February 9, 201115 yr Author Ok, little issue: When I reboot my server, the drive assignments get muffed and I have to re-assign my drives. I've attempted to go into the Utils and "New Slots", but same result after reboot. Idea? This is not good... please send email to [email protected]. BTW: in regards to your avatar, see attached.
February 9, 201115 yr Ok, little issue: When I reboot my server, the drive assignments get muffed and I have to re-assign my drives. I've attempted to go into the Utils and "New Slots", but same result after reboot. Idea? This is not good... please send email to [email protected]. BTW: in regards to your avatar, see attached. RE: Avatar.. No thanks RE: Issue I did a little dance around a fire and sacrificed a pig, rebooted 4 times and got it to work, but I had to do this: Assign all drives to their individual slots NOT start the array Reboot Upon reboot, they all fell into place. the 3 times I tried assigning the drives and then start the array (and then reboot), it came back muffed. If you still want me to e-mail, I can.
February 9, 201115 yr Ok, little issue: When I reboot my server, the drive assignments get muffed and I have to re-assign my drives. I've attempted to go into the Utils and "New Slots", but same result after reboot. Idea? This is not good... please send email to [email protected]. BTW: in regards to your avatar, see attached. RE: Avatar.. No thanks RE: Issue I did a little dance around a fire and sacrificed a pig, rebooted 4 times and got it to work, but I had to do this: Assign all drives to their individual slots NOT start the array Reboot Upon reboot, they all fell into place. the 3 times I tried assigning the drives and then start the array (and then reboot), it came back muffed. If you still want me to e-mail, I can. Did you run the "New Slots" script on the "Utils" page. I did not try a reboot without using it... should have... hindsight is 20-20.
February 9, 201115 yr Read my previous post... Short answer: Yes. Sorry about that. note to self: read previous posts before sticking foot in mouth.
February 9, 201115 yr ok another question now. I'm running beta4 and now my Windows computer that accesses the unRAID server using Samba can no longer read a ton of files. If I go through windows explorer, I can see for example my "Downloads" folder inside one of the shares. However, when I try to click on a file it usually throws an error saying "File not found" or "Location not found". Any thoughts or where I should look?
February 9, 201115 yr For me, same as with beta3, I am unable to access the unraid web interface. I followed all instructions, and before starting I disabled all 3rd party addons, including unmenu. I can telnet and upon inspecting the syslog I see a "segmentation fault" message when the "emhttp &" line of my go script executes. Also, I can execute "uu" and then access the first page of the unmenu web interface. But, that is all I can do. For now, I'm back up and running with v4.7.
February 9, 201115 yr Author For me, same as with beta3, I am unable to access the unraid web interface. I followed all instructions, and before starting I disabled all 3rd party addons, including unmenu. I can telnet and upon inspecting the syslog I see a "segmentation fault" message when the "emhttp &" line of my go script executes. Also, I can execute "uu" and then access the first page of the unmenu web interface. But, that is all I can do. For now, I'm back up and running with v4.7. C'mon man, for a beta release it's even more important to attach the system log.
February 9, 201115 yr Author ok another question now. I'm running beta4 and now my Windows computer that accesses the unRAID server using Samba can no longer read a ton of files. If I go through windows explorer, I can see for example my "Downloads" folder inside one of the shares. However, when I try to click on a file it usually throws an error saying "File not found" or "Location not found". Any thoughts or where I should look? How many is a "ton", 2000? or is it a "metric ton" (2200)? A system log would be nice too. You see that little button on the menu bar named "log" - click that and a real-time system log opens in a new browser window (or tab). You can start that up, then try listing a ton of files and accessing one - does something pop up in the log? I need to see if so. Something else to try, I noticed during development with servers going up/down & protocols changing, etc., occasionally Win7 got "confused" - it would show the shares, but any click inside a share produced "Location not found" - the solution was to reboot windows.
February 9, 201115 yr ok another question now. I'm running beta4 and now my Windows computer that accesses the unRAID server using Samba can no longer read a ton of files. If I go through windows explorer, I can see for example my "Downloads" folder inside one of the shares. However, when I try to click on a file it usually throws an error saying "File not found" or "Location not found". Any thoughts or where I should look? How many is a "ton", 2000? or is it a "metric ton" (2200)? A system log would be nice too. You see that little button on the menu bar named "log" - click that and a real-time system log opens in a new browser window (or tab). You can start that up, then try listing a ton of files and accessing one - does something pop up in the log? I need to see if so. Something else to try, I noticed during development with servers going up/down & protocols changing, etc., occasionally Win7 got "confused" - it would show the shares, but any click inside a share produced "Location not found" - the solution was to reboot windows. I was about to respond that rebooting Windows seemed to have done the trick. I will apologize now for mouthing off at the wrong time
February 9, 201115 yr Wanted to report positive feedback on unRAID 5.0b4. As reported earlier, I successfully added 2 precleared 2TB 4K-aligned drives. As of a few minutes ago, the parity check nocorrect completed successfully while a minor amount of writes and reads were occurring. It seems the format slightly changed, but if I calculated properly it looks like a rate of roughly 75297K/sec. That's right around my old performance when I had 2 less 2TB drives connected (75300-75660K/sec). Feb 8 22:11:40 reaver kernel: mdcmd (2587): check NOCORRECT Feb 8 22:11:40 reaver kernel: md: recovery thread woken up ... Feb 8 22:11:40 reaver kernel: md: recovery thread checking parity... Feb 8 22:11:40 reaver kernel: md: using 3456k window, over a total of 1953514552 blocks. Feb 9 05:24:05 reaver kernel: md: sync done. time=25944sec Feb 9 05:24:05 reaver kernel: md: recovery thread sync completion status: 0
February 9, 201115 yr Ok, little issue: When I reboot my server, the drive assignments get muffed and I have to re-assign my drives. I've attempted to go into the Utils and "New Slots", but same result after reboot. Idea? This is not good... please send email to [email protected]. BTW: in regards to your avatar, see attached. RE: Avatar.. No thanks RE: Issue I did a little dance around a fire and sacrificed a pig, rebooted 4 times and got it to work, but I had to do this: Assign all drives to their individual slots NOT start the array Reboot Upon reboot, they all fell into place. the 3 times I tried assigning the drives and then start the array (and then reboot), it came back muffed. If you still want me to e-mail, I can. Did you run the "New Slots" script on the "Utils" page. I did not try a reboot without using it... should have... hindsight is 20-20. I did not need to reboot after running teh "New Slots" script. When I first booted into unRAID, my disks were listed but not assigned correctly (on the Main screen). I ran the script and when I went back to the main menu, everyhting was hunky-dory.
February 9, 201115 yr I moved from 5.03b to 5.03b and had to a assign the disk drives (a screenshot from the 5.03b was helping) according to the release note. Again thank you for all the things working now. Here is my quick summary: * I´m able to access all shares via AFP, SMB * NFS is working just great * Timemachine running well * The webinterface is working Here is my go script; I had to remove the autoinstall from unMENU as this caused troubles on the webinterface: #!/bin/bash # Start the Management Utility /usr/local/sbin/emhttp & # # unMENU echo "/boot/unmenu/uu" | at now + 1 minute # # cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c # # Set disk read-ahead 2048 for i in /dev/md* do blockdev --setra 2048 $i done # # cpufreq support modprobe speedstep-lib # # TwonkyMedia server # /boot/twonkymedia/twonkymedia -inifile /boot/twonkymedia/twonkymedia-server.ini -logfile /var/log/TwonkyMediaServer-log.txt # # Increase network performance echo nameserver 192.168.178.1 >/etc/resolv.conf echo 192.168.178.28 tower >>/etc/hosts # # Powerbutton clean shutdown CTRLALTDEL=yes installpkg /boot/packages/powerdown-1.02-noarch-unRAID.tgz [ -f /sbin/powerdown ] && mv /usr/local/sbin/powerdown /usr/local/sbin/unraid_powerdown [ -f /sbin/powerdown ] && sed -i "sX/usr/local/sbin/powerdownX/sbin/powerdownX" /etc/acpi/acpi_handler.sh [ ! -f /usr/local/sbin/unraid_powerdown ] && sed -i "sX/sbin/init 0X/sbin/powerdownX" /etc/acpi/acpi_handler.sh sysctl -w kernel.poweroff_cmd=/sbin/powerdown Has anyone any idea, why the "Set disk read-ahead 2048" and the "cpufreq support" stuff is not working? Any other experience with that?
February 9, 201115 yr Edgar, Have you actually tested the NFS shares (mounting from another system)? The most I did was create the shares to make sure teh webgui didn't dump but have not tried to access them yet. John
February 9, 201115 yr edgarwallace can you confirm that plain unmenu works ? i understand that some of the packages would cause issues and that's why you had to comment the auto install line but unmenu works normal and still has all info ? i know it might sound crazy but it is one of the things i couldn't do without so if it works then i can try to upgrade tomorrow
February 9, 201115 yr Yes plain unMenu without the packages is running. I don't know which package is causing the troubles but will switch on package by package soon so that I know which the delinquent is. So I would say give an upgrade to 5.04b a try.
February 9, 201115 yr Have you actually tested the NFS shares (mounting from another system)? The most I did was create the shares to make sure teh webgui didn't dump but have not tried to access them yet. John Yes as described it's working well. I switched on NFS and exported on of the share needed by XBMC as Public.
February 9, 201115 yr I moved from 5.03b to 5.03b and had to a assign the disk drives (a screenshot from the 5.03b was helping) according to the release note. Here is my go script; I had to remove the autoinstall from unMENU as this caused troubles on the webinterface: What effect did it have? I have many packages being installed, but perhaps it is one of those you were installing that I am not that caused your issue. Can you give a bit more detail. As far as the "blockdev --setra"" It is largely unnecessary since unRAID sets the read-ahead buffer to 1024 by default. What was probably happening is you were invoking it too soon after starting emhttp. The /dev/md devices probably did not yet exist as emhttp had not yet started. You might try either a simple sleep delay, or something a bit more intelligent like I use in one of my scripts. /usr/local/sbin/emhttp & # # unMENU echo "/boot/unmenu/uu" | at now + 1 minute # wait until all the configured disks are mounted (array is then available) num_configured=`grep "disk[0-9]*=" /boot/config/disk.cfg | wc -l` while sleep 5 do md_disk_mounted=`mount | grep "/dev/md[0-9]*" | wc -l` [ "$num_configured" = "$md_disk_mounted" ] && break done # then other programs follow... they will not start until all the configured disks are mounted.
February 9, 201115 yr Just trolling through the syslog... Feb 9 04:19:59 unRAID kernel: md: sync done. time=32024sec Feb 9 04:19:59 unRAID kernel: md: recovery thread sync completion status: 0 Feb 9 04:40:01 unRAID logrotate: ALERT - exited abnormally. Any concern?
February 9, 201115 yr I upgraded from 4.7 to 5.0-b4 and everything went smoothly, following the procedure in the release notes. The only things which caused me moments of doubt: 1) I had to stop and start the array, after enabling nfs before nfs shares became accessible. 2) I started up, initially, with a vanila 'go' file, but quickly re-enabled most of my customisation. Squeezeboxserver came up perfectly (after adding my squeeze user back in the passwords file, and resetting all of the file ownerships on the Squeeze-related files). However, on enabling unMenu/auto-installs, the standard web interface went 'odd'. My immediate guess was that it was the php package which would be causing the problems - this turned out to be precisely true ... all my other auto-install packages appear to be working well. The only other items I've not re-enabled yet are my modified mover script and snap. Since 4.6rc3, as expected, the power saving variable processor clock hadn't been working, but it's back in operation now. I still have to go through all my shares enabling nfs and disabling smb, as required, but those I have adjusted are fine. So, as far as I'm concerned, full marks to Tom for a virtually perfect beta! Edit: Oh, of course, the other thing which doesn't work is the 'unRAID Main' page in unMENU. Edit 2: New 'oddities' appearing in my syslog (does anyone have any comments?): Feb 9 22:34:27 Tower kernel: ACPI Warning: 32/64 FACS address mismatch in FADT - two FACS tables! (20100702/tbfadt-369) (Minor Issues) Feb 9 22:34:27 Tower kernel: ACPI Warning: 32/64X FACS address mismatch in FADT - 0xCB61DF40/0x00000000CB61DE40, using 32 (20100702/tbfadt-486) (Minor Issues) ----- Feb 9 22:34:27 Tower logger: ERROR: Module md_mod does not exist in /proc/modules (Errors) ----- Feb 9 22:38:23 Tower kernel: inotifywait[4106]: segfault at a95f2360 ip b77d7978 sp bfb536cc error 6 in ld-2.11.1.so[b77d1000+1d000] (Errors) Feb 9 22:38:23 Tower kernel: inotifywait[4107]: segfault at a95a7360 ip b778c978 sp bf81c92c error 6 in ld-2.11.1.so[b7786000+1d000] (Errors)
February 9, 201115 yr Hi Joe, I can't launch the packages now but I included a screenshot with all packages having "auto_install" in the file name. I guess that shows you what packages are being installed if I would remove the # in front of the cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c line The effect on the webinterface is exactly the same as in the post here: http://lime-technology.com/forum/index.php?topic=10660.msg101428#msg101428
February 9, 201115 yr Also just upgraded from 5b02 to 5b04 everything works fine as far as i can see all unmenu addons that i had running are up already OPENSSL SSH Screen DynDNS Then other stuff that works sickbeard couchpotato SabNbzd only Hamachi has a problem Feb 9 21:04:56 p5bplus kernel: tun: version magic '2.6.32.9-unRAID SMP mod_unload CORE2 ' should be '2.6.36.2-unRAID SMP mod_unload CORE2 ' will put the error in the hamachi Thread the releaser can maybe upgrade the package Also Mymain has issues no ID's and my inventory is empty not the biggest issue of course seems the new beta is a bit faster or is it just an idea of me ?
February 9, 201115 yr The effect on the webinterface is exactly the same as in the post here: http://lime-technology.com/forum/index.php?topic=10660.msg101428#msg101428 that would be uRAID-web that is causing the issue. It is installing "php" on top of the already installed "php" that comes with unRAID 5.X, and the config file unRAID-web installs does not enable the use of the "php short tags" that lime-technology is using in their code. To fix is easy... run the following command. sed -i "s/^short_open_tag = Off/short_open_tag = On/" /boot/custom/php/php.ini You can run it on the command line and you can also put it in your "go" script after the line that does the "auto_install" of the unMENU packages. (and I'll probably need to update the unRAID-web package to make the same change, or perhaps not install "php" at all on a 5.0 install of unRAID-web, since "php" is already installed.) Joe L.
Archived
This topic is now archived and is closed to further replies.