savestheday

I started learning some Apache stuff specifically for running all my python servers behind a reverse proxy. I wanted to access my servers from anywhere but didn't want to open a lot of ports. I learned a lot this from these links: http://www.apachetutor.org/admin/reverseproxies http://code.google.com/p/sickbeard/issues/detail?id=684 http://code.google.com/p/sickbeard/wiki/AdvancedSettings http://help.couchpotatoapp.com/kb/tips/reverse-proxy http://httpd.apache.org/docs/2.0/programs/htpasswd.html First, you'll need to get Apache & its dependencies. I've found these packages on http://pkgs.org: http://slackware.cs.utah.edu/pub/slackware/slackware-13.1/patches/packages/httpd-2.2.21-i486-1_slack13.1.txz http://slackware.cs.utah.edu/pub/slackware/slackware-13.1/patches/packages/apr-1.4.5-i486-1_slack13.1.txz http://slackware.cs.utah.edu/pub/slackware/slackware-13.1/patches/packages/apr-util-1.3.12-i486-1_slack13.1.txz Those are the newest versions I could find as of this writing. They're listed in the Slackware patches section and that's where my Slackware experience is lacking. I'm 99% sure these are fine to use as standalone packages but the word "patch" makes me think you may need a previous version and patch it with these. I'm currently running the following pkgs and they're working fine but I am going to upgrade (and edit this tutorial if need be). I've provided links to these pkgs if the above don't work. ftp://ftp.slackware.com/pub/slackware/slackware-13.1/patches/packages/httpd-2.2.20-i486-1_slack13.1.txz ftp://ftp.slackware.com/pub/slackware/slackware-13.37/slackware/l/apr-1.4.2-i486-2.txz ftp://ftp.slackware.com/pub/slackware/slackware-13.37/slackware/l/apr-util-1.3.10-i486-1.txz So let's break this down.... httpd-2.2.20-i486-1_slack13.1.txz - This is the main Apache install pkg. apr-1.4.2-i486-2.txz - Apache pre-req pkg apr-util-1.3.10-i486-1.txz - Apache pre-req pkg I couldn't launch Apache without those apr pkgs. Again, after some reading on the unRAID forums, I saw that they were pre-reqs. Turns out, as far as I can tell, they enable Apache to run on different Linux/Windows distributions. Let's get started! First SSH/Telnet into your unRAID box. I use SSH. Don't think I need to explain how to do this.... Download the pkg's to your /boot/packages folder cd /boot/packages wget ftp://ftp.slackware.com/pub/slackware/slackware-13.1/patches/packages/httpd-2.2.20-i486-1_slack13.1.txz wget ftp://ftp.slackware.com/pub/slackware/slackware-13.37/slackware/l/apr-1.4.2-i486-2.txz wget ftp://ftp.slackware.com/pub/slackware/slackware-13.37/slackware/l/apr-util-1.3.10-i486-1.txz Now install them: installpkg /boot/packages/httpd-2.2.20-i486-1_slack13.1.txz installpkg /boot/packages/apr-1.4.2-i486-2.txz installpkg /boot/packages/apr-util-1.3.10-i486-1.txz Make new directories on your USB stick mkdir /boot/config/etc/httpd mkdir /boot/config/etc/htpasswd Copy your existing httpd folder contents there: cp -R /etc/httpd/* /boot/config/etc/httpd Let's create a password file to secure the server: /usr/bin/htpasswd -c /boot/config/etc/htpasswd/passwords youruserid Follow the prompts to create a password for your account. Replace youruserid with whatever you want. We're doing this so that we can have a persistent config through reboots. Fire up your favorite txt editor. I use TextMate on the Mac (via Transmit/FTP) or nano from the shell. Open up: /boot/config/etc/httpd/httpd.conf You're going to need to edit the following fields: Listen 5780 This is the port your server will listen on. I strongly suggest changing because the default unRAID web interface runs on the standard port of 80 and Apache won't launch. I changed it something very non standard, port 5780. User apacheacct Group users Apache won't run as the user Apache (unless you create that acct). I'm running this under a user account I created - apacheacct. I don't want to run this as root for security reasons. ServerName servername.dyndns.org:5780 This is another setting you'll want to change. I changed this to my dyndns acct which is the outward hostname I'm running this on. I'd recommend getting a dyndns acct and using that. Add the following Locations to your conf file under the directory section: <Directory /> Options FollowSymLinks AllowOverride None Order deny,allow Deny from all </Directory> <Location /couchpotato> ProxyPass http://localhost:5000 ProxyPassReverse http://localhost:5000 AuthType Basic AuthName "EchoBase" AuthUserFile /boot/config/htpasswd/passwords Require user youruserid </Location> <Location /sabnzbd> ProxyPass http://localhost:2020/sabnzbd ProxyPassReverse http://localhost:2020/sabnzbd AuthType Basic AuthName "EchoBase" AuthUserFile /boot/config/htpasswd/passwords Require user youruserid </Location> <Location /sickbeard> ProxyPass http://localhost:8081/sickbeard ProxyPassReverse http://localhost:8081/sickbeard AuthType Basic AuthName "EchoBase" AuthUserFile /boot/config/htpasswd/passwords Require user youruserid </Location> AuthType - this is the basic auth we're using. I would love to switch to SSL, and started down that road (generating certs, etc), but haven't fleshed it out yet. I'm welcome to suggestions here if you've got experience! AuthName - is the name your server presents itself with. My server is called EchoBase so I chose to match that here. AuthUserFile - is the file we created previously with our account and password (/boot/config/etc/htpasswd/passwords) You'll notice that I run SABnzbd on a non standard port. The other two I run on the default ports. You'll need to update the config.ini files of SickBeard and CouchPotato to reflect the new directories. In SickBeard's config.ini, add/change this setting in your favorite txt editor: web_root = /sickbeard In CouchPotato's config.ini, add/change this setting in your favorite txt editor: urlbase = /couchpotato You will need to restart CouchPotato and Sickbeard to use these new config settings. Add those commands to your GO script. This is how my GO file looks. installpkg /boot/packages/httpd-2.2.20-i486-1_slack13.1.txz installpkg /boot/packages/apr-1.4.2-i486-2.txz installpkg /boot/packages/apr-util-1.3.10-i486-1.txz rm -R /etc/httpd ln -s /boot/config/etc/httpd /etc/httpd apachectl start You'll notice that I'm removing the default /etc/httpd folder created by the pkg and symlinking to our persistent config. apachectl start starts your server. You can play around with the httpd.conf file and restart Apache very easily: apachectl graceful Again, this is a work in progress. I'm no expert in Linux and I welcome any suggestions to this tutorial. Figured some people might benefit from this!

So John, I think I came up with a creative way to mount that drive This mount was rock solid as long as I used a real PCI slot cover and not the one that the Norco includes (which is removable and not stable). But then I researched if there were any PCI slot mounts for a 3.5" HDD. There's not but there IS one for a 2.5" HDD. Not sure if you've seen this before: http://www.amazon.com/gp/product/B002MWDRD6/ref=oh_o00_s00_i00_details I just bought this, should see how nice it is in a few days. Thought this would be perfect for you. I have a 500GB 2.5" hard drive laying around that should do the job for me.

Hi John, I've got the same setup as you (case, mobo). Was wondering if you ever were able to mount a 3.5" drive in the Norco case? I want to use this 750gb black drive as my data store but my drive bays are full. Thx!

So this is really one of the last pesky problems I have. I've got Dropbox running thx to this thread: http://lime-technology.com/forum/index.php?topic=10160.0 Here's my GO script # Start Dropbox echo 100000 > /proc/sys/fs/inotify/max_user_watches ln -s /mnt/cache/.dropboxdb /root/.dropbox ln -s /mnt/cache/.Dropbox /root/Dropbox ln -s /boot/custom/packages/.dropbox-dist /root/.dropbox-dist sleep 20 # Launch Daemon # cd /root/.dropbox-dist # dropboxd & # cd /root # /root/.dropbox-dist/dropboxd & # Launch via dropbox.py cd /boot/scripts python dropbox.py start As you can see, I have tried both launching the daemon manually and launching it from dropbox.py. Here's the weird part: neither work when launching from the GO script. If I launch dropboxd manually, it will launch, but it intermittently quits. If I launch from the GO script, it doesn't respect the symlinks I just created (no matter how long of a wait I put in) and says that the it's not associated with my acct. If I manually launch from the dropbox.py start it starts just fine but from the GO script, it prompts me to install the "dropbox daemon". I can do this via the dropbox.py script but them I'm prompted to accept that it will download the proprietary daemon which you can't suppress via cmd line. All in all, it's very annoying. I'm just not sure what prevents it from launching in the GO script but I know it's gotta be something simple I'm missing. Anyone run into this?

Okay I've moved all my MP3's to disk17 and have Subsonic pointed at that MP3 share (again the MP3 share is only on disk17) and my drives are still awake.

This helped me out a lot. Thank you so much!

Thanks Joe L. I've got a question tho....should I just run mv /mnt/disk17/Movies/HD /mnt/user/Movies/HD ?? I'd feel more comfortable running with RSYNC since it has some verification that things got moved over correctly. Is there a problem doing rsync instead of a move then deleting them off of disk17 when done? I just tried rsyncing the data: rsync -av --stats --progress /mnt/disk17/DVD-R/ /mnt/user/DVD-R/ and RSYNC said no files got transferred: Number of files: 658 Number of files transferred: 0 Total file size: 165617427761 bytes Total transferred file size: 0 bytes Literal data: 0 bytes Matched data: 0 bytes File list size: 17133 File list generation time: 0.001 seconds File list transfer time: 0.000 seconds Total bytes sent: 17660 Total bytes received: 526 Color me confused. I thought those files only existed on disk17. I use MOST FREE as my Allocation method and have excluded DISK17 from all shares. Do I also need to incl. Disk1-16 & 18-20 too?

I'm trying out the new beta. I will say that I can send a test email just fine from my 0.8 SF setup (could NOT do that previously!) but the reports are never sent on the schedule. I checked my config and the only thing that would prevent them from being sent is if all disks are spun down but that's not the case. EDIT Looks like a reboot was necessary for the schedule to kick in. Thx again!

preproman are you seeing what I saw in beta 10? http://lime-technology.com/forum/index.php?topic=14158.msg138076#msg138076 It looks like that last disk, depending on disk assignment, pushes the already assigned disk out of the array. Tom, I attached a syslog on that post if you want to see what was happening on my box. I have not fixed this and I currently have my 24th drive (which I'd like to mount outside of the array) hanging out of my machine because if I power it on, it will push my hitachi out of the array therefore screwing it up.

So the way I set up my unRAID was that my shares span all my drives going to the most free. I have a fairly new 2TB drive that's 17% used. Now, I want to make my (empty) MP3 share to use *only* for that drive and exclude all other shares from that drive. I want to do this because Subsonic keeps my drives awake and with my current MP3 situation, where they're in another share that's spanned, doesn't work. If I exclude this drive from other shares, will the data that's already on those drives move to the next drive with the most free data? If not, can I go in via SFTP or SSH and move them from /mnt/drive20 to /mnt/user/<shares>??

Any fix for the Send mail stuff? I haven't gotten an email update in a month. Thx.

I am running subsonic from the subsonic.sh script. On the top of the script I have: SUBSONIC_HOME=/mnt/cache/.subsonic/home Subsonic runs only from my cache drive. However, it's scanning an MP3 share that spans all of my drives. This constant scanning, I assume, is what keeps my drives awake. I've taken subsonic out of my go script and like clockwork, the drives spundown as they were supposed to. Conventional wisdom says I should just move my mp3's to one drive (and I very may well do that) but what do I do for the time being? I'm sure Subsonic has a scanning frequency but I'm also sure that would conflict with my spindown schedule anyway. Anyone run into this?

The only problem I have, and I posted this in a new thread on the applications forum, is that for some reason, even if running under root, I get these errors: 2011-08-18 12:13:44,149 ERROR: Error removing workdir (/mnt/cache/.sabnzbd/downloads/incomplete/...

Uggh my AFP issues persist. I shouldn't really say that since they got much better since setting up my AFP the way Nezil described in the beta 10 thread http://lime-technology.com/forum/index.php?topic=14158.60 but I have done some testing on a new hackintosh I built. The new hack doesn't have my credentials saved for the AFP share so it connects as guest. When connecting as "Guest", clicking on my unRAID server on the Finder sidebar gives me an instantaneous view of my shares. It's exactly what it should be like! However, when I login with my account ('redfive') on my MacBook Pro, it takes at least a minute to list those same shares. I can only assume it's because it's checking file permissions against my account and when I connect as guest it doesn't? When I connect as 'redfive', I get a bunch of what's listed below in the syslog: Aug 18 09:20:06 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:06 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:06 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:06 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:07 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:10 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/HD/.DS_Store Aug 18 09:20:11 EchoBase shfs/user: duplicate object: /mnt/disk9/Movies/HD/.DS_Store Aug 18 09:20:15 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:15 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store Aug 18 09:20:16 EchoBase shfs/user: duplicate object: /mnt/disk6/Movies/.DS_Store Aug 18 09:20:16 EchoBase shfs/user: duplicate object: /mnt/disk11/Movies/.DS_Store I don't see this connecting as guest. It's getting to the point where I may just open up full access to the shares to everyone because it's quicker viewing them that way. If I connect as 'redfive' on my hack, I'm back to waiting. Maybe Nezil knows what's up with this? Btw this is not really a beta 11 issue, same thing happened in beta 10. Just figured since Tom was able to get the final version of Netatalk that some of the issues might be solved.

Tom, Was the Mover scheduling bug fixed?

I ran this script from Nezil to get rid of all my db files: /etc/rc.d/rc.atalk stop cd /mnt find disk*/ -name .AppleD* -exec rm -r {} \; That fixed it for me. It should recreate them after that. The other thing I did was follow Nezil's instructions a few pages back to edit the default location. Been working a lot better since then.

I ve never seen that message before. Maybe I should feel honored to have found something no one else found! So good news (sort of). I pulled the last (unassigned) SAMSUNG drive from the Norco and left the hitachi in there. I'm now able to assign the Hitachi So I'm surmising that maybe my first theory was true. It sees (4) slots left in your unRAID and chooses the first four drives it finds. Not sure it's a major bug - I'm an edgecase. I had all of these drives left over from two FreeNAS's. Something to keep in mind tho.

I have a pro license, the $119 one. Syslog attached. Thx! (please read my edit above, hope I didn't screw up the process by bringing 3 of the 4 online) Damn syslog wouldn't upload, it's here tho: http://cl.ly/2R3J1e0O3S0L1T1p2D42 Looks like I found the issue! Aug 14 21:40:38 EchoBase emhttp: SAMSUNG_HD204UI_S2H7J1CZC06138 (sdr) 1953514584 Aug 14 21:40:38 EchoBase emhttp: SAMSUNG_HD204UI_S2H7JD2ZA12046 (sdt) 1953514584 Aug 14 21:40:38 EchoBase emhttp: SAMSUNG_HD204UI_S2H7JD2ZA12067 (sdu) 1953514584 Aug 14 21:40:38 EchoBase emhttp: SAMSUNG_HD204UI_S2H7JD2ZA12066 (sdv) 1953514584 Aug 14 21:40:38 EchoBase emhttp: SAMSUNG_HD204UI_S2H7JD2ZA12047 (sdw) 1953514584 Aug 14 21:40:38 EchoBase emhttp: too many devices to add ../../devices/pci0000:00/0000:00:01.1/0000:02:00.0/host2/port-2:7/end_device-2:7/target2:0:7/2:0:7:0/block/sdx Aug 14 21:40:38 EchoBase emhttp: shcmd (2): modprobe md-mod super=/boot/config/super.dat slots=21 |& logger Aug 14 21:40:38 EchoBase kernel: xor: automatically using best checksumming function: pIII_sse Aug 14 21:40:38 EchoBase kernel: pIII_sse : 13513.200 MB/sec Aug 14 21:40:38 EchoBase kernel: xor: using function: pIII_sse (13513.200 MB/sec) Aug 14 21:40:38 EchoBase kernel: md: unRAID driver 2.1.2 installed

Okay having a really odd issue.... I just precleared (5) new hard drives. I have (4) slots open in my unRAID. What I planned on doing was bringing 4 of the 5 online to completely fill my unRAID server. (1) drive is a 2TB Hitachi Coolspin, (4) are the Samsung green 2tb drives. Problem is, I can't choose the Hitachi whatsoever in the GUI. unRAID sees it because it's in unMENU and I can test it with preclear. So a few things I was thinking.... Do you think that unRAID just sees that I have only (4) slots left and automatically chooses the first (4) drives? (this hitachi is technically the last device /dev/sdx) If so, why am I unable to replace one of my 1TB drives with it? Would I have to assign these (4) Samsungs first and then restart and then be able to replace my 1TB with this 2TB hitachi? Second thought was, and I know from a prior post it shouldn't matter, but this Hitach drive is NOT a 4k AF drive but preclear set it up as one based on my preference in the the web gui. You think that has anything to do with it? Again, I highly doubt it, but just throwing that out there. Hope someone can help me! EDIT Had 2 ideas...maybe it was Simple Features? Nope. Happens in default web gui. I also brought 3 of the Samsung drives online thinking that maybe it'd let me chose the Hitachi? Nope that didn't work either

Hi guys, Stupid question but figured I'd ask here instead of on a Mac forum. I just got the new AirPort Extreme but I can't connect to my unRAID via hostname (hothbase). It's a DNS issue but it's only happened since I switched routers. Here's what I've tried: 1.) dscacheutil -flushcache (on my Macs) 2.) Reinitializing the hostname in the web gui: Settings->Identification->Apply/Done It shows up in the router correctly: I do have a DHCP reservation for it: If I use http://hothbase.local it works. Any clue? Is this just an AirPort Extreme problem? BTW, I already expect plenty of jokes about naming all my hosts after Star Wars planets Thanks ahead of time!

Thanks for the update but the new package isn't working at all: Verifying package usenet_apps-0.2-i486-unRAID.tgz. gzip: stdin: unexpected end of file Installing package usenet_apps-0.2-i486-unRAID.tgz: PACKAGE DESCRIPTION: WARNING: Package has not been created with 'makepkg' Package usenet_apps-0.2-i486-unRAID.tgz installed. That's what I get when I install it. Any ideas?

speeding_ant Never properly thanked you for this amazing skin! I know we spoke about the unraid notify not working in the Beta 10 thread but I wanted to bring it up here as well. I'm not getting ANY email from unRAID anymore. Not sure what changed but the last email I remember getting was 3 weeks ago. Is there a new unRAID notify script you could post that fixes things? Is there an unraid_notify log somewhere I can parse to see what's up? I'm stumped. I miss my email notifications especially the ones regarding power outages. Thanks man!

Thanks for this! Only problem I see is misalignment in Google Chrome & Safari on the Mac

So I'm fairly certain that all cron stuff is messed up. I use simplefeatures and it's supposed to send me a status email every 3 hours. It was doing this in beta 9 but I haven't gotten an email in a while. My mover isn't working even if I "apply" it. I move stuff manually now

I know this is an old thread but I just wanted to say thank you!