Annotation of wikisrc/guide/raidframe.mdwn, revision 1.2

1.1       jdf         1: # NetBSD RAIDframe
                      2: 
                      3: ## RAIDframe Introduction
                      4: 
                      5: ### About RAIDframe
                      6: 
                      7: NetBSD uses the [CMU RAIDframe](http://www.pdl.cmu.edu/RAIDframe/) software for 
                      8: its RAID subsystem. NetBSD is the primary platform for RAIDframe development. 
                      9: RAIDframe can also be found in older versions of FreeBSD and OpenBSD. NetBSD 
                     10: also has another way of bundling disks, the 
                     11: [ccd(4)](http://netbsd.gw.com/cgi-bin/man-cgi?ccd+4+NetBSD-5.0.1+i386) subsystem 
                     12: (see [Concatenated Disk Device](guide/ccd)). You should possess some [basic 
                     13: knowledge](http://www.acnc.com/04_00.html) about RAID concepts and terminology 
                     14: before continuing. You should also be at least familiar with the different 
                     15: levels of RAID - Adaptec provides an [excellent 
                     16: reference](http://www.adaptec.com/en-US/_common/compatibility/_education/RAID_level_compar_wp.htm), 
                     17: and the [raid(4)](http://netbsd.gw.com/cgi-bin/man-cgi?raid+4+NetBSD-5.0.1+i386) 
                     18: manpage contains a short overview too.
                     19: 
                     20: ### A warning about Data Integrity, Backups, and High Availability
                     21: 
                     22: RAIDframe is a Software RAID implementation, as opposed to Hardware RAID. As 
                     23: such, it does not need special disk controllers supported by NetBSD. System 
                     24: administrators should give a great deal of consideration to whether software 
                     25: RAID or hardware RAID is more appropriate for their "Mission Critical" 
                     26: applications. For some projects you might consider the use of many of the 
                     27: hardware RAID devices [supported by 
                     28: NetBSD](http://www.NetBSD.org/support/hardware/). It is truly at your discretion 
                     29: what type of RAID you use, but it is recommend that you consider factors such 
                     30: as: manageability, commercial vendor support, load-balancing and failover, etc.
                     31: 
                     32: Depending on the RAID level used, RAIDframe does provide redundancy in the event 
                     33: of a hardware failure. However, it is *not* a replacement for reliable backups! 
                     34: Software and user-error can still cause data loss. RAIDframe may be used as a 
                     35: mechanism for facilitating backups in systems without backup hardware, but this 
                     36: is not an ideal configuration. Finally, with regard to "high availability", RAID 
                     37: is only a very small component to ensuring data availability.
                     38: 
                     39: Once more for good measure: *Back up your data!*
                     40: 
                     41: ### Hardware versus Software RAID
                     42: 
                     43: If you run a server, it will most probably already have a Hardware RAID 
                     44: controller. There are reasons for and against using a Software RAID, depending 
                     45: on the scenario.
                     46: 
                     47: In general, a Software RAID is well suited for low-IO system disks. If you run a 
                     48: Software RAID, you can exchange disks and disk controllers, or even move the 
                     49: disks to a completely different machine. The computational overhead for the RAID 
                     50: is negligible if there is only few disk IO operations.
                     51: 
                     52: If you need much IO, you should use a Hardware RAID. With a Software RAID, the 
                     53: redundancy data has to be transferred via the bus your disk controller is 
                     54: connected to. With a Hardware RAID, you transfer data only once - the redundancy 
                     55: computation and transfer is done by the controller.
                     56: 
                     57: ### Getting Help
                     58: 
                     59: If you encounter problems using RAIDframe, you have several options for 
                     60: obtaining help.
                     61: 
                     62:  1. Read the RAIDframe man pages: 
                     63:     [raid(4)](http://netbsd.gw.com/cgi-bin/man-cgi?raid+4+NetBSD-5.0.1+i386) and 
                     64:     [raidctl(8)](http://netbsd.gw.com/cgi-bin/man-cgi?raidctl+8+NetBSD-5.0.1+i386) 
                     65:     thoroughly.
                     66: 
                     67:  2. Search the mailing list archives. Unfortunately, there is no NetBSD list 
                     68:     dedicated to RAIDframe support. Depending on the nature of the problem, posts
                     69:     tend to end up in a variety of lists. At a very minimum, search
                     70:     [netbsd-help](http://mail-index.NetBSD.org/netbsd-help/),
                     71:     [netbsd-users@NetBSD.org](http://mail-index.NetBSD.org/netbsd-users/),
                     72:     [current-users@NetBSD.org](http://mail-index.NetBSD.org/current-users/). Also
                     73:     search the list for the NetBSD platform on which you are using RAIDframe:
                     74:     port-*`${ARCH}`*@NetBSD.org.
                     75: 
                     76:     ### Caution
                     77: 
                     78:        Because RAIDframe is constantly undergoing development, some information in 
                     79:        mailing list archives has the potential of being dated and inaccurate.
                     80: 
                     81:  3. Search the [Problem Report 
                     82:     database](http://www.NetBSD.org/support/send-pr.html).
                     83: 
                     84:  4. If your problem persists: Post to the mailing list most appropriate 
                     85:     (judgment call). Collect as much verbosely detailed information as possible 
                     86:     before posting: Include your 
                     87:     [dmesg(8)](http://netbsd.gw.com/cgi-bin/man-cgi?dmesg+8+NetBSD-5.0.1+i386) 
                     88:     output from `/var/run/dmesg.boot`, your kernel 
                     89:     [config(5)](http://netbsd.gw.com/cgi-bin/man-cgi?config+5+NetBSD-5.0.1+i386) , 
                     90:     your `/etc/raid[0-9].conf`, any relevant errors on `/dev/console`, 
                     91:     `/var/log/messages`, or to `stdout/stderr` of 
                     92:     [raidctl(8)](http://netbsd.gw.com/cgi-bin/man-cgi?raidctl+8+NetBSD-5.0.1+i386). 
                     93:     The output of **raidctl -s** (if available) will be useful as well. Also 
                     94:     include details on the troubleshooting steps you've taken thus far, exactly 
                     95:     when the problem started, and any notes on recent changes that may have 
                     96:     prompted the problem to develop. Remember to be patient when waiting for a 
                     97:     response.
                     98: 
                     99: ## Setup RAIDframe Support
                    100: 
                    101: The use of RAID will require software and hardware configuration changes.
                    102: 
                    103: ### Kernel Support
                    104: 
                    105: The GENERIC kernel already has support for RAIDframe. If you have built a custom 
                    106: kernel for your environment the kernel configuration must have the following 
                    107: options:
                    108: 
                    109:     pseudo-device   raid            8       # RAIDframe disk driver
                    110:     options         RAID_AUTOCONFIG         # auto-configuration of RAID components
                    111: 
                    112: The RAID support must be detected by the NetBSD kernel, which can be checked by 
                    113: looking at the output of the 
                    114: [dmesg(8)](http://netbsd.gw.com/cgi-bin/man-cgi?dmesg+8+NetBSD-5.0.1+i386) 
                    115: command.
                    116: 
                    117:     # dmesg|grep -i raid
                    118:     Kernelized RAIDframe activated
                    119: 
                    120: Historically, the kernel must also contain static mappings between bus addresses 
                    121: and device nodes in `/dev`. This used to ensure consistency of devices within 
                    122: RAID sets in the event of a device failure after reboot. Since NetBSD 1.6, 
                    123: however, using the auto-configuration features of RAIDframe has been recommended 
                    124: over statically mapping devices. The auto-configuration features allow drives to 
                    125: move around on the system, and RAIDframe will automatically determine which 
                    126: components belong to which RAID sets.
                    127: 
                    128: ### Power Redundancy and Disk Caching
                    129: 
                    130: If your system has an Uninterruptible Power Supply (UPS), if your system has 
                    131: redundant power supplies, or your disk controller has a battery, you should 
                    132: consider enabling the read and write caches on your drives. On systems with 
                    133: redundant power, this will improve drive performance. On systems without 
                    134: redundant power, the write cache could endanger the integrity of RAID data in 
                    135: the event of a power loss.
                    136: 
                    137: The [dkctl(8)](http://netbsd.gw.com/cgi-bin/man-cgi?dkctl+8+NetBSD-5.0.1+i386) 
                    138: utility to can be used for this on all kinds of disks that support the operation 
                    139: (SCSI, EIDE, SATA, ...):
                    140: 
                    141:     # dkctl wd0 getcache
                    142:     /dev/rwd0d: read cache enabled
                    143:     /dev/rwd0d: read cache enable is not changeable
                    144:     /dev/rwd0d: write cache enable is changeable
                    145:     /dev/rwd0d: cache parameters are not savable
                    146:     # dkctl wd0 setcache rw
                    147:     # dkctl wd0 getcache
                    148:     /dev/rwd0d: read cache enabled
                    149:     /dev/rwd0d: write-back cache enabled
                    150:     /dev/rwd0d: read cache enable is not changeable
                    151:     /dev/rwd0d: write cache enable is changeable
                    152:     /dev/rwd0d: cache parameters are not savable
                    153: 
                    154: ## Example: RAID-1 Root Disk
                    155: 
                    156: This example explains how to setup RAID-1 root disk. With RAID-1 components are 
                    157: mirrored and therefore the server can be fully functional in the event of a 
                    158: single component failure. The goal is to provide a level of redundancy that will 
                    159: allow the system to encounter a component failure on either component disk in 
                    160: the RAID and:
                    161: 
                    162:  * Continue normal operations until a maintenance window can be scheduled.
                    163:  * Or, in the unlikely event that the component failure causes a system reboot, 
                    164:    be able to quickly reconfigure the system to boot from the remaining
                    165:    component (platform dependent).
                    166: 
1.2     ! jdf       167: ![RAID-1 Disk Logical Layout](guide/images/rf-raidL1-diskdia.png)
1.1       jdf       168: **RAID-1 Disk Logical Layout**
                    169: 
                    170: Because RAID-1 provides both redundancy and performance improvements, its most 
                    171: practical application is on critical "system" partitions such as `/`, `/usr`, 
                    172: `/var`, `swap`, etc., where read operations are more frequent than write 
                    173: operations. For other file systems, such as `/home` or `/var/`, other RAID 
                    174: levels might be considered (see the references above). If one were simply 
                    175: creating a generic RAID-1 volume for a non-root file system, the cookie-cutter 
                    176: examples from the man page could be followed, but because the root volume must 
                    177: be bootable, certain special steps must be taken during initial setup.
                    178: 
                    179: *Note*: This example will outline a process that differs only slightly between 
                    180: the i386 and sparc64 platforms. In an attempt to reduce excessive duplication of 
                    181: content, where differences do exist and are cosmetic in nature, they will be 
                    182: pointed out using a section such as this. If the process is drastically 
                    183: different, the process will branch into separate, platform dependent steps.
                    184: 
                    185: ### Pseudo-Process Outline
                    186: 
                    187: Although a much more refined process could be developed using a custom copy of 
                    188: NetBSD installed on custom-developed removable media, presently the NetBSD 
                    189: install media lacks RAIDframe tools and support, so the following pseudo process
                    190: has become the de facto standard for setting up RAID-1 Root.
                    191: 
                    192:  1. Install a stock NetBSD onto Disk0 of your system.
                    193: 
1.2     ! jdf       194:     ![Perform generic install onto Disk0/wd0](guide/images/rf-r1r-pp1.png)
1.1       jdf       195:     **Perform generic install onto Disk0/wd0**
                    196: 
                    197:  2. Use the installed system on Disk0/wd0 to setup a RAID Set composed of 
                    198:     Disk1/wd1 only.
                    199: 
                    200:     ![Setup RAID Set](rf-r1r-pp2.png)
                    201:     **Setup RAID Set**
                    202: 
                    203:  3. Reboot the system off the Disk1/wd1 with the newly created RAID volume.
                    204: 
1.2     ! jdf       205:     ![Reboot using Disk1/wd1 of RAID](guide/images/rf-r1r-pp3.png)
1.1       jdf       206:     **Reboot using Disk1/wd1 of RAID**
                    207: 
                    208:  4. Add / re-sync Disk0/wd0 back into the RAID set.
                    209: 
1.2     ! jdf       210:     ![Mirror Disk1/wd1 back to Disk0/wd0](guide/images/rf-r1r-pp4.png)
1.1       jdf       211:     **Mirror Disk1/wd1 back to Disk0/wd0**
                    212: 
                    213: ### Hardware Review
                    214: 
                    215: At present, the alpha, amd64, i386, pmax, sparc, sparc64, and vax NetBSD 
                    216: platforms support booting from RAID-1. Booting is not supported from any other 
                    217: RAID level. Booting from a RAID set is accomplished by teaching the 1st stage 
                    218: boot loader to understand both 4.2BSD/FFS and RAID partitions. The 1st boot 
                    219: block code only needs to know enough about the disk partitions and file systems 
                    220: to be able to read the 2nd stage boot blocks. Therefore, at any time, the 
                    221: system's BIOS / firmware must be able to read a drive with 1st stage boot blocks 
                    222: installed. On the i386 platform, configuring this is entirely dependent on the 
                    223: vendor of the controller card / host bus adapter to which your disks are 
                    224: connected. On sparc64 this is controlled by the IEEE 1275 Sun OpenBoot Firmware.
                    225: 
                    226: This article assumes two identical IDE disks (`/dev/wd{0,1}`) which we are going 
                    227: to mirror (RAID-1). These disks are identified as:
                    228: 
                    229:     # grep ^wd /var/run/dmesg.boot
                    230:     wd0 at atabus0 drive 0: <WDC WD100BB-75CLB0>
                    231:     wd0: drive supports 16-sector PIO transfers, LBA addressing
                    232:     wd0: 9541 MB, 19386 cyl, 16 head, 63 sec, 512 bytes/sect x 19541088 sectors
                    233:     wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 5 (Ultra/100)
                    234:     wd0(piixide0:0:0): using PIO mode 4, Ultra-DMA mode 2 (Ultra/33) (using DMA data transfers)
                    235:     
                    236:     wd1 at atabus1 drive 0: <WDC WD100BB-75CLB0>
                    237:     wd1: drive supports 16-sector PIO transfers, LBA addressing
                    238:     wd1: 9541 MB, 19386 cyl, 16 head, 63 sec, 512 bytes/sect x 19541088 sectors
                    239:     wd1: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 5 (Ultra/100)
                    240:     wd1(piixide0:1:0): using PIO mode 4, Ultra-DMA mode 2 (Ultra/33) (using DMA data transfers)
                    241: 
                    242: *Note*: If you are using SCSI, replace `/dev/{,r}wd{0,1}` with 
                    243: `/dev/{,r}sd{0,1}`.
                    244: 
                    245: In this example, both disks are jumpered as Master on separate channels on the 
                    246: same controller. You usually wouldn't want to have both disks on the same bus on 
                    247: the same controller; this creates a single point of failure. Ideally you would 
                    248: have the disks on separate channels on separate controllers. Nonetheless, in 
                    249: most cases the most critical point is the hard disk, so having redundant 
                    250: channels or controllers is not that important. Plus, having more channels or 
                    251: controllers increases costs. Some SCSI controllers have multiple channels on the 
                    252: same controller, however, a SCSI bus reset on one channel could adversely affect 
                    253: the other channel if the ASIC/IC becomes overloaded. The trade-off with two 
                    254: controllers is that twice the bandwidth is used on the system bus. For purposes 
                    255: of simplification, this example shows two disks on different channels on the 
                    256: same controller.
                    257: 
                    258: *Note*: RAIDframe requires that all components be of the same size. Actually, it 
                    259: will use the lowest common denominator among components of dissimilar sizes. For 
                    260: purposes of illustration, the example uses two disks of identical geometries. 
                    261: Also, consider the availability of replacement disks if a component suffers a 
                    262: critical hardware failure.
                    263: 
                    264: *Tip*: Two disks of identical vendor model numbers could have different 
                    265: geometries if the drive possesses "grown defects". Use a low-level program to 
                    266: examine the grown defects table of the disk. These disks are obviously 
                    267: suboptimal candidates for use in RAID and should be avoided.
                    268: 
                    269: ### Initial Install on Disk0/wd0
                    270: 
                    271: Perform a very generic installation onto your Disk0/wd0. Follow the `INSTALL` 
                    272: instructions for your platform. Install all the sets but do not bother 
                    273: customizing anything other than the kernel as it will be overwritten.
                    274: 
                    275: *Tip*: On i386, during the sysinst install, when prompted if you want to `use 
                    276: the entire disk for NetBSD`, answer `yes`.
                    277: 
                    278:  * [Installing NetBSD: Preliminary considerations and preparations](guide/inst)
                    279:  * [NetBSD/i386 Install Directions](http://ftp.NetBSD.org/pub/NetBSD/NetBSD-5.0.2/i386/INSTALL.html)
                    280:  * [NetBSD/sparc64 Install Directions](http://ftp.NetBSD.org/pub/NetBSD/NetBSD-5.0.2/sparc64/INSTALL.html)
                    281: 
                    282: Once the installation is complete, you should examine the 
                    283: [disklabel(8)](http://netbsd.gw.com/cgi-bin/man-cgi?disklabel+8+NetBSD-5.0.1+i386) 
                    284: and [fdisk(8)](http://netbsd.gw.com/cgi-bin/man-cgi?fdisk+8+NetBSD-5.0.1+i386) / 
                    285: [sunlabel(8)](http://netbsd.gw.com/cgi-bin/man-cgi?sunlabel+8+NetBSD-5.0.1+i386) 
                    286: outputs on the system:
                    287: 
                    288:     # df
                    289:     Filesystem   1K-blocks        Used       Avail %Cap Mounted on
                    290:     /dev/wd0a       9487886      502132     8511360   5% /
                    291: 
                    292: On i386:
                    293: 
                    294:     # disklabel -r wd0
                    295:     type: unknown
                    296:     disk: Disk00
                    297:     label:
                    298:     flags:
                    299:     bytes/sector: 512
                    300:     sectors/track: 63
                    301:     tracks/cylinder: 16
                    302:     sectors/cylinder: 1008
                    303:     cylinders: 19386
                    304:     total sectors: 19541088
                    305:     rpm: 3600
                    306:     interleave: 1
                    307:     trackskew: 0
                    308:     cylinderskew: 0
                    309:     headswitch: 0           # microseconds
                    310:     track-to-track seek: 0  # microseconds
                    311:     drivedata: 0
                    312:     
                    313:     16 partitions:
                    314:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    315:      a:  19276992        63     4.2BSD   1024  8192 46568  # (Cyl.      0* - 19124*)
                    316:      b:    264033  19277055       swap                     # (Cyl.  19124* - 19385)
                    317:      c:  19541025        63     unused      0     0        # (Cyl.      0* - 19385)
                    318:      d:  19541088         0     unused      0     0        # (Cyl.      0 - 19385)
                    319:     
                    320:     # fdisk /dev/rwd0d
                    321:     Disk: /dev/rwd0d
                    322:     NetBSD disklabel disk geometry:
                    323:     cylinders: 19386, heads: 16, sectors/track: 63 (1008 sectors/cylinder)
                    324:     total sectors: 19541088
                    325:     
                    326:     BIOS disk geometry:
                    327:     cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
                    328:     total sectors: 19541088
                    329:     
                    330:     Partition table:
                    331:     0: NetBSD (sysid 169)
                    332:         start 63, size 19541025 (9542 MB, Cyls 0-1216/96/1), Active
                    333:     1: <UNUSED>
                    334:     2: <UNUSED>
                    335:     3: <UNUSED>
                    336:     Bootselector disabled.
                    337:     First active partition: 0
                    338: 
                    339: On Sparc64 the command and output differ slightly:
                    340: 
                    341:     # disklabel -r wd0
                    342:     type: unknown
                    343:     disk: Disk0
                    344:     [...snip...]
                    345:     8 partitions:
                    346:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    347:      a:  19278000         0     4.2BSD   1024  8192 46568  # (Cyl.      0 -  19124)
                    348:      b:    263088  19278000       swap                     # (Cyl.  19125 -  19385)
                    349:      c:  19541088         0     unused      0     0        # (Cyl.      0 -  19385)
                    350:     
                    351:     # sunlabel /dev/rwd0c
                    352:     sunlabel> P
                    353:     a: start cyl =      0, size = 19278000 (19125/0/0 - 9413.09Mb)
                    354:     b: start cyl =  19125, size =   263088 (261/0/0 - 128.461Mb)
                    355:     c: start cyl =      0, size = 19541088 (19386/0/0 - 9541.55Mb)
                    356: 
                    357: ### Preparing Disk1/wd1
                    358: 
                    359: Once you have a stock install of NetBSD on Disk0/wd0, you are ready to begin. 
                    360: Disk1/wd1 will be visible and unused by the system. To setup Disk1/wd1, you will 
                    361: use 
                    362: [disklabel(8)](http://netbsd.gw.com/cgi-bin/man-cgi?disklabel+8+NetBSD-5.0.1+i386) 
                    363: to allocate the entire second disk to the RAID-1 set.
                    364: 
                    365: *Tip*: The best way to ensure that Disk1/wd1 is completely empty is to 'zero' 
                    366: out the first few sectors of the disk with 
                    367: [dd(1)](http://netbsd.gw.com/cgi-bin/man-cgi?dd+1+NetBSD-5.0.1+i386) . This will 
                    368: erase the MBR (i386) or Sun disk label (sparc64), as well as the NetBSD disk 
                    369: label. If you make a mistake at any point during the RAID setup process, you can 
                    370: always refer to this process to restore the disk to an empty state.
                    371: 
                    372: *Note*: On sparc64, use `/dev/rwd1c` instead of `/dev/rwd1d`!
                    373: 
                    374:     # dd if=/dev/zero of=/dev/rwd1d bs=8k count=1
                    375:     1+0 records in
                    376:     1+0 records out
                    377:     8192 bytes transferred in 0.003 secs (2730666 bytes/sec)
                    378: 
                    379: Once this is complete, on i386, verify that both the MBR and NetBSD disk labels 
                    380: are gone. On sparc64, verify that the Sun Disk label is gone as well.
                    381: 
                    382: On i386:
                    383: 
                    384:     # fdisk /dev/rwd1d
                    385:     
                    386:     fdisk: primary partition table invalid, no magic in sector 0
                    387:     Disk: /dev/rwd1d
                    388:     NetBSD disklabel disk geometry:
                    389:     cylinders: 19386, heads: 16, sectors/track: 63 (1008 sectors/cylinder)
                    390:     total sectors: 19541088
                    391:     
                    392:     BIOS disk geometry:
                    393:     cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
                    394:     total sectors: 19541088
                    395:     
                    396:     Partition table:
                    397:     0: <UNUSED>
                    398:     1: <UNUSED>
                    399:     2: <UNUSED>
                    400:     3: <UNUSED>
                    401:     Bootselector disabled.
                    402:     
                    403:     # disklabel -r wd1
                    404:     
                    405:     [...snip...]
                    406:     16 partitions:
                    407:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    408:      c:  19541025        63     unused      0     0        # (Cyl.      0* - 19385)
                    409:      d:  19541088         0     unused      0     0        # (Cyl.      0 - 19385)
                    410: 
                    411: On sparc64:
                    412: 
                    413:     # sunlabel /dev/rwd1c
                    414:     
                    415:     sunlabel: bogus label on `/dev/wd1c' (bad magic number)
                    416:     
                    417:     # disklabel -r wd1
                    418:     
                    419:     [...snip...]
                    420:     3 partitions:
                    421:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    422:      c:  19541088         0     unused      0     0        # (Cyl.      0 -  19385)
                    423:     disklabel: boot block size 0
                    424:     disklabel: super block size 0
                    425: 
                    426: Now that you are certain the second disk is empty, on i386 you must establish 
                    427: the MBR on the second disk using the values obtained from Disk0/wd0 above. We 
                    428: must remember to mark the NetBSD partition active or the system will not boot. 
                    429: You must also create a NetBSD disklabel on Disk1/wd1 that will enable a RAID 
                    430: volume to exist upon it. On sparc64, you will need to simply 
                    431: [disklabel(8)](http://netbsd.gw.com/cgi-bin/man-cgi?disklabel+8+NetBSD-5.0.1+i386) 
                    432: the second disk which will write the proper Sun Disk Label.
                    433: 
                    434: *Tip*: 
                    435: [disklabel(8)](http://netbsd.gw.com/cgi-bin/man-cgi?disklabel+8+NetBSD-5.0.1+i386) 
                    436: will use your shell' s environment variable `$EDITOR` variable to edit the 
                    437: disklabel. The default is 
                    438: [vi(1)](http://netbsd.gw.com/cgi-bin/man-cgi?vi+1+NetBSD-5.0.1+i386)
                    439: 
                    440: On i386:
                    441: 
                    442:     # fdisk -0ua /dev/rwd1d
                    443:     fdisk: primary partition table invalid, no magic in sector 0
                    444:     Disk: /dev/rwd1d
                    445:     NetBSD disklabel disk geometry:
                    446:     cylinders: 19386, heads: 16, sectors/track: 63 (1008 sectors/cylinder)
                    447:     total sectors: 19541088
                    448:     
                    449:     BIOS disk geometry:
                    450:     cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
                    451:     total sectors: 19541088
                    452:     
                    453:     Do you want to change our idea of what BIOS thinks? [n]
                    454:     
                    455:     Partition 0:
                    456:     <UNUSED>
                    457:     The data for partition 0 is:
                    458:     <UNUSED>
                    459:     sysid: [0..255 default: 169]
                    460:     start: [0..1216cyl default: 63, 0cyl, 0MB]
                    461:     size: [0..1216cyl default: 19541025, 1216cyl, 9542MB]
                    462:     bootmenu: []
                    463:     Do you want to change the active partition? [n] y
                    464:     Choosing 4 will make no partition active.
                    465:     active partition: [0..4 default: 0] 0
                    466:     Are you happy with this choice? [n] y
                    467:     
                    468:     We haven't written the MBR back to disk yet.  This is your last chance.
                    469:     Partition table:
                    470:     0: NetBSD (sysid 169)
                    471:         start 63, size 19541025 (9542 MB, Cyls 0-1216/96/1), Active
                    472:     1: <UNUSED>
                    473:     2: <UNUSED>
                    474:     3: <UNUSED>
                    475:     Bootselector disabled.
                    476:     Should we write new partition table? [n] y
                    477:     
                    478:     # disklabel -r -e -I wd1
                    479:     type: unknown
                    480:     disk: Disk1
                    481:     label:
                    482:     flags:
                    483:     bytes/sector: 512
                    484:     sectors/track: 63
                    485:     tracks/cylinder: 16
                    486:     sectors/cylinder: 1008
                    487:     cylinders: 19386
                    488:     total sectors: 19541088
                    489:     [...snip...]
                    490:     16 partitions:
                    491:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    492:      a:  19541025        63       RAID                     # (Cyl.      0*-19385)
                    493:      c:  19541025        63     unused      0     0        # (Cyl.      0*-19385)
                    494:      d:  19541088         0     unused      0     0        # (Cyl.      0 -19385)
                    495: 
                    496: On sparc64:
                    497: 
                    498:     # disklabel -r -e -I wd1
                    499:     type: unknown
                    500:     disk: Disk1
                    501:     label:
                    502:     flags:
                    503:     bytes/sector: 512
                    504:     sectors/track: 63
                    505:     tracks/cylinder: 16
                    506:     sectors/cylinder: 1008
                    507:     cylinders: 19386
                    508:     total sectors: 19541088
                    509:     [...snip...]
                    510:     3 partitions:
                    511:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    512:      a:  19541088         0       RAID                     # (Cyl.      0 -  19385)
                    513:      c:  19541088         0     unused      0     0        # (Cyl.      0 -  19385)
                    514:     
                    515:     # sunlabel /dev/rwd1c 
                    516:     sunlabel> P
                    517:     a: start cyl =      0, size = 19541088 (19386/0/0 - 9541.55Mb)
                    518:     c: start cyl =      0, size = 19541088 (19386/0/0 - 9541.55Mb)
                    519: 
                    520: *Note*: On i386, the `c:` and `d:` slices are reserved. `c:` represents the 
                    521: NetBSD portion of the disk. `d:` represents the entire disk. Because we want to 
                    522: allocate the entire NetBSD MBR partition to RAID, and because `a:` resides 
                    523: within the bounds of `c:`, the `a:` and `c:` slices have same size and offset 
                    524: values and sizes. The offset must start at a track boundary (an increment of 
                    525: sectors matching the sectors/track value in the disk label). On sparc64 however, 
                    526: `c:` represents the entire NetBSD partition in the Sun disk label and `d:` is 
                    527: not reserved. Also note that sparc64's `c:` and `a:` require no offset from the 
                    528: beginning of the disk, however if they should need to be, the offset must start 
                    529: at a cylinder boundary (an increment of sectors matching the sectors/cylinder 
                    530: value).
                    531: 
                    532: ### Initializing the RAID Device
                    533: 
                    534: Next we create the configuration file for the RAID set / volume. Traditionally, 
                    535: RAIDframe configuration files belong in `/etc` and would be read and initialized 
                    536: at boot time, however, because we are creating a bootable RAID volume, the 
                    537: configuration data will actually be written into the RAID volume using the 
                    538: *auto-configure* feature. Therefore, files are needed only during the initial 
                    539: setup and should not reside in `/etc`.
                    540: 
                    541:     # vi /var/tmp/raid0.conf
                    542:     START array
                    543:     1 2 0
                    544:     
                    545:     START disks
                    546:     absent
                    547:     /dev/wd1a
                    548:     
                    549:     START layout
                    550:     128 1 1 1
                    551:     
                    552:     START queue
                    553:     fifo 100
                    554: 
                    555: Note that `absent` means a non-existing disk. This will allow us to establish 
                    556: the RAID volume with a bogus component that we will substitute for Disk0/wd0 at 
                    557: a later time.
                    558: 
                    559: Next we configure the RAID device and initialize the serial number to something 
                    560: unique. In this example we use a "YYYYMMDD*`Revision`*" scheme. The format you 
                    561: choose is entirely at your discretion, however the scheme you choose should 
                    562: ensure that no two RAID sets use the same serial number at the same time.
                    563: 
                    564: After that we initialize the RAID set for the first time, safely ignoring the 
                    565: errors regarding the bogus component.
                    566: 
                    567:     # raidctl -v -C /var/tmp/raid0.conf raid0
                    568:     Ignoring missing component at column 0
                    569:     raid0: Component absent being configured at col: 0
                    570:              Column: 0 Num Columns: 0
                    571:              Version: 0 Serial Number: 0 Mod Counter: 0
                    572:              Clean: No Status: 0
                    573:     Number of columns do not match for: absent
                    574:     absent is not clean!
                    575:     raid0: Component /dev/wd1a being configured at col: 1
                    576:              Column: 0 Num Columns: 0
                    577:              Version: 0 Serial Number: 0 Mod Counter: 0
                    578:              Clean: No Status: 0
                    579:     Column out of alignment for: /dev/wd1a
                    580:     Number of columns do not match for: /dev/wd1a
                    581:     /dev/wd1a is not clean!
                    582:     raid0: There were fatal errors
                    583:     raid0: Fatal errors being ignored.
                    584:     raid0: RAID Level 1
                    585:     raid0: Components: component0[**FAILED**] /dev/wd1a
                    586:     raid0: Total Sectors: 19540864 (9541 MB)
                    587:     # raidctl -v -I 2009122601 raid0
                    588:     # raidctl -v -i raid0
                    589:     Initiating re-write of parity
                    590:     raid0: Error re-writing parity!
                    591:     Parity Re-write status:
                    592:     
                    593:     # tail -1 /var/log/messages
                    594:     Dec 26 00:00:30  /netbsd: raid0: Error re-writing parity!
                    595:     # raidctl -v -s raid0
                    596:     Components:
                    597:               component0: failed
                    598:                /dev/wd1a: optimal
                    599:     No spares.
                    600:     component0 status is: failed.  Skipping label.
                    601:     Component label for /dev/wd1a:
                    602:        Row: 0, Column: 1, Num Rows: 1, Num Columns: 2
                    603:        Version: 2, Serial Number: 2009122601, Mod Counter: 7
                    604:        Clean: No, Status: 0
                    605:        sectPerSU: 128, SUsPerPU: 1, SUsPerRU: 1
                    606:        Queue size: 100, blocksize: 512, numBlocks: 19540864
                    607:        RAID Level: 1
                    608:        Autoconfig: No
                    609:        Root partition: No
                    610:        Last configured as: raid0
                    611:     Parity status: DIRTY
                    612:     Reconstruction is 100% complete.
                    613:     Parity Re-write is 100% complete.
                    614:     Copyback is 100% complete.
                    615: 
                    616: ### Setting up Filesystems
                    617: 
                    618: *Caution*: The root filesystem must begin at sector 0 of the RAID device. If 
                    619: not, the primary boot loader will be unable to find the secondary boot loader.
                    620: 
                    621: The RAID device is now configured and available. The RAID device is a pseudo 
                    622: disk-device. It will be created with a default disk label. You must now 
                    623: determine the proper sizes for disklabel slices for your production environment. 
                    624: For purposes of simplification in this example, our system will have 8.5 
                    625: gigabytes dedicated to `/` as `/dev/raid0a` and the rest allocated to `swap` 
                    626: as `/dev/raid0b`.
                    627: 
                    628: *Caution*: This is an unrealistic disk layout for a production server; the 
                    629: NetBSD Guide can expand on proper partitioning technique. See [Installing 
                    630: NetBSD: Preliminary considerations and preparations*](inst).
                    631: 
                    632: *Note*: Note that 1 GB is 2\*1024\*1024=2097152 blocks (1 block is 512 bytes, or 
                    633: 0.5 kilobytes). Despite what the underlying hardware composing a RAID set is, 
                    634: the RAID pseudo disk will always have 512 bytes/sector.
                    635: 
                    636: *Note*: In our example, the space allocated to the underlying `a:` slice 
                    637: composing the RAID set differed between i386 and sparc64, therefore the total 
                    638: sectors of the RAID volumes differs:
                    639: 
                    640: On i386:
                    641: 
                    642:      # disklabel -r -e -I raid0
                    643:     type: RAID
                    644:     disk: raid
                    645:     label: fictitious
                    646:     flags:
                    647:     bytes/sector: 512
                    648:     sectors/track: 128
                    649:     tracks/cylinder: 8
                    650:     sectors/cylinder: 1024
                    651:     cylinders: 19082
                    652:     total sectors: 19540864
                    653:     rpm: 3600
                    654:     interleave: 1
                    655:     trackskew: 0
                    656:     cylinderskew: 0
                    657:     headswitch: 0 # microseconds
                    658:     track-to-track seek: 0 # microseconds
                    659:     drivedata: 0
                    660:     
                    661:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    662:      a:  19015680         0     4.2BSD      0     0     0  # (Cyl.      0 - 18569)
                    663:      b:    525184  19015680       swap                     # (Cyl.  18570 - 19082*)
                    664:      d:  19540864         0     unused      0     0        # (Cyl.      0 - 19082*)
                    665: 
                    666: On sparc64:
                    667: 
                    668:     # disklabel -r -e -I raid0
                    669:     [...snip...]
                    670:     total sectors: 19539968
                    671:     [...snip...]
                    672:     3 partitions:
                    673:     #        size    offset     fstype [fsize bsize cpg/sgs]
                    674:      a:  19251200         0     4.2BSD      0     0     0  # (Cyl.      0 -  18799)
                    675:      b:    288768  19251200       swap                     # (Cyl.  18800 -  19081)
                    676:      c:  19539968         0     unused      0     0        # (Cyl.      0 -  19081)
                    677: 
                    678: Next, format the newly created `/` partition as a 4.2BSD FFSv1 File System:
                    679: 
                    680:     # newfs -O 1 /dev/rraid0a
                    681:     /dev/rraid0a: 9285.0MB (19015680 sectors) block size 16384, fragment size 2048
                    682:             using 51 cylinder groups of 182.06MB, 11652 blks, 23040 inodes.
                    683:     super-block backups (for fsck -b #) at:
                    684:     32, 372896, 745760, 1118624, 1491488, 1864352, 2237216, 2610080, 2982944,
                    685:     ...............................................................................
                    686:     
                    687:     # fsck -fy /dev/rraid0a
                    688:     ** /dev/rraid0a
                    689:     ** File system is already clean
                    690:     ** Last Mounted on
                    691:     ** Phase 1 - Check Blocks and Sizes
                    692:     ** Phase 2 - Check Pathnames
                    693:     ** Phase 3 - Check Connectivity
                    694:     ** Phase 4 - Check Reference Counts
                    695:     ** Phase 5 - Check Cyl groups
                    696:     1 files, 1 used, 4679654 free (14 frags, 584955 blocks, 0.0% fragmentation)
                    697: 
                    698: ### Migrating System to RAID
                    699: 
                    700: The new RAID filesystems are now ready for use. We mount them under `/mnt` and 
                    701: copy all files from the old system. This can be done using 
                    702: [dump(8)](http://netbsd.gw.com/cgi-bin/man-cgi?dump+8+NetBSD-5.0.1+i386) or 
                    703: [pax(1)](http://netbsd.gw.com/cgi-bin/man-cgi?pax+1+NetBSD-5.0.1+i386).
                    704: 
                    705:     # mount /dev/raid0a /mnt
                    706:     # df -h /mnt
                    707:     Filesystem        Size       Used      Avail %Cap Mounted on
                    708:     /dev/raid0a       8.9G       2.0K       8.5G   0% /mnt
                    709:     # cd /; pax -v -X -rw -pe . /mnt
                    710:     [...snip...]
                    711: 
                    712: The NetBSD install now exists on the RAID filesystem. We need to fix the 
                    713: mount-points in the new copy of `/etc/fstab` or the system will not come up 
                    714: correctly. Replace instances of `wd0` with `raid0`.
                    715: 
                    716: The swap should be unconfigured upon shutdown to avoid parity errors on the RAID 
                    717: device. This can be done with a simple, one-line setting in `/etc/rc.conf`.
                    718: 
                    719:     # vi /mnt/etc/rc.conf
                    720:     swapoff=YES
                    721: 
                    722: Next, the boot loader must be installed on Disk1/wd1. Failure to install the 
                    723: loader on Disk1/wd1 will render the system un-bootable if Disk0/wd0 fails. You
                    724: should hope your system won't have to reboot when wd0 fails.
                    725: 
                    726: *Tip*: Because the BIOS/CMOS menus in many i386 based systems are misleading 
                    727: with regard to device boot order. I highly recommend utilizing the `-o 
                    728: timeout=X` option supported by the i386 1st stage boot loader. Setup unique 
                    729: values for each disk as a point of reference so that you can easily determine 
                    730: from which disk the system is booting.
                    731: 
                    732: *Caution*: Although it may seem logical to install the 1st stage boot block into 
                    733: `/dev/rwd1{c,d}` (which is historically correct with NetBSD 1.6.x 
                    734: [installboot(8)](http://netbsd.gw.com/cgi-bin/man-cgi?installboot+8+NetBSD-5.0.1+i386) 
                    735: , this is no longer the case. If you make this mistake, the boot sector will 
                    736: become irrecoverably damaged and you will need to start the process over again.
                    737: 
                    738: On i386, install the boot loader into `/dev/rwd1a`:
                    739: 
                    740:     # /usr/sbin/installboot -o timeout=30 -v /dev/rwd1a /usr/mdec/bootxx_ffsv1
                    741:     File system:         /dev/rwd1a
                    742:     Primary bootstrap:   /usr/mdec/bootxx_ffsv1
                    743:     Ignoring PBR with invalid magic in sector 0 of `/dev/rwd1a'
                    744:     Boot options:        timeout 30, flags 0, speed 9600, ioaddr 0, console pc
                    745: 
                    746: On sparc64, install the boot loader into `/dev/rwd1a` as well, however the `-o` 
                    747: flag is unsupported (and un-needed thanks to OpenBoot):
                    748: 
                    749:     # /usr/sbin/installboot -v /dev/rwd1a /usr/mdec/bootblk
                    750:     File system:         /dev/rwd1a
                    751:     Primary bootstrap:   /usr/mdec/bootblk
                    752:     Bootstrap start sector: 1
                    753:     Bootstrap byte count:   5140
                    754:     Writing bootstrap
                    755: 
                    756: Finally the RAID set must be made auto-configurable and the system should be 
                    757: rebooted. After the reboot everything is mounted from the RAID devices.
                    758: 
                    759:     # raidctl -v -A root raid0
                    760:     raid0: Autoconfigure: Yes
                    761:     raid0: Root: Yes
                    762:     # tail -2 /var/log/messages
                    763:     raid0: New autoconfig value is: 1
                    764:     raid0: New rootpartition value is: 1
                    765:     # raidctl -v -s raid0
                    766:     [...snip...]
                    767:        Autoconfig: Yes
                    768:        Root partition: Yes
                    769:        Last configured as: raid0
                    770:     [...snip...]
                    771:     # shutdown -r now
                    772: 
                    773: ### Warning
                    774: 
                    775: Always use 
                    776: [shutdown(8)](http://netbsd.gw.com/cgi-bin/man-cgi?shutdown+8+NetBSD-5.0.1+i386) 
                    777: when shutting down. Never simply use 
                    778: [reboot(8)](http://netbsd.gw.com/cgi-bin/man-cgi?reboot+8+NetBSD-5.0.1+i386). 
                    779: [reboot(8)](http://netbsd.gw.com/cgi-bin/man-cgi?reboot+8+NetBSD-5.0.1+i386) 
                    780: will not properly run shutdown RC scripts and will not safely disable swap. This 
                    781: will cause dirty parity at every reboot.
                    782: 
                    783: ### The first boot with RAID
                    784: 
                    785: At this point, temporarily configure your system to boot Disk1/wd1. See notes in 
                    786: [[Testing Boot Blocks|guide/rf#adding-text-boot]] for details on this process. 
                    787: The system should boot now and all filesystems should be on the RAID devices. 
                    788: The RAID will be functional with a single component, however the set is not 
                    789: fully functional because the bogus drive (wd9) has failed.
                    790: 
                    791:     # egrep -i "raid|root" /var/run/dmesg.boot
                    792:     raid0: RAID Level 1
                    793:     raid0: Components: component0[**FAILED**] /dev/wd1a
                    794:     raid0: Total Sectors: 19540864 (9541 MB)
                    795:     boot device: raid0
                    796:     root on raid0a dumps on raid0b
                    797:     root file system type: ffs
                    798:     
                    799:     # df -h
                    800:     Filesystem    Size     Used     Avail Capacity  Mounted on
                    801:     /dev/raid0a   8.9G     196M      8.3G     2%    /
                    802:     kernfs        1.0K     1.0K        0B   100%    /kern
                    803:     
                    804:     # swapctl -l
                    805:     Device      1K-blocks     Used    Avail Capacity  Priority
                    806:     /dev/raid0b    262592        0   262592     0%    0
                    807:     # raidctl -s raid0
                    808:     Components:
                    809:               component0: failed
                    810:                /dev/wd1a: optimal
                    811:     No spares.
                    812:     component0 status is: failed.  Skipping label.
                    813:     Component label for /dev/wd1a:
                    814:        Row: 0, Column: 1, Num Rows: 1, Num Columns: 2
                    815:        Version: 2, Serial Number: 2009122601, Mod Counter: 65
                    816:        Clean: No, Status: 0
                    817:        sectPerSU: 128, SUsPerPU: 1, SUsPerRU: 1
                    818:        Queue size: 100, blocksize: 512, numBlocks: 19540864
                    819:        RAID Level: 1
                    820:        Autoconfig: Yes
                    821:        Root partition: Yes
                    822:        Last configured as: raid0
                    823:     Parity status: DIRTY
                    824:     Reconstruction is 100% complete.
                    825:     Parity Re-write is 100% complete.
                    826:     Copyback is 100% complete.
                    827: 
                    828: ### Adding Disk0/wd0 to RAID
                    829: 
                    830: We will now add Disk0/wd0 as a component of the RAID. This will destroy the 
                    831: original file system structure. On i386, the MBR disklabel will be unaffected 
                    832: (remember we copied wd0's label to wd1 anyway) , therefore there is no need to 
                    833: "zero" Disk0/wd0. However, we need to relabel Disk0/wd0 to have an identical 
                    834: NetBSD disklabel layout as Disk1/wd1. Then we add Disk0/wd0 as "hot-spare" to 
                    835: the RAID set and initiate the parity reconstruction for all RAID devices, 
                    836: effectively bringing Disk0/wd0 into the RAID-1 set and "syncing up" both disks.
                    837: 
                    838:     # disklabel -r wd1 > /tmp/disklabel.wd1
                    839:     # disklabel -R -r wd0 /tmp/disklabel.wd1
                    840: 
                    841: As a last-minute sanity check, you might want to use 
                    842: [diff(1)](http://netbsd.gw.com/cgi-bin/man-cgi?diff+1+NetBSD-5.0.1+i386) to 
                    843: ensure that the disklabels of Disk0/wd0 match Disk1/wd1. You should also backup 
                    844: these files for reference in the event of an emergency.
                    845: 
                    846:     # disklabel -r wd0 > /tmp/disklabel.wd0
                    847:     # disklabel -r wd1 > /tmp/disklabel.wd1
                    848:     # diff /tmp/disklabel.wd0 /tmp/disklabel.wd1
                    849:     # fdisk /dev/rwd0 > /tmp/fdisk.wd0
                    850:     # fdisk /dev/rwd1 > /tmp/fdisk.wd1
                    851:     # diff /tmp/fdisk.wd0 /tmp/fdisk.wd1
                    852:     # mkdir /root/RFbackup
                    853:     # cp -p /tmp/{disklabel,fdisk}* /root/RFbackup
                    854: 
                    855: Once you are sure, add Disk0/wd0 as a spare component, and start reconstruction:
                    856: 
                    857:     # raidctl -v -a /dev/wd0a raid0
                    858:     /netbsd: Warning: truncating spare disk /dev/wd0a to 241254528 blocks
                    859:     # raidctl -v -s raid0
                    860:     Components:
                    861:               component0: failed
                    862:                /dev/wd1a: optimal
                    863:     Spares:
                    864:                /dev/wd0a: spare
                    865:     [...snip...]
                    866:     # raidctl -F component0 raid0
                    867:     RECON: initiating reconstruction on col 0 -> spare at col 2
                    868:      11% |****                                   | ETA:    04:26 \
                    869: 
                    870: Depending on the speed of your hardware, the reconstruction time will vary. You 
                    871: may wish to watch it on another terminal (note that you can interrupt
                    872: `raidctl -S` any time without stopping the synchronisation):
                    873: 
                    874:     # raidctl -S raid0
                    875:     Reconstruction is 0% complete.
                    876:     Parity Re-write is 100% complete.
                    877:     Copyback is 100% complete.
                    878:     Reconstruction status:
                    879:       17% |******                                 | ETA: 03:08 -
                    880: 
                    881: After reconstruction, both disks should be *optimal*.
                    882: 
                    883:     # tail -f /var/log/messages
                    884:     raid0: Reconstruction of disk at col 0 completed
                    885:     raid0: Recon time was 1290.625033 seconds, accumulated XOR time was 0 us (0.000000)
                    886:     raid0:  (start time 1093407069 sec 145393 usec, end time 1093408359 sec 770426 usec)
                    887:     raid0: Total head-sep stall count was 0
                    888:     raid0: 305318 recon event waits, 1 recon delays
                    889:     raid0: 1093407069060000 max exec ticks
                    890:     
                    891:     # raidctl -v -s raid0
                    892:     Components:
                    893:                component0: spared
                    894:                /dev/wd1a: optimal
                    895:     Spares:
                    896:          /dev/wd0a: used_spare
                    897:          [...snip...]
                    898: 
                    899: When the reconstruction is finished we need to install the boot loader on the 
                    900: Disk0/wd0. On i386, install the boot loader into `/dev/rwd0a`:
                    901: 
                    902:     # /usr/sbin/installboot -o timeout=15 -v /dev/rwd0a /usr/mdec/bootxx_ffsv1
                    903:     File system:         /dev/rwd0a
                    904:     Primary bootstrap:   /usr/mdec/bootxx_ffsv1
                    905:     Boot options:        timeout 15, flags 0, speed 9600, ioaddr 0, console pc
                    906: 
                    907: On sparc64:
                    908: 
                    909:     # /usr/sbin/installboot -v /dev/rwd0a /usr/mdec/bootblk
                    910:     File system:         /dev/rwd0a
                    911:     Primary bootstrap:   /usr/mdec/bootblk
                    912:     Bootstrap start sector: 1
                    913:     Bootstrap byte count:   5140
                    914:     Writing bootstrap
                    915: 
                    916: And finally, reboot the machine one last time before proceeding. This is 
                    917: required to migrate Disk0/wd0 from status "used\_spare" as "Component0" to state 
                    918: "optimal". Refer to notes in the next section regarding verification of clean 
                    919: parity after each reboot.
                    920: 
                    921:     # shutdown -r now
                    922: 
                    923: ### Testing Boot Blocks
                    924: 
                    925: At this point, you need to ensure that your system's hardware can properly boot 
                    926: using the boot blocks on either disk. On i386, this is a hardware-dependent 
                    927: process that may be done via your motherboard CMOS/BIOS menu or your controller 
                    928: card's configuration menu.
                    929: 
                    930: On i386, use the menu system on your machine to set the boot device order / 
                    931: priority to Disk1/wd1 before Disk0/wd0. The examples here depict a generic Award 
                    932: BIOS.
                    933: 
1.2     ! jdf       934: ![Award BIOS i386 Boot Disk1/wd1](guide/images/rf-awardbios2.png)
1.1       jdf       935: **Award BIOS i386 Boot Disk1/wd1**
                    936: 
                    937: Save changes and exit:
                    938: 
                    939:     >> NetBSD/i386 BIOS Boot, Revision 5.2 (from NetBSD 5.0.2)
                    940:     >> (builds@b7, Sun Feb 7 00:30:50 UTC 2010)
                    941:     >> Memory: 639/130048 k
                    942:     Press return to boot now, any other key for boot menu
                    943:     booting hd0a:netbsd - starting in 30
                    944: 
                    945: You can determine that the BIOS is reading Disk1/wd1 because the timeout of the 
                    946: boot loader is 30 seconds instead of 15. After the reboot, re-enter the BIOS and 
                    947: configure the drive boot order back to the default:
                    948: 
1.2     ! jdf       949: ![Award BIOS i386 Boot Disk0/wd0](guide/images/rf-awardbios1.png)
1.1       jdf       950: **Award BIOS i386 Boot Disk0/wd0**
                    951: 
                    952: Save changes and exit:
                    953: 
                    954:     >> NetBSD/i386 BIOS Boot, Revision 5.2 (from NetBSD 5.0.2)
                    955:     >> Memory: 639/130048 k
                    956:     Press return to boot now, any other key for boot menu
                    957:     booting hd0a:netbsd - starting in 15
                    958: 
                    959: Notice how your custom kernel detects controller/bus/drive assignments 
                    960: independent of what the BIOS assigns as the boot disk. This is the expected 
                    961: behavior.
                    962: 
                    963: On sparc64, use the Sun OpenBoot **devalias** to confirm that both disks are bootable:
                    964: 
                    965:     Sun Ultra 5/10 UPA/PCI (UltraSPARC-IIi 400MHz), No Keyboard
                    966:     OpenBoot 3.15, 128 MB memory installed, Serial #nnnnnnnn.
                    967:     Ethernet address 8:0:20:a5:d1:3b, Host ID: nnnnnnnn.
                    968:     
                    969:     ok devalias
                    970:     [...snip...]
                    971:     cdrom /pci@1f,0/pci@1,1/ide@3/cdrom@2,0:f
                    972:     disk /pci@1f,0/pci@1,1/ide@3/disk@0,0
                    973:     disk3 /pci@1f,0/pci@1,1/ide@3/disk@3,0
                    974:     disk2 /pci@1f,0/pci@1,1/ide@3/disk@2,0
                    975:     disk1 /pci@1f,0/pci@1,1/ide@3/disk@1,0
                    976:     disk0 /pci@1f,0/pci@1,1/ide@3/disk@0,0
                    977:     [...snip...]
                    978:     
                    979:     ok boot disk0 netbsd
                    980:     Initializing Memory [...]
                    981:     Boot device /pci/pci/ide@3/disk@0,0 File and args: netbsd
                    982:     NetBSD IEEE 1275 Bootblock
                    983:     >> NetBSD/sparc64 OpenFirmware Boot, Revision 1.13
                    984:     >> (builds@b7.netbsd.org, Wed Jul 29 23:43:42 UTC 2009)
                    985:     loadfile: reading header
                    986:     elf64_exec: Booting [...]
                    987:     symbols @ [....]
                    988:      Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
                    989:          2006, 2007, 2008, 2009
                    990:          The NetBSD Foundation, Inc.  All rights reserved.
                    991:      Copyright (c) 1982, 1986, 1989, 1991, 1993
                    992:          The Regents of the University of California.  All rights reserved.
                    993:     [...snip...]
                    994: 
                    995: And the second disk:
                    996: 
                    997:     ok boot disk2 netbsd
                    998:     Initializing Memory [...]
                    999:     Boot device /pci/pci/ide@3/disk@2,0: File and args:netbsd
                   1000:     NetBSD IEEE 1275 Bootblock
                   1001:     >> NetBSD/sparc64 OpenFirmware Boot, Revision 1.13
                   1002:     >> (builds@b7.netbsd.org, Wed Jul 29 23:43:42 UTC 2009)
                   1003:     loadfile: reading header
                   1004:     elf64_exec: Booting [...]
                   1005:     symbols @ [....]
                   1006:      Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
                   1007:          2006, 2007, 2008, 2009
                   1008:          The NetBSD Foundation, Inc.  All rights reserved.
                   1009:      Copyright (c) 1982, 1986, 1989, 1991, 1993
                   1010:          The Regents of the University of California.  All rights reserved.
                   1011:     [...snip...]
                   1012: 
                   1013: At each boot, the following should appear in the NetBSD kernel 
                   1014: [dmesg(8)](http://netbsd.gw.com/cgi-bin/man-cgi?dmesg+8+NetBSD-5.0.1+i386) :
                   1015: 
                   1016:     Kernelized RAIDframe activated
                   1017:     raid0: RAID Level 1
                   1018:     raid0: Components: /dev/wd0a /dev/wd1a
                   1019:     raid0: Total Sectors: 19540864 (9541 MB)
                   1020:     boot device: raid0
                   1021:     root on raid0a dumps on raid0b
                   1022:     root file system type: ffs
                   1023: 
                   1024: Once you are certain that both disks are bootable, verify the RAID parity is 
                   1025: clean after each reboot:
                   1026: 
                   1027:     # raidctl -v -s raid0
                   1028:     Components:
                   1029:               /dev/wd0a: optimal
                   1030:               /dev/wd1a: optimal
                   1031:     No spares.
                   1032:     [...snip...]
                   1033:     Component label for /dev/wd0a:
                   1034:        Row: 0, Column: 0, Num Rows: 1, Num Columns: 2
                   1035:        Version: 2, Serial Number: 2009122601, Mod Counter: 67
                   1036:        Clean: No, Status: 0
                   1037:        sectPerSU: 128, SUsPerPU: 1, SUsPerRU: 1
                   1038:        Queue size: 100, blocksize: 512, numBlocks: 19540864
                   1039:        RAID Level: 1
                   1040:        Autoconfig: Yes
                   1041:        Root partition: Yes
                   1042:        Last configured as: raid0
                   1043:     Component label for /dev/wd1a:
                   1044:        Row: 0, Column: 1, Num Rows: 1, Num Columns: 2
                   1045:        Version: 2, Serial Number: 2009122601, Mod Counter: 67
                   1046:        Clean: No, Status: 0
                   1047:        sectPerSU: 128, SUsPerPU: 1, SUsPerRU: 1
                   1048:        Queue size: 100, blocksize: 512, numBlocks: 19540864
                   1049:        RAID Level: 1
                   1050:        Autoconfig: Yes
                   1051:        Root partition: Yes
                   1052:        Last configured as: raid0
                   1053:     Parity status: clean
                   1054:     Reconstruction is 100% complete.
                   1055:     Parity Re-write is 100% complete.
                   1056:     Copyback is 100% complete.
                   1057: 

CVSweb for NetBSD wikisrc <wikimaster@NetBSD.org> software: FreeBSD-CVSweb