http://permalink.gmane.org/gmane.linux.ataraid hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.linux.ataraid/2874 <pre> With the exception of RAID 0, the advantage of BIOS support is the ability to boot from a degraded array. It is certainly possible to boot from a RAID array without BIOS support as long as the boot loader supports it. I have a number of systems booting from RAID 1 sets using LILO and an initrd with mdadm, but should the primary disk fail, the BIOS will probably still try to boot from the broken drive and fail. The only sensible way to set up fakeRAID is to use a driver with proper metadata support. Right now that driver doesn't exist for the vast majority of RAID5 fakeRAID systems unless one uses an older kernel and dmraid. For RAID1 arrays, dmraid still works with recent kernels. mdadm only supports Intel arrays and according to the developers, this situation is unlikely to change. </pre> Kristian Rasmussen 2013-05-18T13:51:48 http://permalink.gmane.org/gmane.linux.ataraid/2873 <pre>Sorry Kristian, by fakeraid I meant softraid. I actually boot off an mdadm RAID0 on my main system. SO it is not a requirement to use bios raid to boot off of these raid types. On Sat, May 18, 2013 at 2:04 PM, Ian Stakenvicius, Aerobiology Research &lt; ian&lt; at &gt;aerobiology.ca&gt; wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Danny 2013-05-18T13:26:45 http://permalink.gmane.org/gmane.linux.ataraid/2872 <pre> A small bit of terminology difference is being used, I think. Traditionally: hardware raid: raid cards that handle all HDD io independently. Bios raid / fake raid: what dmraid is meant for (or mdraid, for isw) software raid: pure software/kernel solution; using mdraid. now, as for booting -- as far as I know the only way to boot a full raid0 system is to use biosraid, since the bios cannot assemble the striped drives of a mdraid-0 in order to load the bootloader or kernel. yes this is probably a rare config (i would never run it) but if someone so chose, dmraid would be the tool for this job. On 2013-05-18, at 6:00 AM, Danny &lt;danwood76&lt; at &gt;gmail.com&gt; wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Ian Stakenvicius, Aerobiology Research 2013-05-18T13:04:05 http://permalink.gmane.org/gmane.linux.ataraid/2871 <pre>You can boot from mdadm fake RAID and it is far superior in terms of features and support/maintenence and the performance is the same. So anyone with a pure linux install would be better off in mdadm. On Fri, May 17, 2013 at 8:54 PM, Kristian Rasmussen &lt; kristian.rasmussen&lt; at &gt;broadpark.no&gt; wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Danny 2013-05-18T10:00:40 http://permalink.gmane.org/gmane.linux.ataraid/2870 <pre>Or anybody who wants to boot from fakeRAID. </pre> Kristian Rasmussen 2013-05-17T19:54:14 http://permalink.gmane.org/gmane.linux.ataraid/2869 <pre>Dmraid does still work, but installing requires a bit of knowledge of how it all works and a little tinkering (especially on Ubuntu where the installer is a pile of shit). Unfortunately only the isw stuff has been implemented in mdadm so far and there are no plans to support anything else. The problem with dmraid is that the only people who use it are dual booting windows, and unfortunately any dev that delves in is usually going to go all the way to Linux at some point so their reliance on dmraid dies and they move on to more interesting projects. I was helping maintain the Ubuntu stuff for a while but my main machine now uses a RAID0 through mdadm as I have no use for dmraid. I still try to test where I can but upstream never accept any of the patches we proposed so the dmraid source archives available are outdated and broken for the lesser used metadata types. On Fri, May 17, 2013 at 6:16 PM, Bad Bod &lt;badbod&lt; at &gt;badbod.com&gt; wrote: _______________________________________________ Ataraid-list mailing lis</pre> Danny 2013-05-17T18:10:52 http://permalink.gmane.org/gmane.linux.ataraid/2868 <pre>The problem is that mdadm doesn't support 3rd party metadata formats, Intel being a notable exception. There's no way you can get mdadm to detect and activate a fakeRAID set created by a VIA, SiS or Nvidia controller. This is exactly what dmraid was designed to do, and it works very well with mirror sets (RAID 1). dmraid uses the device mapper RAID modules to access RAID sets. Unfortunately, for RAID 4/5 sets it relies on a module not found in the mailline kernel ("dm-raid45"). The kernel currently contains a module providing similar functionality and more ("dm-raid456"), but dmraid in it's current state cannot use it, and the "dm-raid45" patches doesn't work with recent kernels. So, currently dmraid supports fakeRAID 1 sets using the dm-mirror module, and mdadm can read Intel fakeRAID sets. dmraid cannot activate fakeRAID 5/6 sets, and mdadm cannot activate any non-Intel fakeRAID sets regardless of RAID level. (Personally, I wish someone would take it upon themselves to fix dmraid. There's nothing wrong</pre> Kristian Rasmussen 2013-05-17T17:42:38 http://permalink.gmane.org/gmane.linux.ataraid/2867 <pre>Dear Dead, Apologies. this has been a dead project for quite some time now. I think the response from Heinz is an official 'this project is dead' notice. Use mdadm instead. Now find their support group. _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Bad Bod 2013-05-17T17:16:46 http://permalink.gmane.org/gmane.linux.ataraid/2866 <pre>Hi Heinz, Seeing as all the major distro's use this for detecting RAID during install from live CD, and right now they all fail, then you might suggest to them that this project is dead and you no longer support. Quite why they started using this for RAID, I am not sure. It did work, then got broken. Maybe they still think it works? It does not. Please ask the major distro's to start using mdadm instead of this broken thing they are still using. Regards David On Mon, May 13, 2013 at 4:59 PM, Heinz Mauelshagen &lt;heinzm&lt; at &gt;redhat.com&gt;wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Bad Bod 2013-05-17T16:46:48 http://permalink.gmane.org/gmane.linux.ataraid/2865 <pre>Hi Heinz, thanks for your response! I tried with mdadm --detail --scan but don't show anything. Maybe I forgetting something. It's running with this version of mdadm mdadm - v2.6.9 - 10th March 2009 there is a way to fix this with mdadm? Thanks! Date: Mon, 13 May 2013 11:59:43 +0200 From: heinzm&lt; at &gt;redhat.com To: ataraid-list&lt; at &gt;redhat.com Subject: Re: Rebuilding a Raid 1 Have you tried mdadm? isw should be driven via md by now. Heinz On 04/25/2013 06:55 AM, Dead Gardens wrote: dmraid version # dmraid -V dmraid version: 1.0.0.rc13-64 (2010.11.04) debug dmraid library version: 1.0.0.rc13-17 (2009.06.10) device-mapper version: 4.11.5 From: dead_gardens_&lt; at &gt;hotmail.com To: ataraid-list&lt; at &gt;redhat.com Subject: Rebuilding a Raid 1 Date: Thu, 25 Apr 2013 06:52:01 +0200 I'm tryi</pre> Dead Gardens 2013-05-17T16:20:46 http://permalink.gmane.org/gmane.linux.ataraid/2864 <pre> Have you tried mdadm? isw should be driven via md by now. Heinz On 04/25/2013 06:55 AM, Dead Gardens wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Heinz Mauelshagen 2013-05-13T09:59:43 http://permalink.gmane.org/gmane.linux.ataraid/2863 <pre>dmraid version # dmraid -V dmraid version: 1.0.0.rc13-64 (2010.11.04) debug dmraid library version: 1.0.0.rc13-17 (2009.06.10) device-mapper version: 4.11.5 From: dead_gardens_&lt; at &gt;hotmail.com To: ataraid-list&lt; at &gt;redhat.com Subject: Rebuilding a Raid 1 Date: Thu, 25 Apr 2013 06:52:01 +0200 I'm trying to rebuild a raid 1 (dmraid+isw) unsuccessfully. I replaced the failed disk with a new one and the BIOS added it automatically to the raid. Running kernel 2.6.18-194.17.4.el5. # dmraid -r /dev/sda: isw, "isw_babcjifefe", GROUP, ok, 1953525165 sectors, data&lt; at &gt; 0 /dev/sdb: isw, "isw_babcjifefe", GROUP, ok, 1953525165 sectors, data&lt; at &gt; 0 # dmraid -s *** Group superset isw_babcjifefe --&gt; Subset name : isw_babcjifefe_Raid0 size : 1953519616 stride : 128 type : mirror status : nosync subsets: 0 devs : 2 spares : 0 When i try to start the raid i receive the next errors # dmraid -f isw -S -M /dev/sdb ERROR: isw: SPARE disk must use all space on the disk # dmraid -tay isw_ba</pre> Dead Gardens 2013-04-25T04:55:23 http://permalink.gmane.org/gmane.linux.ataraid/2862 <pre> I'm trying to rebuild a raid 1 (dmraid+isw) unsuccessfully. I replaced the failed disk with a new one and the BIOS added it automatically to the raid. Running kernel 2.6.18-194.17.4.el5. # dmraid -r /dev/sda: isw, "isw_babcjifefe", GROUP, ok, 1953525165 sectors, data&lt; at &gt; 0 /dev/sdb: isw, "isw_babcjifefe", GROUP, ok, 1953525165 sectors, data&lt; at &gt; 0 # dmraid -s *** Group superset isw_babcjifefe --&gt; Subset name : isw_babcjifefe_Raid0 size : 1953519616 stride : 128 type : mirror status : nosync subsets: 0 devs : 2 spares : 0 When i try to start the raid i receive the next errors # dmraid -f isw -S -M /dev/sdb ERROR: isw: SPARE disk must use all space on the disk # dmraid -tay isw_babcjifefe_Raid0: 0 1953519616 mirror core 3 131072 sync block_on_error 2 /dev/sda 0 /dev/sdb 0 # dmraid -ay RAID set "isw_babcjifefe_Raid0" was not activated ERROR: device "isw_babcjifefe_Raid0" could not be found # dmraid -f isw -S -M /dev/sdb ERROR: isw: SPARE disk must use all space on the disk # </pre> Dead Gardens 2013-04-25T04:52:01 http://permalink.gmane.org/gmane.linux.ataraid/2861 <pre>Has the checksum error for Via (8237) RAID been fixed. That the checksum could never be met was pointed out some time ago. I do realize that this chipset is now, very probably, 'obsolete' , but it did work once. Regression ignored is surely satanism? :-) Regards David On Fri, Nov 23, 2012 at 9:27 PM, Ian Stakenvicius, Aerobiology Research &lt; ian&lt; at &gt;aerobiology.ca&gt; wrote: _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Bad Bod 2012-11-23T18:12:14 http://permalink.gmane.org/gmane.linux.ataraid/2860 <pre> I'm guessing that what Heinz means is that dmraid reads superblocks or whatever in accordance with the sector size on the hardware itself ("hard" = hardware level); since it does this in order to feed the correct setup info to device-mapper it would not be acting on virtual devices.. Just my interpretation though, please correct me if i'm wrong. _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Ian Stakenvicius, Aerobiology Research 2012-11-23T14:27:05 http://permalink.gmane.org/gmane.linux.ataraid/2859 <pre>-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 11/21/2012 9:19 AM, Heinz Mauelshagen wrote: What do you mean? dm devices are virtual so there is no "hard" sector size; it's all virtual. Are you saying it is hard coded to report 512 bytes and can't be changed? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (MingW32) Comment: Using GnuPG with undefined - http://www.enigmail.net/ iQEcBAEBAgAGBQJQrULGAAoJEJrBOlT6nu75TP8H/0cr8CtKOtVJ/5f1DIny8xdT 5AEPkiSEwfGTPPkS3Xm2Rm0cyvtfNUHGcLIAFqq2hd9Ky5hJjrR8iCRqyPkbcGmi sr6RgjFgZVUt048scPOf5QAGkzV2KzyHkrdAU7nqa7LIAXBJdPCEYOf3qIzeMBDm Z6cTaDAz/pdjYD/7LzNHeQ0KpQzEhmZcBeCu1i87V2be3lG8o7+1sWOygQdPTGOQ DjAIVp4k12Xe3VL5Mbbtu6QS4vYQwPMRz4q9uqGUbC0+kCBgerV4nNkkyJrVs3m5 +/QeKroDeFybdUcTkViyEPZknYJLf1Rq35thPNRYUhkPTgQb8cBnLg45KXwapTY= =OPJq -----END PGP SIGNATURE----- </pre> Phillip Susi 2012-11-21T21:08:22 http://permalink.gmane.org/gmane.linux.ataraid/2858 <pre> Does md support variable sector sizes. That will give me another reason to finish the super-pdc.c for mdadm. -- Mark-Willem _______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Mark-Willem Jansen 2012-11-21T15:45:49 http://permalink.gmane.org/gmane.linux.ataraid/2857 <pre> No, dm goes by hard sector sizes, not pretended ones. Heinz </pre> Heinz Mauelshagen 2012-11-21T14:19:29 http://permalink.gmane.org/gmane.linux.ataraid/2856 <pre>-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 It appears that PDC arrays &gt; 2 TB pretend they have &gt; 512 byte sector sizes. I'm trying to figure out how to patch dmraid to support this but I'm not sure how to tell device-mapper to use a different sector size. Is this possible? -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (MingW32) Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iQEcBAEBAgAGBQJQoUsoAAoJEJrBOlT6nu75n9kIAIS8hdtJHFk5sOPBBcOZ+Htq rUOOBxHW3cCjHyAqXgAlazRDLeP2qSGODjy4e54ovKxGbN1QZ5yJhX/iPNpGVUFb bypZzYj5wjGBHmhmKD+5UKjBxHdrjOkluNCREdGDDoE9aRIyR+1JPjWEzJJDLvaO fnNoiHg3hQi+RgaOj2oHM2eGexYAFAsKI0qswRXbTnAfrcMCfmm2pf49b76zdZ7y Z6c+xVl43PhDfZiy/EiWqYkrNtagvin7CNjahW5zptmNdf1H7JAzpuXbf3uZ6g5F 9MkDn5BFdJOBwEL/WTi9ZBgbIn4aIf8CwsBrsQ9DC4FFoZ3hnZTW2lT6/2UpHMk= =pXpo -----END PGP SIGNATURE----- </pre> Phillip Susi 2012-11-12T19:16:56 http://permalink.gmane.org/gmane.linux.ataraid/2855 <pre>Hello, just have a minute to reply cuz Im working. I'll send you my page on here 4 my pics and some stuff. Give me a call or txt if you are interested. K gotta go lata. Sent from my iPhone_______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Staci Page 2012-11-02T16:18:33 http://permalink.gmane.org/gmane.linux.ataraid/2854 <pre>Hi again! Are ya any good at giving massages? Work is kinda killin me and I haven't had one in awhile. I can host if you'd prefer, just let me know when is good for you. I have a flexible schedule and do online school in the meantime so I will make something work. I've got photos on this page if you would want to look. Do the screening thing and it'll give you my phone # and we can text or talk to get to know each other a little better. So far so good though! I realize it is a little weird to do the screening, but a few of my girl friends who have met up on Craigslist turned me onto it since a friend of ours got raped when she met up with someone from another site and its worked out fine for them. I'm sure you see my logic. Send me a message so I know ya got my number. | |} TYL_______________________________________________ Ataraid-list mailing list Ataraid-list&lt; at &gt;redhat.com https://www.redhat.com/mailman/listinfo/ataraid-list</pre> Staci Page 2012-11-02T15:24:15 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.linux.ataraid