getting SABnzbd and SickBeard working properly in the archVM image

[JustinChase]

I believe I'm using autofs, as that's what I remember IB saying in his setup/install videos.  I don't recall actually setting it up either way, so I may be wrong.  I'm also not sure how to check.  (and it makes me feel stupid).

 

Here is the cfg file I'm using to start the archVM...

 

name = "archVM"		# change me as required
bootloader = "pygrub"
memory = 4096		# update me to 4096 if using sabnzbd
vcpus = '2'		# update me to no of CPUs allocated
disk = [ 
'phy:/mnt/cache/domains/archVM/arch.img,xvda,w',
#	'phy:/mnt/user/Software/data.img,xvdb,w'
]

# replace xx:xx:xx with random a1:b2:c3 mac address
vif = [ 'mac=00:16:3e:a2:b2:c2,bridge=br0' ]

 

I just checked the downloads share, and NFS and SMB are both set to Public.

 

I also don't understand why these issues started creeping up again today.  I watched it finish several downloads and SB sent several files to SAB yesterday, and there were no issues, and no warnings in either program.

 

I suspect it may have something to do with 're-using' the incomplete downloads, but that's just a guess.

 

I just deleted everything in the incomplete folder.  I will just create a whole new location to put everything, and completely delete that folder.

 

I have one file that is 98% done, so I'm waiting for that to finish, then I'll manually repair it, then I'll just delete all folders.

 

If I wanted to not use root:root, I wonder how much work it will be to reset/change the permissions to the updated settings.  I suspect it would just be easier to uninstall both programs and start from scratch again, to just let it use the permissions it defaults to.  That doesn't sound like much fun either

 

PS some days I hate computers

[Ashe]

In sab have you gone to config > folders and set permissions for completed downloads to 770

 

When it's works it's great, when it doesn't.......

[JustinChase]

It's set to 755.  I can change it; not sure if it will 'stick' though.

 

RE: works or not - Agreed.

 

Have you tried NZBGet?  I may give it a shot now, since SAB is fighting me so hard.

[Ashe]

I would run chmod 777 /mnt/cache/ to just be sure or 755 as you have it

 

I have played with NZBGet before but you need to get this permissions issue sorted otherwise the problem will persist

 

 

Sent from my Q5 using Tapatalk

 

 

[JustinChase]

I would run chmod 777 /mnt/cache/ to just be sure or 755 as you have it

 

I have played with NZBGet before but you need to get this permissions issue sorted otherwise the problem will persist

 

agreed.

 

root@media:~# cd /mnt/cache

root@media:/mnt/cache# ls -l

total 0

drwxrwxrwx  3 nobody users  80 Apr 30 12:47 BIN/

drwxrwxrwx  4 nobody users  96 Jul 20  2013 JRiver/

drwxrwxrwx+ 4 nobody users  96 Apr 30 12:01 backup/

drwxrwxrwx  3 nobody users  80 Apr 30 12:47 documents/

drwxrwxrwx  5 nobody users 120 Apr 27 10:27 domains/

drwxrwxrwx  3 nobody users  80 Apr 30 15:05 downloads/

drwxrwxrwx  3 nobody users  80 Apr 24 22:44 plugins/

drwxrwxrwx  3 nobody users  72 Apr 30 03:59 video/

 

probably not related, but what's with the "+" in the backup directory?

[MyKroFt]

my mounts via fstab are in this format:

 

//tower/tv /mnt/TV cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0

 

Myk

 

[JustinChase]

and it just keeps getting weirder...

 

[root@IronicsArchVM_v4 downloads]# cd incomplete

-bash: cd: incomplete: No such file or directory

[root@IronicsArchVM_v4 downloads]# ls

incomplete

[root@IronicsArchVM_v4 downloads]# cd incomplete

-bash: cd: incomplete: No such file or directory

[root@IronicsArchVM_v4 downloads]# ls -l

ls: cannot access incomplete: No such file or directory

total 0

d??? ? ? ? ?            ? incomplete

 

 

what could be causing this???

 

2 minutes of trying to resolve, and now I get this...

 

[root@IronicsArchVM_v4 downloads]# chown -R root:root /net/media/mnt/user/downloads/

chown: cannot access ‘/net/media/mnt/user/downloads/incomplete’: No such file or directory

[root@IronicsArchVM_v4 downloads]# cd /net/media/mnt/user

[root@IronicsArchVM_v4 user]# ls

backup  documents  downloads  music  plugins

BIN    domains    JRiver    photos  video

[root@IronicsArchVM_v4 user]# cd downloads

[root@IronicsArchVM_v4 downloads]# ls

incomplete

[root@IronicsArchVM_v4 downloads]# ls -l

total 0

drwxr-xr-x 1 nobody users 152 Apr 30 14:36 incomplete

 

2 minutes later...

 

[root@IronicsArchVM_v4 downloads]# mv incomplete/ /opt/sabnzbd/Downloads/incomplete/

mv: cannot stat ‘incomplete/’: No such file or directory

[root@IronicsArchVM_v4 downloads]# ls -l

ls: cannot access incomplete: No such file or directory

total 0

d??? ? ? ? ?            ? incomplete

 

 

I'm ready to punch something 

[eroz]

Wow....not sure what is wrong since I am not in expert.  But you thought of just using tretflix since it is an all-in-one.  It's the VM (ubuntu), sabnzbd, couchpotato, sickbeard and i think plex.  It supposed to have everything linked together with the right permissions and be set up to be a "plug and play"?

[JustinChase]

I've thought about that, but it looks to be a bit of a PITA to build it, and I think my issues are probably not arch related, but are unRAID related, so I need to fix them anyway.  We'll see how much longer I can tolerate the frustration, and if it comes to it Tretflix is probably what I'll work on next.

 

I just shut down everything, installed a NIC, swapped around a couple of drives, to make sure I get my SATA3 drives attached to SATA3 controllers, and re-arranged some USB drives to try to get unRAID necessary stuff onto one controller, and stuff I want to pass thru onto a separate controller.  Not sure if I succeeded, but I just rebooted right now, so we'll see.

 

I turned off auto-start for SAB, and will install NZBGet now, and see how that works out.

 

I eliminated all the incomplete downloads, and destroyed he folders that contained them (the ones with the permissions issues), and will create download folders inside the VM, so see if that works better.  hopefully the SickBeard script can take them from there and put them where they belong, without issue.  We'll see.

 

Then, I'm going to work on PCI passthru for the windows 7 VM.  Hopefully I'll have this worked out before too much longer.  I much prefer to enjoy my media, than fight against it!

[Ashe]

if i run ls -l on arch for my network files

 

drwxr-xr-x 2 root root

 

The only difference being is root:root on my arch and yours is nobody:users

 

Did you change sabnzbd.service and sickbeard.service back from root:root in Arch?

[binhex]

if i run ls -l on arch for my network files

 

drwxr-xr-x 2 root root

 

The only difference being is root:root on my arch and yours is nobody:users

 

Did you change sabnzbd.service and sickbeard.service back from root:root in Arch?

 

Ashe, the user and group for files on unraid should be set to nobody:users, if its set to root:root then you will need to include your root password for unraid in fstab/autofs in order for the vm to read/write to that location, its a much better solution to have it set to nobody:users, you will also get issues with any windows machine attempting to access a file/folder set with root:root as any windows machine will not know what the root password is set to and thus fail.

 

also it shouldn't matter what user and group you have sab or sickbeard running as, this ONLY effects the arch vm, NOT files/folder on unraid, as this is handled by fstab/autofs and is independent of the running process in the vm, for instance i have sab running as user sabnzbd and group sabnzbd, but autofs still reads/writes as nobody:users to unraid, it will of course read/write to the arch vm though as sabnzbd:sabnzbd

 

i just thought i would post this to clear any confusion, it might help you in the future too perhaps 

[binhex]

hi justinchase, those are strange issues you have there with your incomplete folder, it looks to me like autofs is loosing connectivity to unraid and thus your getting intermittent visibility of files/folders, this is odd as im running autofs and its solid for me.

 

couple of questions for you, do you have this hack in-place, if so can you delete it from your crontab file:-

 

* Some users have reported an issue where this doesn't quite work perfectly for them so in this case setup a cron job (something that automates actions).

 

Code: [select]

crontab -e

--------

* * * * * /bin/bash cd /net/unraid/mnt/user/Media

 

Press Ctrl + X to exit and save the crontab file, then type:

 

Code: [select]

systemctl enable cronie

systemctl start cronie

 

This should for your VM, every minute to 'change directory' to your unRAID host and thus keep the channel open for all the apps that require it.

 

also can you nano in to /etc/autofs/auto.master, it should look like this:-

 

/net -hosts

 

if it has a timeout period or anything else please remote it, then reboot vm and see how autofs is performing for you then.

 

just a quick sanity check, you are running v6b4 yes?, because v6b3 had issues with nfs, just double checking 

 

one last thing to check is from unraid ssh/telnet in and just double check any location your trying to read/write to from the vm has ownership set to nobody:users and that it has rwx for at least the user, also verify that ALL shares are set to export and public.

 

last thing to do is remove any bonding you have going on and physically disconnect that second nic!, you MAY need to reboot unraid in order to get the vm and host straight after this im not sure, but potentially the vm MIGHT be seeing the second nic, and if so this could indeed cause issues, best to keep it simple and attempt bonding once stable.

[Ashe]

if i run ls -l on arch for my network files

 

drwxr-xr-x 2 root root

 

The only difference being is root:root on my arch and yours is nobody:users

 

Did you change sabnzbd.service and sickbeard.service back from root:root in Arch?

 

Ashe, the user and group for files on unraid should be set to nobody:users, if its set to root:root then you will need to include your root password for unraid in fstab/autofs in order for the vm to read/write to that location, its a much better solution to have it set to nobody:users, you will also get issues with any windows machine attempting to access a file/folder set with root:root as any windows machine will not know what the root password is set to and thus fail.

 

also it shouldn't matter what user and group you have sab or sickbeard running as, this ONLY effects the arch vm, NOT files/folder on unraid, as this is handled by fstab/autofs and is independent of the running process in the vm, for instance i have sab running as user sabnzbd and group sabnzbd, but autofs still reads/writes as nobody:users to unraid, it will of course read/write to the arch vm though as sabnzbd:sabnzbd

 

i just thought i would post this to clear any confusion, it might help you in the future too perhaps 

 

Good info Binhex, nice to have a simple explanation of it all.

 

Root password is set in fstab on my setup. It is not accessed from anywhere else and only me that uses it. Windows does not have any issues with accessing via root, not sure how I set this up previously, been this way for a couple of years

 

Sent from my Q5 using Tapatalk

 

 

[binhex]

Root password is set in fstab on my setup.

 

ahh that explains how your able to read/write to unraid for files/folders set as root:root :-).

 

Windows does not have any issues with accessing via root, not sure how I set this up previously, been this way for a couple of years

 

hmm interesting, in that case i can only assume you either mapped to the share and specified the root username and password, or your doing some sort of credentials pass-through from windows by creating a "root" user account on your windows machine with the same password as your unraid server.

 

thats fine having it set like that, and in some ways is more secure as you need to specify the root password to access any files/folders  on unraid

[Ashe]

 

 

mapped to the share and specified the root username and password

 

thats fine having it set like that, and in some ways is more secure as you need to specify the root password to access any files/folders  on unraid

 

thats what i did

[JustinChase]

hi justinchase, those are strange issues you have there with your incomplete folder, it looks to me like autofs is loosing connectivity to unraid and thus your getting intermittent visibility of files/folders, this is odd as im running autofs and its solid for me.

 

couple of questions for you, do you have this hack in-place, if so can you delete it from your crontab file:-

 

* Some users have reported an issue where this doesn't quite work perfectly for them so in this case setup a cron job (something that automates actions).

 

Code: [select]

crontab -e

--------

* * * * * /bin/bash cd /net/unraid/mnt/user/Media

 

Press Ctrl + X to exit and save the crontab file, then type:

 

Code: [select]

systemctl enable cronie

systemctl start cronie

 

This should for your VM, every minute to 'change directory' to your unRAID host and thus keep the channel open for all the apps that require it.

 

I've never set that up intentionally, so I guess I don't have it, but I'll have to dig around to find the crontab file to confirm.

 

also can you nano in to /etc/autofs/auto.master, it should look like this:-

 

/net -hosts

 

if it has a timeout period or anything else please remote it, then reboot vm and see how autofs is performing for you then.

 

you may have found it.  Mine looks like this...

 

/net -hosts --timeout=60

/media/Tower /etc/autofs/auto.Tower --timeout=0

 

I'm not sure how/why mine is different, I don't recall ever changing it 

 

just a quick sanity check, you are running v6b4 yes?, because v6b3 had issues with nfs, just double checking 

v6b5a now, but yes, was on beta4 when this started.  I never ran beta3.

 

one last thing to check is from unraid ssh/telnet in and just double check any location your trying to read/write to from the vm has ownership set to nobody:users and that it has rwx for at least the user, also verify that ALL shares are set to export and public.

 

Looks good to me...

 

root@media:/mnt/user# ls -l
total 200
drwxrwxrwx  1 nobody users    160 May  1 02:15 BIN/
drwxrwxrwx  1 nobody users     96 Jul 20  2013 JRiver/
drwxrwxrwx  1 nobody users     72 Apr 25 09:16 adult/
drwxrwxrwx+ 1 nobody users    288 May  1 02:15 backup/
drwxrwxrwx  1 nobody users    256 May  1 02:15 documents/
drwxrwxrwx  1 nobody users    120 Apr 27 10:27 domains/
drwxrwxrwx  1 nobody users     80 Apr 30 21:42 download/
drwxrwxrwx  1 nobody users 205264 May  1 02:15 music/
drwxrwxrwx  1 nobody users    192 Apr 29 02:16 photos/
drwxrwxrwx  1 nobody users     80 Apr 24 22:44 plugins/
drwxrwxrwx  1 nobody users     96 May  1 08:39 video/

 

root@media:/mnt/user# cd /mnt/cache
root@media:/mnt/cache# ls -l
total 0
drwxrwxrwx 4 nobody users  96 Jul 20  2013 JRiver/
drwxrwxrwx 5 nobody users 120 Apr 27 10:27 domains/
drwxrwxrwx 3 nobody users  80 Apr 30 21:42 download/
drwxrwxrwx 3 nobody users  80 Apr 24 22:44 plugins/
drwxrwxrwx 4 nobody users  96 May  1 08:39 video/

 

I have all NFS shares set to Export & Public, but some of my SMB shares are set to Private, for reasons stated above.  I assume since this is in Arch, that we're only dealing with NFS shares, so I'm not going to change the SMB security just yet.

 

last thing to do is remove any bonding you have going on and physically disconnect that second nic!, you MAY need to reboot unraid in order to get the vm and host straight after this im not sure, but potentially the vm MIGHT be seeing the second nic, and if so this could indeed cause issues, best to keep it simple and attempt bonding once stable.

 

Yesterday, I installed a NIC in this box, to eventually share with the Windows VM I'm working on, which made 3 NIC's in the machine.  2 on the motherboard and the new card.  I had connected the new NIC to my router, and booted, and got all kinds of issues, so I disconnected it and rebooted, and everything (else) went back to 'normal', so I concur that having the extra NIC connected can cause issues.  With that said, I've had unRAID set to use bonding since it became available, over a year ago, so I don't think it's the source of my problems.

 

Regardless, I will disconnect it now, then reboot both the VM and the server, just to be safe.

 

the user and group for files on unraid should be set to nobody:users, if its set to root:root then you will need to include your root password for unraid in fstab/autofs in order for the vm to read/write to that location, its a much better solution to have it set to nobody:users, you will also get issues with any windows machine attempting to access a file/folder set with root:root as any windows machine will not know what the root password is set to and thus fail.

 

also it shouldn't matter what user and group you have sab or sickbeard running as, this ONLY effects the arch vm, NOT files/folder on unraid, as this is handled by fstab/autofs and is independent of the running process in the vm, for instance i have sab running as user sabnzbd and group sabnzbd, but autofs still reads/writes as nobody:users to unraid, it will of course read/write to the arch vm though as sabnzbd:sabnzbd

 

i just thought i would post this to clear any confusion, it might help you in the future too perhaps 

 

Would you suggest I change mine back to nobody:users in both the service files and the folders /opt/sabnzbd & /opt/sickbeard?  I changed them based on Ashe's recommendation a couple of days ago, but it never 'felt right' to have done so, even though it certainly did seem to help, at least for a little bit.

 

Thanks again for taking the time to help!!

[JustinChase]

so, it keeps getting weirder.

 

I created a new share for holding downloads, located at /mnt/user/download/  (notice that it's download, singular)

 

Here you can see my ability to access it via unraid directly, and its permissions...

root@media:~# cd /mnt/user/download
root@media:/mnt/user/download# ls -l
total 4
drwxrwxrwx 1 nobody users 3976 May  1 15:59 incomplete/

 

and here you can see my ability to access it via archVM, and its permissions...

[root@IronicsArchVM_v4 sabnzbd]# cd /net/media/mnt/user/download
[root@IronicsArchVM_v4 download]# ls -l
total 4
drwxrwxrwx 1 nobody users 3976 May  1 15:59 incomplete

 

So, I sent some nzb's to sabnzbd, and got some failures.  Here is a snipit from the log...

2014-04-24 03:21:35,015::ERROR::[misc:372] Cannot create directory /net/media/mnt/user/downloads/incomplete

 

Notice how it's downloads, plural

 

That is the old folder I was using.  I have no idea where it's getting this as the folder to use.

 

In addition, I'm seeing a couple of other problems lie this in the log...

2014-04-24 03:25:54,484::ERROR::[misc:378] dirscan_dir directory: /net/media/mnt/user/backup/newsgroups/nzb error accessing
2014-04-24 03:26:24,832::ERROR::[misc:378] script_dir directory: /opt/sickbeard/autoProcessTV error accessing

 

I changed the watched folder to /net/media/mnt/user/backup/nzb (no longer in the /newsgroups folder)

and I copied the autoProcessTV folder from /opt/sickbeard to /opt/sabnzbd to make sure that there wouldn't be any permissions issues, by having different users assigned for each program.  Notice that it's still trying to use the folder in the sickbeard folder.

 

I verified that the settings are correct in sabnzbd, and they are.  I then checked the sabnzbd.ini directly, and they are correct in the ini file also.

 

I not only restarted sabnzbd, I restarted the VM, and I also restarted unRAID after making these changes, so I just can't figure out where it's getting the wrong file locations to try.

 

I even checked the sabnzbd.ini.bak file, but it's also got the correct locations in there.

 

Any ideas where else to try looking, or what else to change to get sabnzbd to use the correct locations?

 

full error log and ini files are zipped and attached to this post.

 

sabnzbd-log.zip

sabnzbd-ini.zip

[JustinChase]

I shut everything down, and rebooted everything again.

 

Now it seems that sab is at least trying to use the correct locations, so I successfully added a download that previously failed back to the queue.

 

After about 15 or 20 minutes, I noticed that SAB was paused, and there were 3 new errors...

 

2014-05-01 19:55:58,583::ERROR::[misc:372] Cannot create directory /net/media/mnt/user/download/incomplete
2014-05-01 19:57:01,821::ERROR::[__init__:862] Saving /net/media/mnt/user/download/incomplete/house.of.cards.2013.s01e09.1080p.bluray.x264-rovers/__ADMIN__/SABnzbd_nzo_Ee5VhB failed
2014-05-01 19:57:01,830::ERROR::[assembler:100] Disk error on creating file /net/media/mnt/user/download/incomplete/house.of.cards.2013.s01e09.1080p.bluray.x264-rovers/5S7gHtS14qjuL3X2sLr4.r02

 

Then I tried to add a nzb by putting it the watch folder.  SAB grabbed it, but didn't actually add it to the queue.  It gave this error message instead...

 

2014-05-01 20:35:35,394	ERROR:	Failed making (/net/media/mnt/user/download/incomplete/sfisodnfoddpskfmapmfsfaf)

 

Then I tried to delete folders from the incomplete folder from within windows, and got an error message stating "You require permission from Unix User\nobody to make changes to this file

SABnzbd_attrib"

 

I then clicked the red circle to delete the same file from within SAB, and it deleted just fine.

 

So, for some reason SAB thinks it doesn't have rights to write to this folder, but it seems to have more rights than a windows user to delete files here.

 

I think the theory that it's just losing access to shares after a bit of time; for some reason, which I don't understand, seems like the most plausable at this point.

 

I honestly don't know if this is a VM caused problem, an unRAID problem, a SABnzbd problem, a shares problem, or some weird combination that I'm just unlucky enough to get stuck with.

 

Anyone think running the "New Permissions" script from the unRAID GUI will help anything?

[JustinChase]

another clue???

 

[root@IronicsArchVM_v4 incomplete]# ls -l

ls: cannot open directory .: Stale file handle

 

 

I'm not positive this is related, but it's from the IP of the archVM.  From the syslog...

 

May  1 20:17:35 media rpc.mountd[1829]: authenticated unmount request from 192.168.20.145:849 for /mnt/user/download (/mnt/user/download)
May  1 20:17:44 media rpc.mountd[1829]: authenticated mount request from 192.168.20.145:753 for /mnt/user/download (/mnt/user/download)
May  1 20:23:50 media rpc.mountd[1829]: authenticated unmount request from 192.168.20.145:904 for /mnt/user/download (/mnt/user/download)
May  1 20:24:58 media rpc.mountd[1829]: authenticated mount request from 192.168.20.145:924 for /mnt/user/download (/mnt/user/download)
May  1 20:30:05 media rpc.mountd[1829]: authenticated unmount request from 192.168.20.145:947 for /mnt/user/download (/mnt/user/download)
May  1 20:31:40 media rpc.mountd[1829]: authenticated mount request from 192.168.20.145:846 for /mnt/user/download (/mnt/user/download)

[binhex]

hi justinchase, i have had more of a think about your issues and i think your running into a race condition where autofs is NOT started before sab, and thus sab cant see any of the folders you have specified, i too came across this issue and managed to fix the problems using PeterB's post here:-

 

http://lime-technology.com/forum/index.php?topic=31687.msg300532#msg300532

 

basically this creates two services, one attempts a ping to your router (or other device), the other service then attempts to writes a temp file to a autofs mount point, once both of these services are succesful then it allows sabnzbd to start and voila no more errors.

 

i too was seeing multiple "cannot create directory" errors in sab, thing is sab runs fine once the vm has booted and autofs is running so hadn't noticed the errors before, i think this is def worth a go, if you get stuck with this let me know as i have it running.

 

the alternative to this is to try using cifs and fstab, which then bypasses the issue to some extent, although it is still possible you will get issues with sab coming up before the network is ready.

 

as a side note, once you have created the service files you can use them with any application you may be getting a race condition with, this is not specific to sabnzbd.

[MyKroFt]

This is why i dont use autofs - if you mount via fstab and make a cronjob that mount -a ; df -h every 5 min (takes care of stale handles), on the VM chmod all the mount dirs 777 you will not have any probs.

 

You use the default conf files as far as user:group, and the systemctl enable can be used as default

 

You guys are making this way harder than it needs to be....

 

My Sab, Sb & CP have been running happy for almost 2 months now.....

 

Myk

 

[binhex]

cheers for the post MyKroFt!, i might take a look at that as an alternative way of doing things then, i dont overly like the idea of creating additional scripts/services.

 

can you post your fstab file so i can get an idea as to what flags your using?, im assuming your using nfs and not cifs?.

[MyKroFt]

//tower/tv /mnt/TV cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/sort\040thru /mnt/SortThru cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/BD\040Movies /mnt/BDMovies cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/DVD\040Movies /mnt/DVDMovies cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/Music /mnt/Music cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/HD\040Misc /mnt/HDMisc cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/ebooks /mnt/eBooks cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/Pictures /mnt/Pictures cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0

 

cifs

 

I also use a 2nd .img file and its mounted as a local path in the VM for the SB download dir and plex library dir

 

Myk

[JustinChase]

hi justinchase, i have had more of a think about your issues and i think your running into a race condition where autofs is NOT started before sab, and thus sab cant see any of the folders you have specified, i too came across this issue and managed to fix the problems using PeterB's post here:-

 

If it's a race condition as you suspect, I suppose I could just not auto-start SAB with the VM, then manually start it once I see the VM is running.  My plan is to have this running pretty much all the time, so this would really only be a small inconvenience every once in a while.  Plus, hopefully TOM can figure out a way to fix this issue in a future release so that autofs just works as it should.

 

Weird thing happened, after I went to bed last night, one file finished and unpacked and moved itself just fine, and the next file started downloading and seems to have been downloading all night, without issue.  The last error in the log is from around 10pm last night.

 

This is why i dont use autofs - if you mount via fstab and make a cronjob that mount -a ; df -h every 5 min (takes care of stale handles), on the VM chmod all the mount dirs 777 you will not have any probs.

 

I suppose that's an option, but I've never created a cronjob, and I don't suspect your text above is the 'right' thing to try to post into a cronjob, so I'll have to research cronjobs and see if I can figure out how to convert your ideas into an actual script.

 

You use the default conf files as far as user:group, and the systemctl enable can be used as default

 

This doesn't sound like anything different to what I'm doing, which I suppose is good.

 

You guys are making this way harder than it needs to be....

 

Well, that's certainly not my intention!!!!  I'm not a Linux expert, and only know a little, from what I've had to learn since installing unRAID a couple of years ago.

 

My Sab, Sb & CP have been running happy for almost 2 months now.....

 

Myk

 

 

Good to know it can be done

 

//tower/tv /mnt/TV cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/sort\040thru /mnt/SortThru cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/BD\040Movies /mnt/BDMovies cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/DVD\040Movies /mnt/DVDMovies cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/Music /mnt/Music cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/HD\040Misc /mnt/HDMisc cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/ebooks /mnt/eBooks cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0
//tower/Pictures /mnt/Pictures cifs rw,guest,username=nobody,iocharset=utf8,file_mode=0777,dir_mode=0777,noperm 0 0

 

Could this be amended to use NFS instead?  Is there any advantage to one over the other?  I heard IB say in his videos that NFS took less resources, and was more efficient (or something to that effect).

 

I also use a 2nd .img file and its mounted as a local path in the VM for the SB download dir and plex library dir

 

Myk

 

I have no idea what that means

 

Regardless, thanks for all the help.

[MyKroFt]

in my archVM.cfg file

 

disk = [
        'phy:/mnt/appdisk/ArchVM/arch.img,xvda,w',
        'phy:/mnt/appdisk/ArchVM/data.img,xvdb,w'
       ]

 

the data.img is just a 100g image file for storage since the default arch.img is like only 15g if I remember correctly.

 

I am not using my cache drive, I have a 240g ssd mounted outside the array for my VM stuff.

 

this data.img is mounted as the 2nd drive when the vm started, I then mount this drive, and have told sab and plex to use this drive for its database/download stuff as whats left of 15g will not hole alot of sab downloads at one time

 

/dev/xvdb1 on /mnt/data type ext4 (rw,relatime,data=ordered)

 

is my data.img that is mount on /mnt/data for useage

 

in my sab config for folder section I have:

 

Default Base Folder: /opt/sabnzbd

Temporary Download Folder /mnt/data/sab-downloads/incomplete

Completed Download Folder /mnt/data/sab-downloads/complete

Permissions for completed downloads 777

Post-Processing Scripts Folder /opt/sabnzbd/scripts

 

I made the sab-downloads folder to keep them seperate from my plex stuff which is in Library - both are located in /mnt/data

 

also in my VM i did a chmod -R 777 * in my /mnt folder for the unraid mounts

 

[root@IronicsArchVM_v4 mnt]# ls -la
total 12
drwxr-xr-x 12 root root 4096 Apr 21 13:34 .
drwxr-xr-x 19 root root 4096 Feb  4 07:58 ..
drwxrwxrwx  2 root root    0 Jul 21  2013 Adult
drwxrwxrwx  2 root root    0 Apr 28 03:04 BDMovies
drwxrwxrwx  4 root root 4096 Apr 19 00:11 data
drwxrwxrwx  2 root root    0 Apr 20 09:21 DVDMovies
drwxrwxrwx  2 root root    0 Dec 16 22:55 eBooks
drwxrwxrwx  2 root root    0 Sep 20  2013 HDMisc
drwxrwxrwx  2 root root    0 Apr 28 03:04 Music
drwxrwxrwx  2 root root    0 Apr 20 08:46 Pictures
drwxrwxrwx  2 root root    0 Apr 29 03:00 SortThru
drwxrwxrwx  2 root root    0 May  2 04:04 TV

 

in the vm they are owned by root.root, but in unraid they are nobody.users, the extra params in the fstab i posted earlier takes care of the ownership prob

 

this way it does not matter what sab is running as - sabnzbd i think if i remember correctly, and sickbeard as sickbeard also

 

Everything has been running like this for almost 2 months now

 

Myk