josetann Posted April 3, 2008 Share Posted April 3, 2008 For anyone reading this for the first time, yes it's working! The how-to is at http://www.thetechguide.com/howto/unraid-on-hard-drive.html . Now back to your regularly scheduled post: I figured it'd make things easier to just have a full Slackware install that runs unRAID. I'm having a bit of a problem though. When trying to access the built-in webserver, it's giving me this message in Firefox: A user name and password are being requested by http://192.168.1.182. The site says: "'unRAID" The message in Internet Explorer is saying this: The server 192.168.1.182 at 'unRAID SMU' requires a username and password. I'm trying this using a virtual server in VMWare, hard disks are configured as ide. This happens whether I have a usb flash drive installed or not. Also happens whether the usb drive is mounted at /boot or elsewhere (thought maybe emhttp was requiring the usb drive to be mounted at /boot instead of elsewhere, that was a no go). I know that emhttp sees the usb drive, it recognizes my license key. Here's the syslog messages from /var/log/messages: Apr 3 21:38:22 Slackware emhttp: unRAID System Management Utility version 4.3-beta5 Apr 3 21:38:22 Slackware emhttp: Copyright © 2005-2008, Lime Technology, LLC Apr 3 21:38:22 Slackware emhttp: Pro key detected, registered to: ******* Apr 3 21:38:22 Slackware emhttp: Device inventory: Apr 3 21:38:22 Slackware emhttp: pci-0000:00:07.1-ide-0:0 (hda) ata-VMware_Virtual_IDE_Hard_Drive_00000000000000000001 Apr 3 21:38:22 Slackware emhttp: shcmd (1): rmmod md-mod >>/var/log/go 2>&1 Apr 3 21:38:22 Slackware kernel: md: unRAID driver removed. Apr 3 21:38:22 Slackware emhttp: shcmd (2): modprobe md-mod super=/boot/config/super.dat slots=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 >>/var/log/go 2>&1 Apr 3 21:38:23 Slackware kernel: md: unRAID driver 0.94.0 installed Apr 3 21:38:23 Slackware emhttp: shcmd (3): cp /var/spool/cron/crontabs/root- /var/spool/cron/crontabs/root Apr 3 21:38:23 Slackware emhttp: shcmd (4): echo '# Generated mover schedule:' >>/var/spool/cron/crontabs/root Apr 3 21:38:23 Slackware emhttp: shcmd (5): echo '40 3 * * * /usr/local/sbin/mover | logger 2>&1' >>/var/spool/cron/crontabs/root Apr 3 21:38:23 Slackware emhttp: shcmd (6): crontab /var/spool/cron/crontabs/root Apr 3 21:38:23 Slackware emhttp: shcmd (7): killall -w smbd nmbd Apr 3 21:38:25 Slackware emhttp: shcmd (: /usr/sbin/nmbd -D Apr 3 21:38:25 Slackware emhttp: shcmd (9): /usr/sbin/smbd -D Apr 3 21:38:36 Slackware emhttp: unRAID System Management Utility version 4.3-beta5 Apr 3 21:38:36 Slackware emhttp: Copyright © 2005-2008, Lime Technology, LLC Apr 3 21:38:36 Slackware emhttp: Pro key detected, registered to: ************** Apr 3 21:38:36 Slackware emhttp: Device inventory: Apr 3 21:38:36 Slackware emhttp: pci-0000:00:07.1-ide-0:0 (hda) ata-VMware_Virtual_IDE_Hard_Drive_00000000000000000001 Apr 3 21:38:36 Slackware emhttp: shcmd (1): rmmod md-mod >>/var/log/go 2>&1 Apr 3 21:38:36 Slackware kernel: md: unRAID driver removed. Apr 3 21:38:36 Slackware emhttp: shcmd (2): modprobe md-mod super=/boot/config/super.dat slots=0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 >>/var/log/go 2>&1 Apr 3 21:38:36 Slackware kernel: md: unRAID driver 0.94.0 installed Apr 3 21:38:36 Slackware emhttp: shcmd (3): cp /var/spool/cron/crontabs/root- /var/spool/cron/crontabs/root Apr 3 21:38:36 Slackware emhttp: shcmd (4): echo '# Generated mover schedule:' >>/var/spool/cron/crontabs/root Apr 3 21:38:37 Slackware emhttp: shcmd (5): echo '40 3 * * * /usr/local/sbin/mover | logger 2>&1' >>/var/spool/cron/crontabs/root Apr 3 21:38:37 Slackware emhttp: shcmd (6): crontab /var/spool/cron/crontabs/root Apr 3 21:38:37 Slackware emhttp: shcmd (7): killall -w smbd nmbd Apr 3 21:38:39 Slackware emhttp: shcmd (: /usr/sbin/nmbd -D Apr 3 21:38:39 Slackware emhttp: shcmd (9): /usr/sbin/smbd -D In case it helps, this is what I get without the usb key inserted: Apr 3 21:36:50 Slackware emhttp: unRAID System Management Utility version 4.3-beta5 Apr 3 21:36:50 Slackware emhttp: Copyright © 2005-2008, Lime Technology, LLC Apr 3 21:36:50 Slackware emhttp: Unregistered Apr 3 21:36:50 Slackware emhttp: Device inventory: Apr 3 21:36:50 Slackware emhttp: pci-0000:00:07.1-ide-0:0 (hda) ata-VMware_Virtual_IDE_Hard_Drive_00000000000000000001 Apr 3 21:36:50 Slackware emhttp: shcmd (1): rmmod md-mod >>/var/log/go 2>&1 Apr 3 21:36:50 Slackware emhttp: shcmd (2): modprobe md-mod super=/boot/config/super.dat slots=0,0,0,0,0,0 >>/var/log/go 2>&1 Apr 3 21:36:50 Slackware kernel: md: unRAID driver 0.94.0 installed Apr 3 21:36:50 Slackware emhttp: shcmd (3): killall -w smbd nmbd Apr 3 21:36:50 Slackware emhttp: shcmd (4): /usr/sbin/nmbd -D Apr 3 21:36:50 Slackware emhttp: shcmd (5): /usr/sbin/smbd -D In /var/log/syslog, I'm getting some stuff too: Apr 3 21:38:22 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-0:0:0:0: No such file or directory Apr 3 21:38:22 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-1:0:0:0: No such file or directory Apr 3 21:38:22 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-2:0:0:0: No such file or directory Apr 3 21:38:23 Slackware kernel: md: xor using function: pIII_sse (5026.000 MB/sec) Apr 3 21:38:23 Slackware kernel: md0: removed Apr 3 21:38:23 Slackware kernel: md1: missing Apr 3 21:38:23 Slackware kernel: md2: missing Apr 3 21:38:23 Slackware kernel: md3: missing Apr 3 21:38:23 Slackware kernel: md0: removed Apr 3 21:38:23 Slackware kernel: md1: missing Apr 3 21:38:23 Slackware kernel: md2: missing Apr 3 21:38:23 Slackware kernel: md3: missing Apr 3 21:38:25 Slackware emhttp: main: can't bind listener socket: Address already in use Apr 3 21:38:36 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-0:0:0:0: No such file or directory Apr 3 21:38:36 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-1:0:0:0: No such file or directory Apr 3 21:38:36 Slackware emhttp: restart_md_driver: stat pci-0000:00:1f.2-scsi-2:0:0:0: No such file or directory Apr 3 21:38:36 Slackware kernel: md: xor using function: pIII_sse (5096.000 MB/sec) Apr 3 21:38:36 Slackware kernel: md0: removed Apr 3 21:38:36 Slackware kernel: md1: missing Apr 3 21:38:36 Slackware kernel: md2: missing Apr 3 21:38:36 Slackware kernel: md3: missing Apr 3 21:38:37 Slackware kernel: md0: removed Apr 3 21:38:37 Slackware kernel: md1: missing Apr 3 21:38:37 Slackware kernel: md2: missing Apr 3 21:38:37 Slackware kernel: md3: missing I'm pretty sure some of these messages are just because I'm running in a different environment than that usb drive was configured for. I'm also pretty sure that it should just work without a usb key in Basic mode, but it's not. It's still requiring a username and password. Here's the same syslog messages without the usb key inserted: Apr 3 21:36:36 Slackware kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through Apr 3 21:36:36 Slackware kernel: sd 0:0:0:0: [sda] Assuming drive cache: write through Apr 3 21:36:50 Slackware emhttp: shcmd: shcmd (1): exit status: 1 Apr 3 21:36:50 Slackware kernel: md: xor using function: p5_mmx (11267.600 MB/sec) Apr 3 21:36:50 Slackware kernel: read_file: error 2 opening /boot/config/super.dat Apr 3 21:36:50 Slackware kernel: md: could not read superblock from /boot/config/super.dat Apr 3 21:36:50 Slackware kernel: md: warning! initializing superblock Apr 3 21:36:50 Slackware emhttp: shcmd: shcmd (3): exit status: 1 Has anyone else tried this and if so, how did you get past this point? Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 Ok, I figured it out. unRAID doesn't like shadow passwords. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 New problem. unRAID doesn't like smp support in the kernel. I have VMWare setup to use two cpus, I recompiled the kernel to enable smp support (and also to have the various IDE and SATA drivers built-in instead of being modular, eliminating the requirement for an initrd). I can start emhttp just fine, was able to assign two drives, started the array, formatted the disks, etc. But when I went to stop the array, I had problems. I was able to reproduce this, after rebooting I could start emhttp, it'd start the array just fine, but when I stopped it'd choke. Recompiling the kernel with the EXACT same config file, minus smp support, does not have this issue. Here's the syslog entry: Apr 3 23:31:00 Slackware kernel: ------------[ cut here ]------------ Apr 3 23:31:00 Slackware kernel: kernel BUG at drivers/md/unraid.c:278! Apr 3 23:31:00 Slackware kernel: invalid opcode: 0000 [#1] SMP Apr 3 23:31:00 Slackware kernel: Modules linked in: md_mod pcnet32 mii Apr 3 23:31:00 Slackware kernel: Apr 3 23:31:00 Slackware kernel: Pid: 1918, comm: emhttp Not tainted (2.6.24.4-unRAID #2) Apr 3 23:31:00 Slackware kernel: EIP: 0060:[<e09ced03>] EFLAGS: 00210202 CPU: 0 Apr 3 23:31:00 Slackware kernel: EIP is at get_free_stripe+0x12/0x177 [md_mod] Apr 3 23:31:00 Slackware kernel: EAX: df0faf01 EBX: df0faf80 ECX: c1406334 EDX: c048a254 Apr 3 23:31:00 Slackware kernel: ESI: df0faf80 EDI: dc4e888c EBP: dc4e8000 ESP: dcbf1d40 Apr 3 23:31:00 Slackware kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Apr 3 23:31:00 Slackware kernel: Process emhttp (pid: 1918, ti=dcbf0000 task=df8bc540 task.ti=dcbf0000) Apr 3 23:31:00 Slackware kernel: Stack: c03408bb c0119410 c1409080 df833a40 00000001 df833a40 00200286 c011aff5 Apr 3 23:31:00 Slackware kernel: df0faf80 c048a250 c048a24c df0faf80 df0faf80 dc4e888c dc4e8000 e09cefc2 Apr 3 23:31:00 Slackware kernel: 00100100 00200200 df833a40 dc4e8000 df0faf80 dc4e8000 dc4e888c dc4e8000 Apr 3 23:31:00 Slackware kernel: Call Trace: Apr 3 23:31:00 Slackware kernel: [<c03408bb>] schedule_timeout+0x13/0x8d Apr 3 23:31:00 Slackware kernel: [<c0119410>] enqueue_task+0xa/0x14 Apr 3 23:31:00 Slackware kernel: [<c011aff5>] try_to_wake_up+0x22d/0x237 Apr 3 23:31:00 Slackware kernel: [<e09cefc2>] shrink_stripes+0x72/0x93 [md_mod] Apr 3 23:31:00 Slackware kernel: [<e09d0143>] unraid_stop+0x19/0x32 [md_mod] Apr 3 23:31:00 Slackware kernel: [<e09caad1>] do_stop+0x8f/0xb4 [md_mod] Apr 3 23:31:00 Slackware kernel: [<e09cb941>] stop_array+0x82/0xa5 [md_mod] Apr 3 23:31:00 Slackware kernel: [<e09cc37b>] md_cmd_proc_write+0x9ef/0xb7c [md_mod] Apr 3 23:31:00 Slackware kernel: [<c0209905>] number+0x159/0x22f Apr 3 23:31:00 Slackware kernel: [<c016dde4>] mntput_no_expire+0x11/0x56 Apr 3 23:31:00 Slackware kernel: [<c0122760>] do_wait+0x946/0x9dc Apr 3 23:31:00 Slackware kernel: [<c015b1b5>] get_unused_fd_flags+0x53/0xc5 Apr 3 23:31:00 Slackware kernel: [<c016468f>] do_path_lookup+0x15a/0x175 Apr 3 23:31:00 Slackware kernel: [<c015d69c>] get_empty_filp+0x8a/0x11b Apr 3 23:31:00 Slackware kernel: [<c016260b>] permission+0xaf/0xc6 Apr 3 23:31:00 Slackware kernel: [<c0163542>] may_open+0xa6/0x188 Apr 3 23:31:00 Slackware kernel: [<c01841cd>] pde_users_dec+0xb/0x28 Apr 3 23:31:00 Slackware kernel: [<c018468b>] proc_reg_open+0x43/0x4a Apr 3 23:31:00 Slackware kernel: [<c0184648>] proc_reg_open+0x0/0x4a Apr 3 23:31:00 Slackware kernel: [<c015b30c>] __dentry_open+0xde/0x16b Apr 3 23:31:00 Slackware kernel: [<c015b413>] nameidata_to_filp+0x24/0x33 Apr 3 23:31:00 Slackware kernel: [<c015b454>] do_filp_open+0x32/0x39 Apr 3 23:31:00 Slackware kernel: [<e09cb98c>] md_cmd_proc_write+0x0/0xb7c [md_mod] Apr 3 23:31:00 Slackware kernel: [<c0187a17>] proc_file_write+0x2b/0x34 Apr 3 23:31:00 Slackware kernel: [<c01879ec>] proc_file_write+0x0/0x34 Apr 3 23:31:00 Slackware kernel: [<c01844ec>] proc_reg_write+0x5c/0x6f Apr 3 23:31:00 Slackware kernel: [<c0184490>] proc_reg_write+0x0/0x6f Apr 3 23:31:00 Slackware kernel: [<c015cd68>] vfs_write+0x8a/0x10c Apr 3 23:31:00 Slackware kernel: [<c015d2d3>] sys_write+0x41/0x67 Apr 3 23:31:00 Slackware kernel: [<c0103dbe>] syscall_call+0x7/0xb Apr 3 23:31:00 Slackware kernel: ======================= Apr 3 23:31:00 Slackware kernel: Code: 0f b6 c0 89 44 24 04 c7 04 24 d6 1b 9d e0 e8 a1 1a 75 df 31 c0 5a 59 5b c3 55 57 56 53 83 ec 2c 89 44 24 20 8a 40 3c 84 c0 7e 04 <0f> 0b eb fe 8b 4c 24 20 8b 44 24 20 8b 59 20 83 c0 20 c7 44 24 Apr 3 23:31:00 Slackware kernel: EIP: [<e09ced03>] get_free_stripe+0x12/0x177 [md_mod] SS:ESP 0068:dcbf1d40 Apr 3 23:31:00 Slackware kernel: ---[ end trace cabe9def06455936 ]--- This may need to be posted elsewhere, but since I'm doing nonstandard stuff (running in a VMWare server) with a nonstandard kernel (with smp support enabled), figured I'd just post it here. Quote Link to comment
bubbaQ Posted April 4, 2008 Share Posted April 4, 2008 SMP is (currently) a known no-no with unRAID. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 SMP is (currently) a known no-no with unRAID. Why exactly? Has anyone else tried it and had problems too? I searched (terms like smp and dual-core), I don't see any mention of problems. I think just not many people are trying it. It'd be useful so I could run unRAID on my server (which I'm doing now, still booting off the flash drive though), then run Windows under VMWare. VMWare is going to eat a lot of resources, so I figure I can help out by 1) Give it a raw partition instead of a virtual disk, 2) Throw another 2GB of RAM in the server, and 3) give VMWare its own cpu (or rather, one of the cores). Quote Link to comment
WeeboTech Posted April 4, 2008 Share Posted April 4, 2008 I never saw it posted that it's a no-no.. It was posted as being "untested" So here's our first test. the md driver is not smp safe I guess. More to come in the future. Quote Link to comment
bubbaQ Posted April 4, 2008 Share Posted April 4, 2008 unRAID + SMP = core dump.... definitely in unraid.c, and probably in the rewritten md.c I don't believe SMP would increase typical unRAID performance much since the bottleneck is disk and lan. Check CPU utilization... you'll find it is rarely high. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 unRAID + SMP = core dump.... definitely in unraid.c, and probably in the rewritten md.c I don't believe SMP would increase typical unRAID performance much since the bottleneck is disk and lan. Check CPU utilization... you'll find it is rarely high. Yes, with a server that runs unRAID and ONLY unRAID, there's really no need for more than one cpu. I'm not arguing that. But what about running unRAID, plus asterisk, plus VMWare? One core can handle unRAID and asterisk just fine, but it might be a bottleneck if it has to handle unRAID, asterisk, and VMWare running Windows. I need asterisk to have all the cpu it wants (it doesn't want much, but it needs to get what little it wants), so I'd like to have everything under linux run on the first core (so basically, unRAID plus asterisk) and anything VMWare does put on the other core (so if it gets bogged down, won't affect our phone system or unRAID). Plus, I hate having two cores without being able to see them both working. Now, I don't really have much experience debugging kernel drivers, but if someone can point me in the right direction I'm willing to try. I've already gotten this far without knowing what I was doing, might as well keep going. Quote Link to comment
bubbaQ Posted April 4, 2008 Share Posted April 4, 2008 Create a single VM for unraid.... a separate vm for other apps. Quote Link to comment
WeeboTech Posted April 4, 2008 Share Posted April 4, 2008 Create a single VM for unraid... I thiink the down side of this is unRaid under VMWare will only see and power manage 4 IDE drives. I had a thought today about getting XEN running under unRaid. Might be easier to get going, but then again, if the unraid code is not SMP safe, there might be little point to it. I don't believe SMP would increase typical unRAID performance much since the bottleneck is disk and lan. Where it would help is in simultaneous multiple drive access. Quote Link to comment
jimwhite Posted April 4, 2008 Share Posted April 4, 2008 Funny this thread should appear today... I was just about to post a similar discussion. I also wanted to build up a "super server"/developement type system that would allow what you have mentioned and more.... thoughts of SMP and a Q6600 qhuad-core danced in my head seriously though, to start I wanted to ask how difficult it would be to set up a full slackware install and have the un-raid fs installed... I'm a Linux nooby but very eager to learn... also, what is the possibility/probability of being able to bring up the un-raid fs on another distro such as ubuntu or something else? Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 seriously though, to start I wanted to ask how difficult it would be to set up a full slackware install and have the un-raid fs installed... I'm a Linux nooby but very eager to learn... also, what is the possibility/probability of being able to bring up the un-raid fs on another distro such as ubuntu or something else? It's not that hard to get going on Slackware 12 once you know what to do (and if you don't, it'll take well over 12 hours of figuring out what went wrong THIS time). Once you get it to boot a kernel with the unRAID drivers compiled in, it's not too much trouble. Take out shadow passwords, copy over all files from the bzroot that don't already exist (I cheated and did a cp -ri * /, then just kept the Enter key pressed down, effectively saying no when it asked to overwrite a file, least I hope that's what it did). Then reboot and you're good to go. You'll need to have emhttp load up, right now I'm doing it manually, still testing stuff out. As far as putting it on another distro, that would require more work. You couldn't just blindly copy files over and hope it works (well you could, who knows it might actually work). I'd recommend just sticking with Slackware, since that's what unRAID is based off of. Once I get a little more time, I'll try to post a write-up on my site on how to do this. I'll have to do the Slackware install using VMWare and a usb drive, since I don't think the AB9 Pro supports an IDE CD-ROM drive (or at least won't boot from it). I'll do the install on my laptop, then see if I can pop the drive in the server. Should work. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 Create a single VM for unraid.... a separate vm for other apps. It has a four drive limit for IDE drives (I know I can actually use any drive type, but you're limited to four emulated IDE drives, whether they're RAW or not), and SCSI isn't working too well (I did compile in the Bus Logic driver, but unRAID still couldn't see the VMWare SCSI drive). Plus I'd rather have unRAID be the host OS, since it's the main part of the whole server. The VMWare Windows install would be secondary, so I'd like it to be the emulated machine. If VMWare didn't have that four drive limitation I'd consider it, but I'm already at four drives and about to add more. Quote Link to comment
WeeboTech Posted April 4, 2008 Share Posted April 4, 2008 > I'll try to post a write-up on my site on how to do this. Where is your site? (URL) Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 I had a thought today about getting XEN running under unRaid. Might be easier to get going, but then again, if the unraid code is not SMP safe, there might be little point to it. It's been a while since I fooled around with XEN, but if it worked, you might be able to have unRAID be the guest OS with access to only one cpu, and as many drives as you want. That'd be great if you needed multiple linux installs, but not so much if you want to run Windows. True, you could always just run VMWare in another XEN guest, or in the host OS, but that's still adding an extra layer of emulation. I like to keep things as streamlined as possible. I don't believe SMP would increase typical unRAID performance much since the bottleneck is disk and lan. Where it would help is in simultaneous multiple drive access. How much would SMP actually help that? Unless you're talking about writing, or reading when a data drive is down. It's an honest question. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 > I'll try to post a write-up on my site on how to do this. Where is your site? (URL) It's http://www.thetechguide.com. Don't worry, I'll post here if/when I do write it up. Hope to get a crack at it today, it's my last day to work until next week (wife works this weekend, and it's near impossible to get anything done while watching the little one). Quote Link to comment
WeeboTech Posted April 4, 2008 Share Posted April 4, 2008 How much would SMP actually help that? Unless you're talking about writing, or reading when a data drive is down. It's an honest question. It all depends on what you are doing and how much of it. For example, I'll have one machine ripping a DVD to an ISO, another ripping as an mpeg2 and a third VMware autogk'ing a stack of MP2's to MP4 for my Archos. I use different drives, so that's where the extra core can help. Now if I could port the vmware environment right onto the unRaid machine I would be able to re-use some horse power. It's not that there are not enough cycles in my machine, SMP helps when there are multiple processes waiting for something. Quote Link to comment
jimwhite Posted April 4, 2008 Share Posted April 4, 2008 Once I get a little more time, I'll try to post a write-up on my site on how to do this. THAT would be great !!! Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 Since I'm lazy I'll ask here. Is there a quick and easy way to do a diff across two different directories? It doesn't need to analyze bit-by-bit, perhaps just analyze dates and sizes. I'd like to eventually get a script to be able to check a newer bzroot with an older one, and only attempt to copy over newer files. Quote Link to comment
WeeboTech Posted April 4, 2008 Share Posted April 4, 2008 diff -r OLDDIR NEWDIR and dircmp (if you have it installed). I used to have a perl utility do it all, I cannot seem to locate it. I'll look deeper this weekend. EDIT: OK I found the basics of my tools, maybe this can get you started. First dir_mklist creates a list of all files down a tree in a format that you can use fo rcomparison. dir_match.pl then looks at the list (OLD) and compares it to another list (NEW) and creates a shell script to resolve differences. (This is what I needed at the time, you can use the match for what ever you want to do the base shell code is there). Other choices are to use dir_mklist.pl and sort then diff to find the deltas. Quote Link to comment
josetann Posted April 4, 2008 Author Share Posted April 4, 2008 Ok, I'm starting on the how-to. Not sure how long it'll take to get done, might be tonight, could be next week, could be never (a lot of my stuff is never fully completed, but still good enough for most). In case you want to try this out before I'm done, I'll let you see what I've got up so far, this should help you figure it out on your own. http://www.thetechguide.com/howto/unraid-on-hard-drive.html Currently, it's not more than a few sentences, but I'll keep updating it as time permits. Quote Link to comment
limetech Posted April 5, 2008 Share Posted April 5, 2008 Heh - you're getting close One suggestion - use grub instead of lilo. Quote Link to comment
josetann Posted April 5, 2008 Author Share Posted April 5, 2008 Well, I've gotten it to work using lilo, even though it gave me lots of trouble. If I get some spare time I might try grub on a virtual machine though. Quote Link to comment
WeeboTech Posted April 5, 2008 Share Posted April 5, 2008 > Convince WeeboTech that this whole endeavor only took ten minutes and each time a new unRAID version comes out, it'd only take Tom a couple minutes max to make this unRAID-max distro. LOL!!!!! Quote Link to comment
musicmann Posted April 5, 2008 Share Posted April 5, 2008 josetann, This is great work indeed! So this setup means you also have swap already and that changes (such as new installations) would survive a reboot, right? If so, this would probably be the easiest way to get any desired software installed and runnable by an end user....that is, it would be much easier to get VMware running this way than what I'm doing trying to turn it into a package. A couple of questions: Does the installation hard drive count against the unRAID hard drive limit? I'm assuming not because it's not in the array (A more general question to everybody) Can the device address for a hard drive change as you add new drives? For example, say I have multiple SATA drives and unRAID is currently installed on a drive that comes up as /dev/sda. Is there any chance that, that same physical card drive's /dev/sda address could change after I add more hard drives (say, with an additional SATA controller added)? Let me know if this question is not clear. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.