Software-RAID mini-HOWTO Linas Vepstas, linas@linas.org v0.38 28 December 1997 RAID stands for ''Redundant Array of Inexpensive Disks'', and is meant to be a way of creating a fast and reliable disk-drive subsystem out of individual disks. This document is a tutorial/HOWTO/FAQ for users of the Linux MD kernel extension, the associated tools, and their use. The MD extension implements RAID-0 (stripping), RAID-1 (mirroring), RAID-4 and RAID-5 in software. That is, with MD, no special hardware or disk controllers are required to get many of the benefits of RAID. This document is NOT an introduction to RAID; you must find this else­ where. Preamble This document is GPL'ed by Linas Vepstas (linas@linas.org). Permission to use, copy, distribute this document for any purpose is hereby granted, provided that the author's / editor's name and this notice appear in all copies and/or supporting documents; and that an unmodified version of this document is made freely available. This document is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY, either expressed or implied. While every effort has been taken to ensure the accuracy of the information documented herein, the author / editor / maintainer assumes NO RESPONSIBILITY for any errors, or for any damages, direct or consequential, as a result of the use of the information documented herein. RAID, although designed to improve system reliability by adding redundancy, can also lead to a false sense of security and confidence when used improperly. This false confidence can lead to even greater disasters. In particular, note that RAID is designed to protect against *disk* failures, and not against *power* failures. A power failure can damage data on the disks in such a way that it is not recoverable! RAID is *not* a substitute for proper backup of your system. Know what you are doing, test, be knowledgeable and aware! 1. Introduction 1. Q: What is RAID? A: RAID stands for "Redundant Array of Inexpensive Disks", and is meant to be a way of creating a fast and reliable disk-drive subsystem out of individual disks. 2. Q: What is this document? A: This document is a tutorial/HOWTO/FAQ for users of the Linux MD kernel extension, the associated tools, and their use. The MD extension implements RAID-0 (stripping), RAID-1 (mirroring), RAID-4 and RAID-5 in software. That is, with MD, no special hardware or disk controllers are required to get many of the benefits of RAID. This document is NOT an introduction to RAID; you must find this elsewhere. 3. Q: What levels of RAID does the Linux kernel implement? A: Striping (RAID-0) and linear concatenation are a part of the stock 2.x series of kernels. This code is of production quality; it is well understood and well maintained. It is being used in some very large USENET news servers. RAID-1, RAID-4 & RAID-5 are not present in the stock kernel; a separate patch needs to be applied to get this functional­ ity. The current snapshots should be considered beta qual­ ity; that is, there are no known bugs but there are some rough edges and untested system setups. RAID-1 hot reconstruction has been recently introduced (August 1997) and should be considered alpha quality. RAID-5 hot reconstruction will be alpha quality any day now. 4. Q: Where do I get it? A: Software RAID-0 and linear mode are a stock part of all current Linux kernels. Patches for Software RAID-1,4,5 are available from . See also the quasi-mirror for patches, tools and other goodies. 5. Q: Are there other Linux RAID references? A: · Generic RAID overview: . · General Linux RAID options: . · Linux-RAID mailing list archive: . · Linux Software RAID Home Page: . · Linux Software RAID tools: . · How to setting up linear/stripped Software RAID: . · Bootable RAID mini-HOWTO: . · Linux RAID-Geschichten: . 6. Q: Who do I blame for this document? A: Linas Vepstas slapped this thing together. However, most of the information, and some of the words were supplied by · Bradley Ward Allen · Luca Berra · Brian Candler · Bohumil Chalupa · Anton Hristozov · Miguel de Icaza · Ingo Molnar · Alvin Oga · Gadi Oxman · Michael Robinton · Martin Schulze · Geoff Thompson · Edward Welbon · Rod Wilkens · Johan Wiltink · Leonard N. Zubkoff · Marc ZYNGIER Copyrights · Copyright (C) 1994-96 Marc ZYNGIER · Copyright (C) 1997 Gadi Oxman, Ingo Molnar, Miguel de Icaza · Copyright (C) 1997 Linas Vepstas · By copyright law, additional copyrights are implicitly held by the contributors listed above. Thanks all for being there! 2. Setup & Installation Considerations 1. Q: I must soon install Linux on new system, one requirement is to have RAID1. Now I'm wondering what is the easiest way to do it. A: I keep rediscovering that file-system planning is one of the more difficult Unix configuration tasks. To answer your question, I can describe what we did. We planned the following setup: · two EIDE disks, 2.1.gig each. disk partition mount pt. size device 1 1 / 300M /dev/hda1 1 2 swap 64M /dev/hda2 1 3 /home 800M /dev/hda3 1 4 /var 900M /dev/hda4 2 1 /root 300M /dev/hdc1 2 2 swap 64M /dev/hdc2 2 3 /home 800M /dev/hdc3 2 4 /var 900M /dev/hdc4 · each disk is on a separate controller (& ribbon cable). The theory is that a controller failure and/or ribbon failure won't disable both disks. Possibly get performance boost from parallel operations? · Install linux on / in /dev/hda1 this will allow booting and subsequent installation of raid patches, etc. · /dev/hdc1 will contain a ``cold'' copy of /dev/hda1. This is NOT a raid copy, just a copy-copy. It's there just in case disk1 fails completely; we can use a rescue disk, mark /dev/hdc1 as bootable, and use that to keep going, without having to reinstall the system. The theory here is that in case of severe failure, I can still boot the system without worrying about raid superblock-corruption or other raid failure modes & gotchas that I don't understand. · /dev/hda3 and /dev/hdc3 will be mirrors /dev/md0. · /dev/hda4 and /dev/hdc4 will be mirrors /dev/md1. · we picked /var and /home to be mirrored, and in separate partitions, under the following (convoluted ???) logic: · / will contain non-changing data --- for all practical purposes, it will be read-only without actually being read-only. · /home will contain slowly changing data --- an almost- read-only system. · /var will contain rapidly changing data, including mail spools, database contents and web server logs. The theory is that if for some bizarre reason, the operating system goes wild, corruption is limited to one partition. Thus, if for some unlikely, hypothetical reason, the database starts scribbling everywhere, it might clobber mail and log files, but not /home. I am not entirely satisfied with my logic & reasoning, but it was the best I could do on short notice. I would like to have some scheme that verifies that files in /usr and /home are not changed, e.g. some MD5 signature scheme that is run regularly. The idea is to detect hacker intrusion as well as corruption. Similarly, the database contents are quite valuable, and I don't have a fault- tolerant plan for that that will let me sleep well at night. So, to complete the answer to your question: · install redhat on disk 1, partition 1. do NOT mount any of the other partitions. · install raid per instructions. · configure md0 and md1. · convince yourself that you know what to do in case of a disk failure! Discover sysadmin mistakes now, and not during an actual crisis. Experiment! (we turned off power during disk activity --- this proved to be ugly but informative). · do some ugly mount/copy/unmount/rename/reboot scheme to move /var over to the /dev/md1. Done carefully, this is not dangerous. · enjoy! 2. Q: Can I strip/mirror the root partition (/)? Why can't I boot Linux directly from the md disks? A: Both Lilo and Loadlin need an non-stripped/mirrored par­ tition to read the kernel image from. If you want to strip/mirror the root partition (/), then create an unstriped/mirrored partition. Typically, this is /boot. Then you either use the initial ramdisk support, or some old patches that were posted a while back, to allow your root device to be striped. There are several approaches that can be used. One approach is documented in detail in the Bootable RAID mini-HOWTO: . Alternately, use mkinitrd to build the ramdisk image, see below. Edward Welbon writes: · To mount an md filesystem as root, the main thing is to build an initial file system image that has the needed modules and md tools to start md. I have a simple script that does this. · For boot media, I have a small cheap SCSI disk (170MB I got it used for $20). This disk runs on a AHA1452, but it could just as well be an inexpensive IDE disk on the native IDE. The disk need not be very fast since it is mainly for boot. · This disk has a small file system which contains the kernel and the file system image for initrd. The initial file system image has just enough stuff to allow me to load the raid SCSI device driver module and start the raid partition that will become root. I then do an echo 0x900 > /proc/sys/kernel/real-root-dev (0x900 is for /dev/md0) and exit linuxrc. The boot proceeds normally from there. · I have built most support as a module except for the AHA1452 driver that brings in the initrd filesystem. So I have a fairly small kernel. The method is perfectly reliable, I have been doing this since before 2.1.26 and have never had a problem that I could not easily recover from. The file systems even survived several 2.1.4[45] hard crashes with no real problems. · At one time I had partitioned the raid disks so that the initial cylinders of the first raid disk held the kernel and the initial cylinders of the second raid disk hold the initial file system image, instead I made the initial cylinders of the raid disks swap since they are the fastest cylinders (why waste them on boot?). · The nice thing about having an inexpensive device dedicated to boot is that it is easy to boot from and can also serve as a rescue disk if necessary. If you are interested, you can take a look at the script that builds my initial ram disk image and then runs lilo. It is current enough to show the picture. It isn't espe­ cially pretty and it could certainly build a much smaller filesystem image for the initial ram disk. It would be easy to a make it more efficient. But it uses lilo as is. If you make any improvements, please forward a copy to me. 8-) 3. Q: I have heard that I can run mirroring over striping. Is this true? Can I run mirroring over the loopback device? A: Yes, but not the reverse. That is, you can put a stripe over several disks, and then build a mirror on top of this. However, striping cannot be put on top of mirroring. A brief technical explanation is that the linear and stripe personalities use the ll_rw_blk routine for access. The ll_rw_blk routine maps disk devices and sectors, not blocks. Block devices can be layered one on top of the other; but devices that do raw, low-level disk accesses, such as ll_rw_blk, cannot. Currently (November 1997) RAID cannot be run over the loop­ back devices, although this should be fixed shortly. 4. Q: I have two small disks and three larger disks. Can I concatenate the two smaller disks with RAID-0, and then create a RAID-5 out of that and the larger disks? A: Currently (November 1997), for a RAID-5 array, no. Cur­ rently, one can do this only for a RAID-1 on top of the con­ catenated drives. 5. Q: What is the difference between RAID-1 and RAID-5 for a two-disk configuration (i.e. the difference between a RAID-1 array built out of two disks, and a RAID-5 array built out of two disks)? A: There is no difference in storage capacity. Nor can disks be added to either array to increase capacity (see the question below for details). RAID-1 offers a performance advantage for reads: the RAID-1 driver uses distributed-read technology to simultaneously read two sectors, one from each drive, thus doubling read performance. The RAID-5 driver, although it contains many optimizations, does not currently (September 1997) realize that the parity disk is actually a mirrored copy of the data disk. Thus, it serializes data reads. 6. Q: How can I guard against a two-disk failure? A: Some of the RAID algorithms do guard against multiple disk failures, but these are not currently implemented for Linux. However, the Linux Software RAID can guard against multiple disk failures by layering an array on top of an array. For example, nine disks can be used to create three raid-5 arrays. Then these three arrays can in turn be hooked together into a single RAID-5 array on top. In fact, this kind of a configuration will guard against a three-disk failure. Note that a large amount of disk space is ''wasted'' on the redundancy information. For an NxN raid-5 array, N=3, 5 out of 9 disks are used for parity (=55%percent;) N=4, 7 out of 16 disks N=5, 9 out of 25 disks ... N=9, 17 out of 81 disks (=~20%percent;) In general, an MxN arrary will use M+N-1 disks for parity. The least amount of space is "wasted" when M=N. Another alternative is to create a RAID-1 array with three disks. Note that since all three disks contain identical data, that 2/3's of the space is ''wasted''. 7. Q: I'd like to understand how it'd be possible to have something like fsck: if the partition hasn't been cleanly unmounted, fsck runs and fixes the filesystem by itself more than 90%percent; of the time. Since the machine is capable of fixing it by itself with ckraid --fix, why not make it automatic? A: Brian Candler responds: Then you just put ckraid into your system initialization scripts, like fsck is. After the root partition is mounted, add the following to /etc/rc.d/rc.sysinit: mdadd /dev/md0 /dev/hda1 /dev/hdc1 || { ckraid --fix /etc/raid.usr.conf mdadd /dev/md0 /dev/hda1 /dev/hdc1 } mdadd /dev/md1 /dev/hda2 /dev/hdc2 || { ckraid --fix /etc/raid.var.conf mdadd /dev/md0 /dev/hda2 /dev/hdc2 } (Modify the above to suit your system.) Gadi Oxman explains the operation: In an unclean shutdown, Linux might be in one of the following states: · The in-memory disk cache was in sync with the RAID set when the unclean shutdown occurred; no data was lost. · The in-memory disk cache was newer than the RAID set contents when the crash occurred; this results in a corrupted filesystem and potentially in data loss. This state can be further divided to the following two states: · Linux was writing data when the unclean shutdown occurred. · Linux was not writing data when the crash occurred. Suppose we were using a RAID-1 array. In (2a), it might happen that before the crash, a small number of data blocks were successfully written only to some of the mirrors, so that on the next reboot, the mirrors will no longer contain the same data. If we ignore the mirror differences, the 0.36.3 read- balancing code might choose to read the above data blocks from any of the mirrors, which will result in inconsistent behavior (for example, the output of e2fsck -n /dev/md0 can differ from run to run). Since RAID doesn't protect against unclean shutdowns, usually there isn't any ''obviously correct'' way to fix the mirror differences and the filesystem corruption. For example, by default ckraid --fix will choose the first operational mirror and update the other mirrors with its contents. However, depending on the exact timing at the crash, the data on another mirror might be more recent, and we might want to use it as the source mirror instead, or perhaps use another method for recovery. If you wish to run ckraid --fix automatically, you can check the return code of mdrun for errors. For example: mdrun -p1 /dev/md0 if [ $? -gt 0 ] ; then ckraid --fix /etc/raid1.conf mdrun -p1 /dev/md0 fi 3. Error Recovery 1. Q: I have a RAID-1 (mirroring) setup, and lost power while there was disk activity. Now what do I do? A: The redundancy of RAID levels is designed to protect against a disk failure, not against a power failure. There are several ways to recover from this situation. · Method (1): Use the raid tools. These can be used to sync the raid arrays. They do not fix file-system damage; after the raid arrays are sync'ed, then the file- system still has to be fixed with fsck. Raid arrays can be checked with ckraid /etc/raid1.conf (for RAID-1, else, /etc/raid5.conf, etc.) Calling ckraid /etc/raid1.conf --fix will pick one of the disks in the array (usually the first), and use that as the master copy, and copy its blocks to the others in the mirror. To designate which of the disks should be used as the master, you can use the --force-source flag: for example, ckraid /etc/raid1.conf --fix --force-source /dev/hdc3 The ckraid command can be safely run without the --fix option to verify the inactive RAID array without making any changes. When you are comfortable with the proposed changes, supply the --fix option. · Method (2): Paranoid, time-consuming, not much better than the first way. Lets assume a two-disk RAID-1 array, consisting of partitions /dev/hda3 and /dev/hdc3. You can try the following: a. fsck /dev/hda3 b. fsck /dev/hdc3 c. decide which of the two partitions had fewer errors, or were more easily recovered, or recovered the data that you wanted. Pick one, either one, to be your new ``master'' copy. Say you picked /dev/hdc3. d. dd if=/dev/hdc3 of=/dev/hda3 e. mkraid raid1.conf -f --only-superblock Instead of the last two steps, you can instead run ckraid /etc/raid1.conf --fix --force-source /dev/hdc3 which should be a bit faster. · Method (3): Lazy man's version of above. If you don't want to wait for long fsck's to complete, it is perfectly fine to skip the first three steps above, and move directly to the last two steps. Just be sure to run fsck /dev/md0 after you are done. Method (3) is actually just method (1) in disguise. In any case, the above steps will only sync up the raid arrays. The file system probably needs fixing as well: for this, fsck needs to be run on the active, unmounted md device. With a three-disk RAID-1 array, there are more possibilities, such as using two disks to ''vote'' a majority answer. Tools to automate this do not currently (September 97) exist. 2. Q: I have a RAID-4 or a RAID-5 (parity) setup, and lost power while there was disk activity. Now what do I do? A: The redundancy of RAID levels is designed to protect against a disk failure, not against a power failure. Since the disks in a RAID-4 or RAID-5 array do not contain a file system that fsck can read, there are fewer repair options. You cannot use fsck to do preliminary checking and/or repair; you must use ckraid first. The ckraid command can be safely run without the --fix option to verify the inactive RAID array without making any changes. When you are comfortable with the proposed changes, supply the --fix option. If you wish, you can try designating one of the disks as a ''failed disk''. Do this with the --suggest-failed-disk- mask flag. Only one bit should be set in the flag: RAID-5 cannot recover two failed disks. The mask is a binary bit mask: thus: 0x1 == first disk 0x2 == second disk 0x4 == third disk 0x8 == fourth disk, etc. Alternately, you can choose to modify the parity sectors, by using the --suggest-fix-parity flag. This will recompute the parity from the other sectors. The flags --suggest-failed-dsk-mask and --suggest-fix-parity can be safely used for verification. No changes are made if the --fix flag is not specified. Thus, you can experiment with different possible repair schemes. 3. Q: My RAID-1 device, /dev/md0 consists of two hard drive partitions: /dev/hda3 and /dev/hdc3. Recently, the disk with /dev/hdc3 failed, and was replaced with a new disk. My best friend, who doesn't understand RAID, said that the correct thing to do now is to ''dd if=/dev/hda3 of=/dev/hdc3''. I tried this, but things still don't work. A: You should keep your best friend away from you computer. Fortunately, no serious damage has been done. You can recover from this by running: mkraid raid1.conf -f --only-superblock By using dd, two identical copies of the partition were cre­ ated. This is almost correct, except that the RAID-1 kernel extension expects the RAID superblocks to be different. Thus, when you try to reactivate RAID, the software will notice the problem, and deactivate one of the two parti­ tions. By re-creating the superblock, you should have a fully usable system. 4. Q: My RAID-1 device, /dev/md0 consists of two hard drive partitions: /dev/hda3 and /dev/hdc3. My best (girl?)friend, who doesn't understand RAID, ran fsck on /dev/hda3 while I wasn't looking, and now the RAID won't work. What should I do? A: You should re-examine your concept of ``best friend''. In general, fsck should never be run on the individual par­ titions that compose a RAID array. Assuming that neither of the partitions are/were heavily damaged, no data loss has occurred, and the RAID-1 device can be recovered as follows: a. make a backup of the file system on /dev/hda3 b. dd if=/dev/hda3 of=/dev/hdc3 c. mkraid raid1.conf -f --only-superblock This should leave you with a working disk mirror. 5. Q: Why does the above work as a recovery procedure? A: Because each of the component partitions in a RAID-1 mir­ ror is a perfectly valid copy of the file system. In a pinch, mirroring can be disabled, and one of the partitions can be mounted and safely run as an ordinary, non-RAID file system. When you are ready to restart using RAID-1, then unmount the partition, and follow the above instructions to restore the mirror. Note that the above works ONLY for RAID-1, and not for any of the other levels. It may make you feel more comfortable to reverse the direc­ tion of the copy above: copy from the disk that was untouched to the one that was. Just be sure to fsck the final md. 6. Q: I am confused by the above questions, but am not yet bailing out. Is it safe to run fsck /dev/md0 ? A: Yes, it is safe to run fsck on the md devices. In fact, this is the only safe place to run fsck. 7. Q: If a disk is slowly failing, will it be obvious which one it is? I am concerned that it won't be, and this confusion could lead to some dangerous decisions by a sysadmin. A: Once a disk fails, an error code will be returned from the low level driver to the RAID driver. The RAID driver will mark it as ``bad'' in the RAID superblocks of the ``good'' disks (so we will later know which mirrors are good and which aren't), and continue RAID operation on the remaining operational mirrors. This, of course, assumes that the disk and the low level driver can detect a read/write error, and will not silently corrupt data, for example. This is true of current drives (error detection schemes are being used internally), and is the basis of RAID operation. 8. Q: What about hot-repair? A: Work is underway to complete ``hot reconstruction''. With this feature, one can add several ``spare'' disks to the RAID set (be it level 1 or 4/5), and once a disk fails, it will be reconstructed on one of the spare disks in run time, without ever needing to shut down the array. However, to use this feature, the spare disk must have been declared at boot time, or it must be hot-added, which requires the use of special cabinets and connectors that allow a disk to be added while the electrical power is on. As of October 97, there is a beta version of MD that allows: · RAID 1 and 5 reconstruction on spare drives · RAID-5 parity reconstruction after an unclean shutdown · spare disk to be hot-added to an already running RAID 1 or 4/5 array By default, automatic reconstruction is (Dec 97) currently disabled by default, due to the preliminary nature of this work. It can be enabled by changing the value of SUPPORT_RECONSTRUCTION in include/linux/md.h. If spare drives were configured into the array when it was created and kernel-based reconstruction is enabled, the spare drive will already contain a RAID superblock (written by mkraid), and the kernel will reconstruct its contents automatically (without needing the usual mdstop, replace drive, ckraid, mdrun steps). If you are not running automatic reconstruction, and have not configured a hot-spare disk, the procedure described by Gadi Oxman is recommended: · Currently, once the first disk is removed, the RAID set will be running in degraded mode. To restore full operation mode, you need to: · stop the array (mdstop /dev/md0) · replace the failed drive · run ckraid raid.conf to reconstruct its contents · run the array again (mdadd, mdrun). At this point, the array will be running with all the drives, and again protects against a failure of a single drive. Currently, it is not possible to assign single hot-spare disk to several arrays. Each array requires it's own hot-spare. 9. Q: I would like to have an audible alarm for ``you schmuck, one disk in the mirror is down'', so that the novice sysadmin knows that there is a problem. A: The kernel is logging the event with a ``KERN_ALERT'' priority in syslog. There are several software packages that will monitor the syslog files, and beep the PC speaker, call a pager, send e-mail, etc. automatically. 10. Q: How do I run RAID-5 in degraded mode (with one disk failed, and not yet replaced)? A: Gadi Oxman writes: · Normally, to run a RAID-5 set of n drives you have to: mdadd /dev/md0 /dev/disk1 ... /dev/disk(n) mdrun -p5 /dev/md0 Even if one of the disks has failed, you still have to mdadd it as you would in a normal setup. Then, · The array will be active in degraded mode with (n - 1) drives. If ``mdrun'' fails, the kernel has noticed an error (for example, several faulty drives, or an unclean shutdown). Use ``dmesg'' to display the kernel error messages from ``mdrun''. If the raid-5 set is corrupted due to a power loss, rather than a disk crash, one can try to recover by creating a new RAID superblock: mkraid -f --only-superblock raid5.conf A RAID array doesn't provide protection against a power failure or a kernel crash, and can't guarantee correct recovery. Rebuilding the superblock will simply cause the system to ignore the condition by marking all the drives as ``OK'', as if nothing happened. 11. Q: How does RAID-5 work when a disk fails? A: The typical operating scenario is as follows: · A RAID-5 array is active. · One drive fails while the array is active. · The drive firmware and the low-level Linux disk/controller drivers detect the failure and report an error code to the MD driver. · The MD driver continues to provide an error-free /dev/md0 device to the higher levels of the kernel (with a performance degradation) by using the remaining operational drives. · The sysadmin can umount /dev/md0 and mdstop /dev/md0 as usual. · If the failed drive is not replaced, the sysadmin can still start the array in degraded mode as usual, by running mdadd and mdrun. 12. Q: The QuickStart says that mdstop is just to make sure that the disks are sync'ed. Is this REALLY necessary? Isn't unmounting the file systems enough? A: The command mdstop /dev/md0 will: · mark it ''clean''. This allows us to detect unclean shutdowns, for example due to a power failure or a kernel crash. · sync the array. This is less important after unmounting a filesystem, but is important if the /dev/md0 is accessed directly rather than through a filesystem (for example, by e2fsck). 13. Q: A: 14. Q: Why is there no question 13? A: If you are concerned about RAID, High Availability, and UPS, then its probably a good idea to be superstitious as well. 4. Troubleshooting Install Problems 1. Q: What is the current best known-stable patch for RAID in the 2.0.x series kernels? A: As of 18 Sept 1997, it is "2.0.30 + pre-9 2.0.31 + Werner Fink's swapping patch + the alpha RAID patch". As of Novem­ ber 1997, it is 2.0.31 + ... !? 2. Q: The RAID patches will not install cleanly for me. What's wrong? A: Make sure that /usr/include/linux is a symbolic link to /usr/src/linux/include/linux. Make sure that the new files raid5.c, etc. have been copied to their correct locations. Sometimes the patch command will not create new files. Try the -f flag on patch. 3. Q: While compiling raidtools 0.42, compilation stops trying to include but it doesn't exist in my system. How do I fix this? A: raidtools-0.42 requires linuxthreads-0.6 from: Alternately, use glibc v2.0. 4. Q: I get the message: mdrun -a /dev/md0: Invalid argument A: Use mkraid to initialize the RAID set prior to the first use. mkraid ensures that the RAID array is initially in a consistent state by erasing the RAID partitions. In addi­ tion, mkraid will create the RAID superblocks. 5. Q: I get the message: mdrun -a /dev/md0: Invalid argument The setup was: · raid1 build as a kernel module · normal install procedure followed ... mdcreate, mdadd, etc. · cat /proc/mdstat shows Personalities : read_ahead not set md0 : inactive sda1 sdb1 6313482 blocks md1 : inactive md2 : inactive md3 : inactive · mdrun -a creates the error message /dev/md0: Invalid argument A: Try lsmod to see if the modules is loaded, and if not, load it with modprobe raid1. 6. Q: Truxton Fulton wrote: On my Linux 2.0.30 system, while doing a mkraid for a RAID-1 device, during the clearing of the two individual parti­ tions, I got "Cannot allocate free page" errors appearing on the console, and "Unable to handle kernel paging request at virtual address ..." errors in the system log. At this time, the system became quite unusable, but it appears to recover after a while. The operation appears to have com­ pleted with no other errors, and I am successfully using my RAID-1 device. The errors are disconcerting though. Any ideas? A: This was a well-known bug in the 2.0.30 kernels. It is fixed in the 2.0.31 kernel; alternately, fall back to 2.0.29. 7. Q: I'm not able to mdrun a RAID-1, RAID-4 or RAID-5 device. If I try to mdrun a mdadd'ed device I get the message ''invalid raid superblock magic''. A: Make sure that you've run the mkraid part of the install procedure. 8. Q: When I access /dev/md0, the kernel spits out a lot of errors like md0: device not running, giving up ! and I/O error.... I've successfully added my devices to the virtual device. A: To be usable, the device must be running. Use mdrun -px /dev/md0 where x is l for linear, 0 for RAID-0 or 1 for RAID-1, etc. 9. Q: I've created a linear md-dev with 2 devices. cat /proc/mdstat shows the total size of the device, but df only shows the size of the first physical device. A: You must mkfs your new md-dev before using it the first time, so that the filesystem will cover the whole device. 10. Q: I've set up /etc/mdtab using mdcreate, I've mdadd'ed, mdrun and fsck'ed my two /dev/mdX partitions. Everything looks okay before a reboot. As soon as I reboot, I get an fsck error on both partitions: fsck.ext2: Attempt to read block from filesystem resulted in short read while trying too open /dev/md0. Why?! How do I fix it?! A: During the boot process, the RAID partitions must be started before they can be fsck'ed. This must be done in one of the boot scripts. For some distributions, fsck is called from /etc/rc.d/rc.S, for others, it is called from /etc/rc.d/rc.sysinit. Change this file to mdadd -ar *before* fsck -A is executed. Better yet, it is suggested that ckraid be run if mdadd returns with an error. How do do this is discussed in greater detail in question 14 of the section ''Error Recovery''. 11. Q: I get the message invalid raid superblock magic while trying to run an array which consists of partitions which are bigger than 4GB. A: This bug is now fixed. (September 97) Make sure you have the latest raid code. 12. Q: I get the message Warning: could not write 8 blocks in inode table starting at 2097175 while trying to run mke2fs on a partition which is larger than 2GB. A: This seems to be a problem with mke2fs (November 97). A temporary work-around is to get the mke2fs code, and add #undef HAVE_LLSEEK to e2fsprogs-1.10/lib/ext2fs/llseek.c just before the first #ifdef HAVE_LLSEEK and recompile mke2fs. 13. Q: ckraid currently isn't able to read /etc/mdtab A: The RAID0/linear configuration file format used in /etc/mdtab is obsolete, although it will be supported for a while more. The current, up-to-date config files are cur­ rently named /etc/raid1.conf, etc. 14. Q: The personality modules (raid1.o) are not loaded automatically; they have to be manually modprobe'd before mdrun. How can this be fixed? A: To autoload the modules, we can add the following to /etc/conf.modules: alias md-personality-3 raid1 alias md-personality-4 raid5 15. Q: I've mdadd'ed 13 devices, and now I'm trying to mdrun -p5 /dev/md0 and get the message: /dev/md0: Invalid argument A: The default configuration for software RAID is 8 real devices. Edit linux/md.h, change #define MAX_REAL=8 to a larger number, and rebuild the kernel. 16. Q: I can't make md work with partitions on our latest SPARCstation 5. I suspect that this has something to do with disk-labels. A: Sun disk-labels sit in the first 1K of a partition. For RAID-1, the Sun disk-label is not an issue since ext2fs will skip the label on every mirror. For other raid levels (0, linear and 4/5), this appears to be a problem; it has not yet (Dec 97) been addressed. 5. Supported Hardware 1. Q: I have SCSI adapter brand XYZ (with or without several channels), and disk brand(s) PQR and LMN, will these work with md to create a linear/stripped/mirrored personality? A: Yes! Software RAID will work with any disk controller (IDE or SCSI) and any disks. The disks do not have to be identical, nor do the controllers. For example, a RAID mir­ ror can be created with one half the mirror being a SCSI disk, and the other an IDE disk. The disks do not even have to be the same size. There are no restrictions on the mixing & matching of disks and controllers. This is because Software RAID works with disk partitions, not with the raw disks themselves. The only recommendation is that for RAID levels 1 and 5, the disk partitions that are used as part of the same set be the same size. If the partitions used to make up the RAID 1 or 5 array are not the same size, then the excess space in the larger partitions is wasted (not used). 2. Q: I have a twin channel BT-952, and the box states that it supports hardware RAID 0, 1 and 0+1. I have made a RAID set with two drives, the card apparently recognizes them when it's doing it's BIOS startup routine. I've been reading in the driver source code, but found no reference to the hardware RAID support. Anybody out there working on that? A: The Mylex/BusLogic FlashPoint boards with RAIDPlus are actually software RAID, not hardware RAID at all. RAIDPlus is only supported on Windows 95 and Windows NT, not on Net­ ware or any of the Unix platforms. Aside from booting and configuration, the RAID support is actually in the OS drivers. While in theory Linux support for RAIDPlus is possible, the implementation of RAID-0/1/4/5 in the Linux kernel is much more flexible and should have superior performance, so there's little reason to support RAIDPlus directly. 6. Modifying an Existing Installation 1. Q: Are linear MD's expandable? Can a new hard-drive/partition be added, and the size of the existing file system expanded? A: Miguel de Icaza writes: I changed the ext2fs code to be aware of multiple-devices instead of the regular one device per file system assump­ tion. So, when you want to extend a file system, you run a utility program that makes the appropriate changes on the new device (your extra partition) and then you just tell the system to extend the fs using the specified device. You can extend a file system with new devices at system operation time, no need to bring the system down (and when­ ever I get some extra time, you will be able to remove devices from the ext2 volume set, again without even having to go to single-user mode or any hack like that). You can get the patch for 2.1.x kernel from my web page: 2. Q: Can I add disks to a RAID-5 array? A: Currently, (September 1997) no, not without erasing all data. A conversion utility to allow this does not yet exist. The problem is that the actual structure and layout of a RAID-5 array depends on the number of disks in the array. Of course, one can add drives by backing up the array to tape, deleting all data, creating a new array, and restoring from tape. 3. Q: What would happen to my RAID1/RAID0 sets if I shift one of the drives from being /dev/hdb to /dev/hdc? Because of cabling/case size/stupidity issues, I had to make my RAID sets on the same IDE controller (/dev/hda and /dev/hdb). Now that I've fixed some stuff, I want to move /dev/hdb to /dev/hdc. What would happen if I just change the /etc/mdtab and /etc/raid1.conf files to reflect the new location? A: For RAID-0/linear, one must be careful to specify the drives in exactly the same order. Thus, in the above exam­ ple, if the original config is mdadd /dev/md0 /dev/hda /dev/hdb Then the new config *must* be mdadd /dev/md0 /dev/hda /dev/hdc For RAID-1/4/5, the drive's ''RAID number'' is stored in its RAID superblock, and therefore the order in which the disks are specified is not important. RAID-0/linear does not have a superblock due to it's older design, and the desire to maintain backwards compatibility with this older design. 7. Performance, Tools & General Bone-headed Questions 1. Q: I've created a RAID-0 device on /dev/sda2 and /dev/sda3. The device is a lot slower than a single partition. Isn't md a pile of junk? A: To have a RAID-0 device running a full speed, you must have partitions from different disks. Besides, putting the two halves of the mirror on the same disk fails to give you any protection whatsoever against disk failure. 2. Q: I have 2 Brand X super-duper hard disks and a Brand Y controller. and am considering using md. Does it significantly increase the throughput? Is the performance really noticeable? A: The answer depends on the configuration that you use. Linux MD RAID-0 (striping) performance: Must wait for all disks to read/write the stripe. Linux MD RAID-1 (mirroring) read performance: MD implements read balancing. In a low-IO situation, this won't change performance. But, with two disks in a high-IO environment, this could as much as double the read performance. For N disks in the mirror, this could improve performance N-fold. Linux MD RAID-1 (mirroring) write performance: Must wait for the write to occur to all of the disks in the mirror. 3. Q: What is the optimal RAID-5 configuration for performance? A: Since RAID-5 attempts to equally distribute the I/O load across several drives, the best performance will be obtained when the RAID set is balanced by using identical drives, identical controllers, and the same (low) number of drives on each controller. Note, however, that using identical components might raise the probability of multiple drives failures. 4. Q: What is the optimal block size for a RAID-4/5 array? A: When using the current (November 1997) RAID-4/5 implemen­ tation, it is strongly recommended that the file system be created with mke2fs -b 4096 instead of the default 1024 byte filesystem block size. This is because the current RAID-5 implementation allocates one 4K memory page per disk block; thus 75% of the memory which RAID-5 is allocating for pending I/O is not being used. With a 4096 block size, it will potentially queue 4 times as much pending I/O to the low level drivers without allocating additional memory. Note: the 4K memory page size applies to the Intel x86 architecture. I think memory pages are 8K on Alpha/Sparc (????), and thus the above figures should be adjusted accordingly. Note: if your file system has a lot of small files (files less than 10KBytes in size), a considerable fraction of the disk space might be wasted. This is because disk space is allocated in multiples of the block size. Allocating large blocks for small files clearly results in a waste of disk space. Note: the above remarks do NOT apply to Software RAID-0/1/linear. Note: most ''typical'' systems do not have that many small files. That is, although there might be thousands of small files, this would lead to only some 10 to 100MB wasted space, which is probably an acceptable tradeoff for perfor­ mance on a multi-gigabyte disk. Note: for news servers, there might be tens or hundreds of thousands of small files. In such cases, the smaller block size may be more important than the improved performance. Note: there exists an experimental file system for Linux which packs small files and file chunks onto a single block. It apparently has some very positive performance implica­ tions when the average file size is much smaller than the block size. Note: Future versions may implement schemes that obsolete the above discussion. However, this is difficult to imple­ ment, since dynamic run-time allocation can lead to dead- locks; the current implementation performs a static pre- allocation. 5. Q: How does the chunk size influence the speed of my RAID device? A: The chunk size is the amount of data contiguous on the virtual device that is also contiguous on the physical device. Depending on your workload, the best is to let the chunk size match the size of your requests, so two requests have chances to be on different disks, and to be run the same time. This suppose a lot of testing with different chunk sizes to match the average request size, and to have the best performances. 6. Q: Where can I put the md commands in the startup scripts, so that everything will start automatically at boot time? A: Rod Wilkens writes: What I did is put ``mdadd -ar'' in the ``/etc/rc.d/rc.sysinit'' right after the kernel loads the modules, and before the ``fsck'' disk check. This way, you can put the ``/dev/md?'' device in the ``/etc/fstab''. Then I put the ``mdstop -a'' right after the ``umount -a'' unmounting the disks, in the ``/etc/rc.d/init.d/halt'' file. For raid-5, you will want to look at the return code for mdadd, and if it failed, do a ckraid --fix /etc/raid5.conf to repair any damage. 7. Q: I was wondering if it's possible to setup stripping with more than 2 devices in md0? This is for a news server, and I have 9 drives... Needless to say I need much more than two. Is this possible? A: Yes. (describe how to do this) 8. Q: When is Software RAID superior to Hardware RAID? A: Normally, Hardware RAID is considered superior to Soft­ ware RAID, because hardware controllers often have a large cache, and can do a better job of scheduling operations in parallel. However, integrated Software RAID can (and does) gain certain advantages from being close to the operating system. For example, ... ummm. Opaque description of caching of reconstructed blocks in buffer cache elided ... On a dual PPro SMP system, it has been reported that Soft­ ware-RAID performance exceeds the performance of a well- known hardware-RAID board vendor by a factor of 2 to 5. Software RAID is also a very interesting option for high- availability redundant server systems. In such a configura­ tion, two CPU's are attached to one set or SCSI disks. If one server crashes or fails to respond, then the other server can mdadd, mdrun and mount the software RAID array, and take over operations. This sort of dual-ended operation is not always possible with many hardware RAID controllers, because of the state configuration that the hardware con­ trollers maintain. 9. Q: If I upgrade my version of raidtools, will it have trouble manipulating older raid arrays? In short, should I recreate my RAID arrays when upgrading the raid utilities? A: No, not unless the major version number changes. An MD version x.y.z consists of three sub-versions: x: Major version. y: Minor version. z: Patchlevel version. Version x1.y1.z1 of the RAID driver supports a RAID array with version x2.y2.z2 in case (x1 == x2) and (y1 >= y2). Different patchlevel (z) versions for the same (x.y) version are designed to be mostly compatible. The minor version number is increased whenever the RAID array layout is changed in a way which is incompatible with older versions of the driver. New versions of the driver will maintain compatibility with older RAID arrays. The major version number will be increased if it will no longer make sense to support old RAID arrays in the new kernel code. For RAID-1, it's not likely that the disk layout nor the superblock structure will change anytime soon. Most all Any optimization and new features (reconstruction, multithreaded tools, hot-plug, etc.) doesn't affect the physical layout. 10. Q: The command mdstop /dev/md0 says that the device is busy. A: There's a process that has a file open on /dev/md0, or /dev/md0 is still mounted. Terminate the process or umount /dev/md0. 11. Q: Are there performance tools? A: There is also a new utility called iotrace in the linux/iotrace directory. It reads /proc/io-trace and analy­ ses/plots it's output. If you feel your system's block IO performance is too low, just look at the iotrace output. 12. Q: I was reading the RAID source, and saw the value SPEED_LIMIT defined as 1024K/sec. What does this mean? Does this limit performance? A: SPEED_LIMIT is used to limit RAID reconstruction speed during automatic reconstruction. Basically, automatic reconstruction allows you to e2fsck and mount immediately after an unclean shutdown, without first running ckraid. Automatic reconstruction is also used after a failed hard drive has been replaced. In order to avoid overwhelming the system while reconstruc­ tion is occurring, the reconstruction thread monitors the reconstruction speed and slows it down if its too fast. The 1M/sec limit was arbitrarily chosen as a reasonable rate which allows the reconstruction to finish reasonably rapidly, while creating only a light load on the system so that other processes are not interfered with. 13. Q: What about ''spindle synchronization'' or ''disk synchronization''? A: Spindle synchronization is used to keep multiple hard drives spinning at exactly the same speed, so that their disk platters are always perfectly aligned. This is used by some hardware controllers to better organize disk writes. However, for software RAID, this information is not used, and spindle synchronization might even hurt performance. 8. Questions Waiting for Answers 1. Q: What are the option you have used for formating the (raid) disks? I used: mke2fs -b 4096 -R stride=4 ... blah or is it supposed to be 64K × 4 drives: mke2fs -b 4096 -R stride=262000 ... blah are there any other options ? stride blocks are filesystem blocks, not virtual memory pages. Is there a paper somewhere about what does the stride option do ? Also, in relation with the ``md'' device driver ? The stride option serves only one purpose: it tells mke2fs how many file system blocks will be written to each member in turn. This allows mke2fs to allocate the block and inode bitmaps so that they don't all end up on the same physical drive. I noticed last spring that one drive in a pair always had a larger I/O count, and tracked it down to the these meta-data blocks. Ted added the -R stride= option in response to my explanation and request for a workaround. > For a 4KB block file system, with stripe size 32KB, one would use -R > stride=8. 2. Q: For testing the raw disk thru put... is there a character device for raw read/raw writes instead of /dev/sdaxx that we can use to measure performance on the raid drives?? is there a GUI based tool to use to watch the disk thru-put?? 9. Wish List of Enhancements to MD and Related Software Bradley Ward Allen wrote: Ideas include: · Bootup parameters to tell the kernel which devices are to be MD devices (no more ``mdadd'') · Making MD transparent to ``mount''/``umount'' such that there is no ``mdrun'' and ``mdstop'' · Integrating ``ckraid'' entirely into the kernel, and letting it run as needed (So far, all I've done is suggest getting rid of the tools and putting them into the kernel; that's how I feel about it, this is a filesystem, not a toy.) · Deal with arrays that can easily survive N disks going out simultaneously or at separate moments, where N is a whole number > 0 settable by the administrator · Handle kernel freezes, power outages, and other abrupt shutdowns better · Don't disable a whole disk if only parts of it have failed, e.g., if the sector errors are confined to less than 50% of access over the attempts of 20 dissimilar requests, then it continues just ignoring those sectors of that particular disk. · Bad sectors: · A mechanism for saving which sectors are bad, someplace onto the disk. · If there is a generalized mechanism for marking degraded bad blocks that upper filesystem levels can recognize, use that. Program it if not. · Perhaps alternatively a mechanism for telling the upper layer that the size of the disk got smaller, even arranging for the upper layer to move out stuff from the areas being eliminated. This would help with a degraded blocks as well. · Failing the above ideas, keeping a small (admin settable) amount of space aside for bad blocks (distributed evenly across disk?), and using them (nearby if possible) instead of the bad blocks when it does happen. Of course, this is inefficient. Furthermore, the kernel ought to log every time the RAID array starts each bad sector and what is being done about it with a ``crit'' level warning, just to get the administrator to realize that his disk has a piece of dust burrowing into it (or a head with platter sickness). · Software-switchable disks: ``disable this disk'' would block until kernel has completed making sure there is no data on the disk being shut down that is needed (e.g., to complete an XOR/ECC/other error correction), then release the disk from use (so it could be removed, etc.); ``enable this disk'' would mkraid a new disk if appropriate and then start using it for ECC/whatever operations, enlarging the RAID5 array as it goes; ``resize array'' would respecify the total number of disks and the number of redundant disks, and the result would often be to resize the size of the array; where no data loss would result, doing this as needed would be nice, but I have a hard time figuring out how it would do that; in any case, a mode where it would block (for possibly hours (kernel ought to log something every ten seconds if so)) would be necessary; ``enable this disk while saving data'' which would save the data on a disk as-is and move it to the RAID5 system as needed, so that a horrific save and restore would not have to happen every time someone brings up a RAID5 system (instead, it may be simpler to only save one partition instead of two, it might fit onto the first as a gzip'd file even); finally, ``re-enable disk'' would be an operator's hint to the OS to try out a previously failed disk (it would simply call disable then enable, I suppose).