[Support] binhex - rTorrentVPN

[MyKroFt]

ok, if I run rtorrent.sh reload and not try and put it in the background during a docker bash shell - it restarts with current session data intact

 

but if I try and run ./rtorrent.sh reload &

 

it stops for some reason

 

Myk

 

[binhex]

ok, if I run rtorrent.sh reload and not try and put it in the background during a docker bash shell - it restarts with current session data intact

 

but if I try and run ./rtorrent.sh reload &

 

it stops for some reason

 

Myk

 

not sure where your getting the reload parameter from?, but in any case the script handles all the starting of rtorrent should the process die for you. it will auto start the rtorrent process every 10 mins if the process is not running.

 

just to be clear i am not seeing 100% uptime for rtorrent, it def is a bit flaky, if i add a torrent i will see high cpu spikes and sometimes the rtorrent process will just die, (causing it to be restarted, see above comment). ive spent a LOT of time trying to debug this and in the end come to conclusion rtorrent just is a bit crap when it comes to stability, ive even gone as far as trying other docker images to see if its something i may of misconfigured, but im seeing the same sort of thing with other rtorrent dockers too.

 

now it is possible the issue im seeing is related to a btrfs bug, docker bug and/or bug in the kernel for unraid 6.1.9, so im not saying ive given up completely, and its possible moving to 6.2 may improve things, but as i dont have a test rig ive got to wait until this at least goes to RC before i will touch it, feel free to try 6.2 beta and let me know if it improves things if your feeling brave :-).

 

for now deluge seems rock solid, and always has been for me, i honestly cant remember seeing it crash, so if your looking for a more solid solution i would recommend that for now, at least until i get this issue resolved (assuming it can be?!)

[MyKroFt]

I have used Deluge, but like the web interface of rutorrent 100% better - am willing to put up with its problems since it seems to be the program instead of the docker itself.

 

 

Thanks for looking

 

 

Myk

 

 

[MyKroFt]

any know of a option to add to the .rc file to tell rtorrent to log to a file?  Am still trying to track down the crashing and need more info.

 

 

After a crash and waiting for it to restart and fail or restarting the docker and it fails a crashes right away - IF I manualy start rtorrent not in a tmux it starts right up and continues where everything left off.

 

 

So I am going to ASSUME one of the command line parameters is getting corrupted and rtorrent dont like it - but again and trying to get more log info

 

 

Thanks

Myk

 

 

[noobraider]

I am trying to move from deluge (standard) to rtorrentvpn as I've recently signed up with PIA for torrenting. Deluge's scheduler plugin has always seemed lackluster which is why I want to give rtorrent a try...

 

I am able to get the docker started and the VPN connection appears to come up and I can add/remove torrents... however the configuration doesn't appear to be saving certain values. Every time it starts I need to configure the Download and Upload bandwidth limits.

 

Once in a while trying to apply new configuration settings I am seeing:

 

"error,getsettings XMLRPC call Failed"

 

Values like the scheduler and download/automove paths appear to be saving, it's just other values are not.

[noobraider]

I thought I had this figured this out on my own... I thought perhaps certain settings like this had to be modified directly in rtorrent.rc for it to actually stick... I was wrong even doing the following in rtorrent.rc still causes rtorrent to set its global down down and up to "102400" when the docker restarts:

 

# Global upload and download rate in KiB. "0" for unlimited.

download_rate = 48000

upload_rate = 150

 

Also other settings that I initially thought were sticking, like download folder and auto-move when completed directory also seems to be resetting... which is very frustrating as I can't figure out why.

 

Edit: Nevermind. It was a combination of permission issues on rtorrent.rc, bad values being set, and the scheduler plugin overriding my settings. I think I'm finally good and using rtorrentvpn now! Thanks binhex!

[binhex]

I thought I had this figured this out on my own... I thought perhaps certain settings like this had to be modified directly in rtorrent.rc for it to actually stick... I was wrong even doing the following in rtorrent.rc still causes rtorrent to set its global down down and up to "102400" when the docker restarts:

 

# Global upload and download rate in KiB. "0" for unlimited.

download_rate = 48000

upload_rate = 150

 

Also other settings that I initially thought were sticking, like download folder and auto-move when completed directory also seems to be resetting... which is very frustrating as I can't figure out why.

 

Edit: Nevermind. It was a combination of permission issues on rtorrent.rc, bad values being set, and the scheduler plugin overriding my settings. I think I'm finally good and using rtorrentvpn now! Thanks binhex!

 

excellent, glad you got it sorted, can i ask a favour, if your running unraid 6.2 beta can you let me know if you are seeing any issues connecting to rtorrent via the rutorrent webui when adding a torrent, im seeing high cpu load issues and timeouts because of it when running unraid 6.1.9, just wondered if this might be fixed by a later build of unraid (diff kernel, docker etc).

 

also just to confirm for anybody else reading this, you are correct about your earlier assumption, settings are split into two, settings for rutorrent and settings for rtorrent, just because you can set values for rtorrent via rutorrent webui does NOT necessarily mean they are permanent, ONLY settings for ruttorent plugins or webui itself are persistent via the webui, all other settings for rtorrent have to be done via the rtorrent.rc config file, this is just how it is. 

[MyKroFt]

The spike/timeout happens with the latest beta as well

[binhex]

The spike/timeout happens with the latest beta as well

 

thank you for confirming this, i kinda wish you hadn't :-) oh well back to the drawing board on the high cpu issue then. if anybody out there is running this container and is NOT using btrfs i would be interested to hear about their experience, i have heard rtorrent is quite sensitive to the fs used, ext3 seems to be the fs of choice for this app, just wondered if maybe the overhead of btrfs is causing a problem?.

[MyKroFt]

Where I see the spike/lag most of the time is when rtorrent is moving the download to its completed location - ex from incomplete to complete.  I need to get a bigger SSD 1st - but am gonna try and move the incomplete/complete to the SSD cache drive

 

Myk

 

[MyKroFt]

Modification Suggestion

 

 

I changed the sleep variable in rtorrent.sh to .25 - since it likes to crash alot with queue errors - this way it dont take 10 min to come back

 

 

Thanks

Myk

 

 

[binhex]

Modification Suggestion

 

 

I changed the sleep variable in rtorrent.sh to .25 - since it likes to crash alot with queue errors - this way it dont take 10 min to come back

 

 

Thanks

Myk

 

i wouldnt advise this change, this means your going to be hitting pia's api every 15 seconds to check the incoming port, this in itself can cause issues as they might block you and thus cause issues when attempting to re-established a incoming port.

[roppy84]

I have spent the last few days trying to get the webui to pull up. I set up a VPN service with PIA, I followed the guide on http://lime-technology.com/forum/index.php?topic=45812.0 on how to fill in the fields including finding my LAN_NETWORK. At this point if when I try t go the Webui ruTorrent comes up 50% of the time but once up I get these:

[04.07.2016 13:25:02] WebUI started.

[04.07.2016 13:25:23] Bad response from server: (0 [error,getplugins])

[04.07.2016 13:25:24] Bad response from server: (0 [error,getuisettings])

 

Now if I go in and edit the docker and turn off the VPN everything works great...

 

I am a complete noob and for the past month have been learning as I go. I have finally hit a wall I do not know what is going on.

 

Thank you for your help.

 

rutorrent.txt

[MyKroFt]

Modification Suggestion

 

 

I changed the sleep variable in rtorrent.sh to .25 - since it likes to crash alot with queue errors - this way it dont take 10 min to come back

 

 

Thanks

Myk

 

i wouldnt advise this change, this means your going to be hitting pia's api every 15 seconds to check the incoming port, this in itself can cause issues as they might block you and thus cause issues when attempting to re-established a incoming port.

 

Am using AirVPN - no issues so far - but if I leave it at 10 min - sometimes it will take 40 min to start as it crashes upon loading the session directory.....

its a known problem - but they have not fixed it yet

 

Myk

[binhex]

Modification Suggestion

 

 

I changed the sleep variable in rtorrent.sh to .25 - since it likes to crash alot with queue errors - this way it dont take 10 min to come back

 

 

Thanks

Myk

 

i wouldnt advise this change, this means your going to be hitting pia's api every 15 seconds to check the incoming port, this in itself can cause issues as they might block you and thus cause issues when attempting to re-established a incoming port.

 

Am using AirVPN - no issues so far - but if I leave it at 10 min - sometimes it will take 40 min to start as it crashes upon loading the session directory.....

its a known problem - but they have not fixed it yet

 

Myk

 

ahh right, ok fair enough if your not using pia then the port forward api call doesnt come into play so thats fine.

 

yes there are a number of fairly major issues with rtorrent, just have a look at the github issues list, its quite long, really hoping they do fix these, for me right now ive switched back to deluge as its always been very solid for me, shame as i love the ui and dl speed rutorrent+rtorrent gives you but uptime is more important to me.

[MyKroFt]

Ok, so yesterday I had a aprox 6 hr power outage while I was at work - UPS did its job and cleanly shutdown unRAID....

 

Question, when the container is told to stop, how does it do it.

 

All my session data was corrupted for rTorrent and I had to wipe it and manually readd the torrents.

 

Does docker send a unit style shutdown and wait for it to acknowledge that shutdown - or does it just kill it?

 

It seems to just kill it and that when I have session data problems

 

If I bash into the container and manually run rtorrent, and shut it down with ctrl-q - it does not shut down instant.  It finished sending any blocks out that it is working on and then shuts down.....

 

Ideas?

 

Thanks

Myk

[MyKroFt]

ok, according to this I found

 

 

When you issue a docker stop command Docker will first ask nicely for the process to stop and if it doesn't comply within 10 seconds it 
will forcibly kill it. If you've ever issued a docker stop and had to wait 10 seconds for the command to return you've seen this in action

 

 

Looks like this container needs more time if it is busy (mine is seeding 100% of the time) to finish the in-progress blocks.

 

 

Don't think this is something you can do anything about,  Will have to post over in the main docker section and see what they say

 

 

Myk

 

 

[binhex]

ok, according to this I found

 

 

When you issue a docker stop command Docker will first ask nicely for the process to stop and if it doesn't comply within 10 seconds it 
will forcibly kill it. If you've ever issued a docker stop and had to wait 10 seconds for the command to return you've seen this in action

 

 

Looks like this container needs more time if it is busy (mine is seeding 100% of the time) to finish the in-progress blocks.

 

 

Don't think this is something you can do anything about,  Will have to post over in the main docker section and see what they say

 

 

Myk

 

no thats external to the container, tbh unraid itself might do something similar as well on a shutdown, possibly waiting to see if a docker stop command stops all container nicely, and if all docker container are not stopped within X then doing a kill of the docker daemon process, this is a guess but i would assume this is probably the case to prevent "stuck" shutdowns, which was the bane of my life back when plugins were prolific in unraid.

[LaserAllan]

Hey there, first of all i wanna thank you allot for all the hard work you've done here.

 

But for some weird reason I've run into some problems I am not sure how to solve.

First of all, when I did a restart of the host system that runs the docker I get this error when i try to launch the container:

 

2016/07/07 12:11:20 [crit] 804#804: *111 open() "/var/lib/nginx/fastcgi/7/00/0000000007" failed (13: Permission denied) while reading upstream, client: 192.168.1.123, server: localhost, request: "GET /php/getplugins.php HTTP/1.1", upstream: "fastcgi://127.0.0.1:7777", host: "192.168.1.215:9080", referrer: "http://192.168.1.215:9080/"

 

I can't seem to be able to access the webui from another computer than the localhost from the actual machine who runs the container. It used to work though but not anymore.

 

Any idea how i should fix this?

 

Anyway, thanks allot for the work your doing

Awesome!

 

Oh, and I also seem to not have the "Tracker" plugin on the left bar.

 

 

[binhex]

Hey there, first of all i wanna thank you allot for all the hard work you've done here.

 

But for some weird reason I've run into some problems I am not sure how to solve.

First of all, when I did a restart of the host system that runs the docker I get this error when i try to launch the container:

 

2016/07/07 12:11:20 [crit] 804#804: *111 open() "/var/lib/nginx/fastcgi/7/00/0000000007" failed (13: Permission denied) while reading upstream, client: 192.168.1.123, server: localhost, request: "GET /php/getplugins.php HTTP/1.1", upstream: "fastcgi://127.0.0.1:7777", host: "192.168.1.215:9080", referrer: "http://192.168.1.215:9080/"

 

I can't seem to be able to access the webui from another computer than the localhost from the actual machine who runs the container. It used to work though but not anymore.

 

Any idea how i should fix this?

 

Anyway, thanks allot for the work your doing

Awesome!

 

Oh, and I also seem to not have the "Tracker" plugin on the left bar.

 

ok can you post your supervisord.log file (scan and remove username and passwords) , also post a screenshot of your docker container configuration (minus username and password), or if your not using unraid then your docker run command.

[LaserAllan]

Where can i find the supervisord.log file?

 

Docker run command is as follows:

 

 

 

docker run -d \

    --cap-add=NET_ADMIN \

    -p 9080:9080 \

    -p 9443:9443 \

    -p 8118:8118 \

    --name=torr \

    -v /mnt/a:/data \

    -e VPN_ENABLED=yes \

    -e VPN_USER=myusername \

    -e VPN_PASS=mypassword \

    -e VPN_REMOTE=nl.privateinternetaccess.com \

    -e VPN_PORT=1194 \

    -e VPN_PROTOCOL=udp \

    -e VPN_PROV=pia \

    -e ENABLE_PRIVOXY=yes \

    -e LAN_NETWORK=192.168.1.0/24 \

    -e DEBUG=false \

    -e PHP_TZ=UTC \

    -e PUID=0 \

    -e PGID=0 \

    binhex/arch-rtorrentvpn

[MyKroFt]

Where can i find the supervisord.log file?

 

Docker run command is as follows:

 

 

 

docker run -d \

    --cap-add=NET_ADMIN \

    -p 9080:9080 \

    -p 9443:9443 \

    -p 8118:8118 \

    --name=torr \

    -v /mnt/a:/data \

    -e VPN_ENABLED=yes \

    -e VPN_USER=myusername \

    -e VPN_PASS=mypassword \

    -e VPN_REMOTE=nl.privateinternetaccess.com \

    -e VPN_PORT=1194 \

    -e VPN_PROTOCOL=udp \

    -e VPN_PROV=pia \

    -e ENABLE_PRIVOXY=yes \

    -e LAN_NETWORK=192.168.1.0/24 \

    -e DEBUG=false \

    -e PHP_TZ=UTC \

    -e PUID=0 \

    -e PGID=0 \

    binhex/arch-rtorrentvpn

 

there should be a -v config= path\

 

where the docker configuration file are kept

 

Myk

[LaserAllan]

Where can i find the supervisord.log file?

 

Docker run command is as follows:

 

 

 

docker run -d \

    --cap-add=NET_ADMIN \

    -p 9080:9080 \

    -p 9443:9443 \

    -p 8118:8118 \

    --name=torr \

    -v /mnt/a:/data \

    -e VPN_ENABLED=yes \

    -e VPN_USER=myusername \

    -e VPN_PASS=mypassword \

    -e VPN_REMOTE=nl.privateinternetaccess.com \

    -e VPN_PORT=1194 \

    -e VPN_PROTOCOL=udp \

    -e VPN_PROV=pia \

    -e ENABLE_PRIVOXY=yes \

    -e LAN_NETWORK=192.168.1.0/24 \

    -e DEBUG=false \

    -e PHP_TZ=UTC \

    -e PUID=0 \

    -e PGID=0 \

    binhex/arch-rtorrentvpn

 

there should be a -v config= path\

 

where the docker configuration file are kept

 

Myk

 

 

So I should add that then?

It's worked so far apart from that i dont have the tracker labels on the left and now that it for some reason doesn't accept any connections.

 

Here's a link on the pastebin of the information i can find in the logs when i start the container:

http://pastebin.com/ADF4VWEj

 

Anyway, I am not sure what path to specify for the config files then. Would be helpful to know ehre the supervisor log is.

[MyKroFt]

I have it defined here on my unRAID set - its stores the log(s) - rtorrent and nginx config files, rutorrent plugin files and rtorrent session stuff

 

 

Myk

 

 

[LaserAllan]

I have it defined here on my unRAID set - its stores the log(s) - rtorrent and nginx config files, rutorrent plugin files and rtorrent session stuff

 

 

Myk

 

Ok, it seems like it found 9 supervisord.log files, not sure which one I should look so I will have to look in all of them and paste it here.

Could you just give me a sample path that i can use?, becuase I guess I can use any path on the docker container drive?

 

binhex:

http://pastebin.com/5dvmZKUD

here is the Supervisor.log file, I grabbed the entries from today.

 

Thanks allot for your hard work btw.