
you can concatenate any number of disks with linear, the problem lies in extending the filesystem on them. options are: formatting buying some product from PowerQuest and obtaining the resize2fs tool if i remember correctly ext2-volume was a patch Miguel de Icaza wrote a long ago and it was not definitely intended for the faint of heart, really if you don't know what ext2fs.h it is better you leave that patch alone. there is a rumor that someone is being paid by RedHat to develop a volume manager for linux, but i dun know nothing about this, regards, Luca On Sun, Oct 18, 1998 at 12:05:11PM +0100, Johan Grönvall wrote: Hi, I'm new to this list so please bare with me if I ask stupid questions. I'm looking for a kind of linear solution. I have however got the impression that you can only 'concatenate' 2 disks or partitions to make a single md device. Correct? And both disks need to be reformatted. Right? If that's the case, then who's actually using linear mode? Anyway, I found something that was called ext2-volume, a kind of extension to the ext2 filesystem, that made it possible to extend a mounted partition on the fly! Cool, but I don't know how to build it. It seems that I lack a file called ext2fs.h. Anyone tried this? Thanks, Johan -- Johan Grönvall Grönegatan 17B S-222 24 LUND home:+46 (0)46 184528 cellular:+46 0708 627588 [EMAIL PROTECTED] --- -- Luca Berra -- [EMAIL PROTECTED] System and Network Manager - CoMedia s.r.l.
