itimpi Posted August 24, 2014 Share Posted August 24, 2014 I stopped the array, added the disk to the next disk slot, and in the FS column it says reiserfs (I did not start the array). So I backed off (removed the precleared disk from the array) to wait for instruction. What is the proper technique to add a precleared disk to the array with an XFS or BTRFS file system. On my system if I stop the array and then click on a disk name (e.g. disk1) then is a drop-down that allows you to select the file system type (if the array is not stopped then this list is locked to the current value). Changing it and then restarting the array brings up the disk as unformatted with the option to format it enabled. I assume that the same would happen if you are adding a new disk. Quote Link to comment
Joe L. Posted August 24, 2014 Share Posted August 24, 2014 If we upgrade from a 2TB to a 4TB disk, do we have the option to make the new 4TB disk XFS even if the old disk was ReiserFS If you do have that option to change file-system type, I expect you will then need to reformat that disk, and you will lose any/all files that are currently on it. There is no magical convert the current file-system (with its files) into a different file-system (retaining all the files) command. You've been warned. Quote Link to comment
SSD Posted August 24, 2014 Share Posted August 24, 2014 If we upgrade from a 2TB to a 4TB disk, do we have the option to make the new 4TB disk XFS even if the old disk was ReiserFS If you do have that option to change file-system type, I expect you will then need to reformat that disk, and you will lose any/all files that are currently on it. There is no magical convert the current file-system (with its files) into a different file-system (retaining all the files) command. You've been warned. Correct. To "convert" a disk to a different filesystem you'd have to start with a new disk, format it with the new filesystem, and copy all the files from the RFS disk to the new XFS or BTRFS disk. You could then change the FS on the RFS disk, reformat it, and then use it as to convert another disk. Etc, etc. There are no tools supplied to convert a filesystem in place. Quote Link to comment
SSD Posted August 24, 2014 Share Posted August 24, 2014 I saw those instructions but it reads only v1,2,3,4 This was very misleading. Yes, this was an oversight in the readme. It was noted in an earlier post in this thread. Betas almost always require just copying a few files. Understand it is a little confusing, but it is a brand new beta. If in doubt ask questions. If you are doing a flashdrive rebuild, and backup your existing config directory while the array is started, and then use it on your new flashdrive, a parity check will start. That is because there is a file in the config directory that tells unRaid whether the array was stopped properly, and since the backup was taken when the array was started, unRaid will assume you had a dirty shutdown and start the parity check. Quote Link to comment
Jaco2k Posted August 24, 2014 Share Posted August 24, 2014 I am slightly disappointed to see that the bug about the unclean shutdown when pressing the power button is still there... You promised it would be fixed on the next release but, apparently not. This seriously affects my usage right now - I run it headless and sometimes I am simply streaming to a XBMC machine on my living room where there is no web browser. Having to startup my desktop just to come and shutdown UnRaid via the web interface is far from good usage. So, when is then the fix coming for this "known" error? Given the current average release schedule, I hope it is not 2 months... --- Edit: Reviewed the original reply and I correct that it was not promised for next release. Nevertheless, this was working fine on the 5.x branches, so why not just transport the same shutdown method when pressing the power button? Mind you, I just tested on the 6.0 b7 and I updated, started the system, did NOT access any files and simply shutdown via power button. It still started a parity check on the next boot. I have absolutely no mods or plugins on my system - it is a totally default install that has been working well for a long time. In 'stock' unraid, the power switch event is tied to same code that webGui uses to shutdown, so is most cases it should work correctly. Depending on plugins and system activity it might not work - this is a known problem and will be fixed. In some ways it is a tough problem because forcing a shutdown can result in data loss. Consider what windows does: if it can't shut something down without possibly losing data, you get a nice dialog asking you what to do. Anyway, this is FYI - please let's not discuss this issue in this thread. A place to post bugs and features is coming soon, please everyone have just a little more patience. Quote Link to comment
jonp Posted August 24, 2014 Share Posted August 24, 2014 I tried to copy a 16G file from one disk share to another disk share using Samba. The destination was almost full. I was using Windows 8.1 File Explorer. The server was recently rebooted and all disks were spun up. The copy dialog sat there at 0% copied for a very long time - I'd say a minute or even longer, before giving an error (see image below - note the file name is masked). I had noticed, about 20 seconds or so waiting for the copy to start, that the file explorer window that was displaying the source disk of the file i was copying was having trouble refreshing (the path field started filling with green from left to right). I refreshed the WebGUI to see if the server was unresponsive and it responded quickly while this copy delay was occurring and the file explorer window continued to try to refresh. Nothing in the log. After I got the message I tried it again. This time there was a short delay maybe 5-10 seconds, and the copy began and completed without issue. It did ask if I wanted to overwrite the file (the file didn't exist on the disk before I initiated the copy the first time). This kind of behavior has been occurring going back to sometime during the 5.0 betas I think, and has happened on multiple different servers / motherboards, and using various flavors of Windows from XP to Windows 8.1. I have noticed that this problem tends to happen and get worse especially with bigger disks and when the destination disk becomes fuller. I do not have any problem coping files between disks - even to near full disks - when doing the operations on the server (cp/mv). Feedback like this should probably get moved into the defect reports thread. Quote Link to comment
SSD Posted August 24, 2014 Share Posted August 24, 2014 Feedback like this should probably get moved into the defect reports thread. Done - my post removed from this thread. Quote Link to comment
jonp Posted August 24, 2014 Share Posted August 24, 2014 Feedback like this should probably get moved into the defect reports thread. Done - my post removed from this thread. Thank you! Quote Link to comment
jonp Posted August 24, 2014 Share Posted August 24, 2014 I am slightly disappointed to see that the bug about the unclean shutdown when pressing the power button is still there... You promised it would be fixed on the next release but, apparently not. This seriously affects my usage right now - I run it headless and sometimes I am simply streaming to a XBMC machine on my living room where there is no web browser. Having to startup my desktop just to come and shutdown UnRaid via the web interface is far from good usage. So, when is then the fix coming for this "known" error? Given the current average release schedule, I hope it is not 2 months... --- Edit: Reviewed the original reply and I correct that it was not promised for next release. Nevertheless, this was working fine on the 5.x branches, so why not just transport the same shutdown method when pressing the power button? Mind you, I just tested on the 6.0 b7 and I updated, started the system, did NOT access any files and simply shutdown via power button. It still started a parity check on the next boot. I have absolutely no mods or plugins on my system - it is a totally default install that has been working well for a long time. In 'stock' unraid, the power switch event is tied to same code that webGui uses to shutdown, so is most cases it should work correctly. Depending on plugins and system activity it might not work - this is a known problem and will be fixed. In some ways it is a tough problem because forcing a shutdown can result in data loss. Consider what windows does: if it can't shut something down without possibly losing data, you get a nice dialog asking you what to do. Anyway, this is FYI - please let's not discuss this issue in this thread. A place to post bugs and features is coming soon, please everyone have just a little more patience. Will take a look at this next week, but I have had great success using powerbutton to shutdown unRAID without undergoing parity checks on reboot. Quote Link to comment
GHunter Posted August 24, 2014 Share Posted August 24, 2014 Just thought I would let you know that my Windows 7 VM on Xen works great under beta 7. It ran good on beta 5a, but crashed every few hours on beta 6. Another thing I noticed is an increase in network speed. I use MetaBrowser 2 an my scan would normally take 70 seconds to read through 1000 media folders. Under beta 7, I am getting speeds of 45-50 seconds. This is exactly the same VM with no changes made to software or hardware under the above mentioned beta's. Thanks Quote Link to comment
jonp Posted August 24, 2014 Share Posted August 24, 2014 Just thought I would let you know that my Windows 7 VM on Xen works great under beta 7. It ran good on beta 5a, but crashed every few hours on beta 6. Another thing I noticed is an increase in network speed. I use MetaBrowser 2 an my scan would normally take 70 seconds to read through 1000 media folders. Under beta 7, I am getting speeds of 45-50 seconds. This is exactly the same VM with no changes made to software or hardware under the above mentioned beta's. Thanks That's great feedback. Thank you for sharing!! Quote Link to comment
jumperalex Posted August 25, 2014 Share Posted August 25, 2014 I hope this is where you want stuff like this Tom / Jon ... about 10 minutes in: Aug 23 14:56:58 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 19 slots Aug 23 14:58:34 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 20 slots ... Aug 23 15:01:45 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 19 slots Aug 23 15:01:45 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 20 slots ... started another torrent ... Aug 23 15:19:00 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 20 slots Aug 23 15:19:00 arch_vm kernel: xen_netfront: xennet: skb rides the rocket: 20 slots I used to get a few of these in the past i think so this isn't too bad. It is still early days, but I would say beta7 and Xen are pretty stable. Hmm, those are the errors we are looking to find and were hoping were squashed. Let me know if they persist and how they affect your VMs. Also post your config for the VM and your network.cfg settings. Need to see some details to help diagnose. Well like I said, I was seeing these in beta5a too I'm about 99% sure. They never caused a problem because they were about as frequent as this. I'll try to dig more into my old logs to confirm and of course let you know if I encounter problems. I expect running more simultaneous torrents (with more open connections) might push it; I'll try that in a bit when I have things to download. VM Config: name = "arch" memory = '4096' vcpus = '8' vif = [ 'bridge=xenbr0,mac=00:16:3E:1E:66:1F' ] disk = ['file:/mnt/appdisk/vm/arch/arch.img,xvda,w','file:/mnt/appdisk/vm/arch/data.img,xvdb,w'] bootloader = "pygrub" Network Config: # Generated settings: USE_DHCP="no" IPADDR="10.0.0.4" NETMASK="255.255.255.0" GATEWAY="10.0.0.1" DHCP_KEEPRESOLV="no" DNS_SERVER1="10.0.0.1" DNS_SERVER2="" DNS_SERVER3="" BONDING="no" BONDING_MODE="1" BRIDGING="yes" BRNAME="xenbr0" BRSTP="no" Syslinux Config: default /syslinux/menu.c32 menu title Lime Technology prompt 0 timeout 50 label unRAID OS # menu default kernel /bzimage append initrd=/bzroot label unRAID OS Safe Mode (no plugins) kernel /bzimage append initrd=/bzroot unraidsafemode label Memtest86+ kernel /memtest label Xen/unRAID OS menu default kernel /syslinux/mboot.c32 # append /xen dom0_max_vcpus=1 dom0_vcpus_pin dom0_mem=2048M,max:2048M --- /bzimage --- /bzroot append /xen dom0_mem=2048M,max:2048M --- /bzimage --- /bzroot label Xen/unRAID OS Safe Mode (no plugins) kernel /syslinux/mboot.c32 append /xen --- /bzimage --- /bzroot unraidsafemode I've had some more batches of rocket riding but so far it hasn't caused a problem. I have however seen something new Aug 24 04:32:08 Tower kernel: vif vif-1-0 vif1.0: draining TX queue Aug 24 04:32:48 Tower last message repeated 4 times Aug 24 04:33:49 Tower last message repeated 6 times Aug 24 04:34:59 Tower last message repeated 7 times Aug 24 04:36:09 Tower last message repeated 7 times Aug 24 04:37:19 Tower last message repeated 7 times Aug 24 04:38:29 Tower last message repeated 7 times Aug 24 04:39:39 Tower last message repeated 7 times Aug 24 04:40:49 Tower last message repeated 7 times Aug 24 04:41:59 Tower last message repeated 7 times Aug 24 04:43:00 Tower last message repeated 6 times Aug 24 04:44:10 Tower last message repeated 7 times Aug 24 04:45:20 Tower last message repeated 7 times Aug 24 04:46:30 Tower last message repeated 7 times Aug 24 04:46:40 Tower kernel: vif vif-1-0 vif1.0: draining TX queue Aug 24 04:46:48 Tower kernel: mdcmd (55): spindown 1 Aug 24 04:46:48 Tower kernel: mdcmd (56): spindown 2 Aug 24 04:46:50 Tower kernel: vif vif-1-0 vif1.0: draining TX queue Aug 24 04:47:00 Tower kernel: vif vif-1-0 vif1.0: draining TX queue Aug 24 04:47:09 Tower kernel: mdcmd (57): spindown 0 Aug 24 04:47:09 Tower kernel: mdcmd (58): spindown 3 Aug 24 04:47:10 Tower kernel: vif vif-1-0 vif1.0: draining TX queue Aug 24 04:47:50 Tower last message repeated 4 times Aug 24 04:49:00 Tower last message repeated 7 times Aug 24 04:50:10 Tower last message repeated 7 times Aug 24 04:51:11 Tower last message repeated 6 times Aug 24 04:52:21 Tower last message repeated 7 times Aug 24 04:53:31 Tower last message repeated 7 times Aug 24 04:54:41 Tower last message repeated 7 times Aug 24 04:55:51 Tower last message repeated 7 times Aug 24 04:57:01 Tower last message repeated 7 times Aug 24 04:58:11 Tower last message repeated 7 times Aug 24 04:59:12 Tower last message repeated 6 times Aug 24 05:00:22 Tower last message repeated 7 times Aug 24 05:01:32 Tower last message repeated 7 times Aug 24 05:02:42 Tower last message repeated 7 times Aug 24 05:03:52 Tower last message repeated 7 times Aug 24 05:05:02 Tower last message repeated 7 times Aug 24 05:06:12 Tower last message repeated 7 times Aug 24 05:07:13 Tower last message repeated 6 times Aug 24 05:08:23 Tower last message repeated 7 times Again no problems, just feeding the beast Oh and this is from my unraid logs. The rocket rides are of course from the Xen Arch VM logs. Quote Link to comment
Sandorr Posted August 25, 2014 Share Posted August 25, 2014 Tom, can you elaborate on this one ... - emhttp: removed automatic file system expand when small drive is replaced by bigger drive. Does this mean that if you are upsizing a disk, say from 2T to 4T, the 4T would continue to be limited to 2T? Can it be manually expanded? Will beta8 detect this condition and auto-expand any replaced disks? Would you advise users to boot back into 6b6 to do the replacement? Code used to just do an unconditional "resize" upon every mount, and that was not an issue because with reiserfs it just was a no-op. But with btrfs/xfs there needs to be a little more sophistication in the coding. I didn't want to delay beta7 another week to get this in, so left auto-resize out until next release. If someone needs to do an expansion before beta8 I will post instructions. Somewhat related - if you add a btrfs or xfs disk in 6b7, can you boot back into 6b6? (I realize the XFS and BRTFS disks would appear unformatted and their contents not accessible, but would parity be maintained and therefore ability to do a disk rebuild?) Yes. Can you post the instructions for 'If someone needs to do an expansion before beta8 I will post instructions.'? [Edited to delete lines that can cause invalid parity - limetech] Quote Link to comment
limetech Posted August 25, 2014 Author Share Posted August 25, 2014 Can you post the instructions for 'If someone needs to do an expansion before beta8 I will post instructions.'? The command you used would be ok if that was your 'cache' device, but if that was an array device, resizing the partition like that invalidates parity. For that reason, I edited your post and deleted those lines because I don't want anyone else to do that. I'll post correct instructions later... Quote Link to comment
Sandorr Posted August 25, 2014 Share Posted August 25, 2014 Can you post the instructions for 'If someone needs to do an expansion before beta8 I will post instructions.'? The command you used would be ok if that was your 'cache' device, but if that was an array device, resizing the partition like that invalidates parity. For that reason, I edited your post and deleted those lines because I don't want anyone else to do that. I'll post correct instructions later... Sounds good, thanks for the update. I will watch for the correct instructions later. Meanwhile...I will start a parity rebuild... Quote Link to comment
coppit Posted August 25, 2014 Share Posted August 25, 2014 I can also confirm that this beta allows my Windows VM to stay up. It's been rock-solid. Also, if anyone is wondering if reformatting one of your drives as btrfs is worth doing in order to use Docker, the answer is "yes". Quote Link to comment
SSD Posted August 25, 2014 Share Posted August 25, 2014 Can you post the instructions for 'If someone needs to do an expansion before beta8 I will post instructions.'? The command you used would be ok if that was your 'cache' device, but if that was an array device, resizing the partition like that invalidates parity. For that reason, I edited your post and deleted those lines because I don't want anyone else to do that. I'll post correct instructions later... Sounds good, thanks for the update. I will watch for the correct instructions later. Meanwhile...I will start a parity rebuild... Parity check should be fine. If the replacement disk was precleared you should not see many parity errors. If not you'd see a bunch. But either way the parity check will correct then (if you run a correcting check) and be faster than a build. Quote Link to comment
Sandorr Posted August 25, 2014 Share Posted August 25, 2014 Can you post the instructions for 'If someone needs to do an expansion before beta8 I will post instructions.'? The command you used would be ok if that was your 'cache' device, but if that was an array device, resizing the partition like that invalidates parity. For that reason, I edited your post and deleted those lines because I don't want anyone else to do that. I'll post correct instructions later... Sounds good, thanks for the update. I will watch for the correct instructions later. Meanwhile...I will start a parity rebuild... Parity check should be fine. If the replacement disk was precleared you should not see many parity errors. If not you'd see a bunch. But either way the parity check will correct then (if you run a correcting check) and be faster than a build. Thanks for the info, I will give it a shot. Quote Link to comment
hackztor Posted August 26, 2014 Share Posted August 26, 2014 How would checkdsks be performed on xfs and btrfs? Also when parity is created it calculates file system right? So would it be any different with xfs or btrfs vs reiserfs. Is mix and matching okay. Quote Link to comment
itimpi Posted August 26, 2014 Share Posted August 26, 2014 Also when parity is created it calculates file system right? So would it be any different with xfs or btrfs vs reiserfs. Is mix and matching okay. Parity is file system agnostic - it works at the physical sector level. This means that mixing and matching does not affect how parity is handled. Quote Link to comment
dlandon Posted August 26, 2014 Share Posted August 26, 2014 This beta is working fine on my server with a Xen Debian VM running Owncloud and a few other apps. Quote Link to comment
2stroke Posted August 27, 2014 Share Posted August 27, 2014 Great work can't wait till beta is finished and we see an RC, already got my new rig close to up and going, the poor old c2d p965 is feeling its age and want some immou love from the xeon Quote Link to comment
shremi Posted August 28, 2014 Share Posted August 28, 2014 I am having problems ..... I was using beta 6 i stopped the array copied the files as per the readme instructed rebooted everything is ok i can see my drives but i cant mount the array via the web gui..... Also i am seeing 24 available drives and i have the Plus version dont know if that is normal Any help would be appreciated Quote Link to comment
SSD Posted August 28, 2014 Share Posted August 28, 2014 I am having problems ..... I was using beta 6 i stopped the array copied the files as per the readme instructed rebooted everything is ok i can see my drives but i cant mount the array via the web gui..... Also i am seeing 24 available drives and i have the Plus version dont know if that is normal Any help would be appreciated unRAID now allows you to partition drives between which can be assigned to the array and which can be assigned to the cache pool. The more in the cache pool, the fewer available for array disks. unRAID will still limit your ability to assign more than the prescribed number of disks. At least that is what was stated previously in this thread, you can always try it and see for sure. Quote Link to comment
pras1011 Posted August 29, 2014 Share Posted August 29, 2014 I am assuming that this is still not safe to use on my main server? 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.