L0rdRaiden Posted March 22, 2018 Share Posted March 22, 2018 @Jcloud Where you able to run at the same time more nodes than cpu cores? I read this limitation somewhere Quote Link to comment
Jcloud Posted March 22, 2018 Author Share Posted March 22, 2018 Just now, L0rdRaiden said: Where you able to run at the same time more nodes than cpu cores? I read this limitation somewhere I honestly haven't tried that, but I can tonight, after work (~4hrs), I'm only up to ten nodes -- and I thought that was plenty. Quote Link to comment
NewDisplayName Posted March 22, 2018 Share Posted March 22, 2018 (edited) Look at the plugin user scripts, there u can write and use shell bash e.g. And we may could install them easy. Edited March 22, 2018 by nuhll 1 Quote Link to comment
Jcloud Posted March 23, 2018 Author Share Posted March 23, 2018 (edited) 4 hours ago, Jcloud said: 4 hours ago, L0rdRaiden said: Where you able to run at the same time more nodes than cpu cores? I read this limitation somewhere I honestly haven't tried that, but I can tonight, after work (~4hrs), I'm only up to ten nodes -- and I thought that was plenty. Confirmed. You can not start more nodes than one has system cores. root@HYDRA:/mnt/user/storj# docker exec Storj storjshare start --config /storj/Node_32/config.json failed to start node, reason: insufficient system resources available I also took this experiment to watch the Container in my Netdata Docker Container. With each additional farming node the CPU usage stays the same, but each node adds ~50MB to allocated memory. Edited March 23, 2018 by Jcloud grammar atk. Quote Link to comment
L0rdRaiden Posted March 23, 2018 Share Posted March 23, 2018 @Jcloud First try, no luck. I have move my old storjshare to Node1, and I have creates 4 more nodes/folders. Then I run this command root@MediaCenter:~# docker exec Storj storjshare-create --storj ******************* --storage /Storj/Node2 --size 200GB --outfile /Storj/Node2/config.json --rpcaddress 0.0.0.0 --rpcport 4004 --manualforwarding true --logdir /Storj/Node2 failed to write config, reason: ENOENT: no such file or directory, open '/Storj/Node2/config.json' Why? I think I found a solution here https://github.com/Storj/storjshare-daemon/issues/122 But I don't know where should I create the folder they mention. I would appreciate if you can do a guide for dummies On the other hand how do you have configured the docker template? everytime I run it even after setting up the script it run a node at the level of /Storj and doesn't even run my original storjshare in Node1 folder For the script this is the setup (image) Quote Link to comment
Jcloud Posted March 23, 2018 Author Share Posted March 23, 2018 (edited) 1 hour ago, L0rdRaiden said: I have move my old storjshare to Node1 No don't do that, just leave your current share in original folder think of it as "Node0" and just start/make node1. 1 hour ago, L0rdRaiden said: failed to write config, reason: ENOENT: no such file or directory, open '/Storj/Node2/config.json' Linux is CaSe SENsiTIVe, so are storj is lower case (on mine at least), and is Node2 folder capitalized on your system? If so, it's okay, if not that needs to change. I'm assuming since you moved your first node, that in your /user/mnt/storj location (unRAID user share for storj) you ran mkdir Node2 to make the new folder, before running the really long create command. Also, I saw your -rpcaddress is 0.0.0.0 if you're using dynamicDNS you can remove the "0.0.0.0" and put in your dDNS address -- if you're drawing a blank on this, don't worry about it for now. 1 hour ago, L0rdRaiden said: On the other hand how do you have configured the docker template? At work, can't make the answer now. I'll do screen shots later tonight and edit this post. Quote I would appreciate if you can do a guide for dummies I'll see what this weekend/ next week looks like. Edited March 23, 2018 by Jcloud Quote Link to comment
L0rdRaiden Posted March 23, 2018 Share Posted March 23, 2018 2 hours ago, Jcloud said: No don't do that, just leave your current share in original folder think of it as "Node0" and just start/make node1. Linux is CaSe SENsiTIVe, so are storj is lower case (on mine at least), and is Node2 folder capitalized on your system? If so, it's okay, if not that needs to change. I'm assuming since you moved your first node, that in your /user/mnt/storj location (unRAID user share for storj) you ran mkdir Node2 to make the new folder, before running the really long create command. Also, I saw your -rpcaddress is 0.0.0.0 if you're using dynamicDNS you can remove the "0.0.0.0" and put in your dDNS address -- if you're drawing a blank on this, don't worry about it for now. At work, can't make the answer now. I'll do screen shots later tonight and edit this post. I'll see what this weekend/ next week looks like. Ok I will leave as "Node0" Regarding the sensitive case I think is fine, I have taken that into account the nodes are really under \Storj\Node2 RPCaddress is fine I just replaced it by 0-0-0-0 to post it here, I'm using duckdns Do I have to create the folders in CLI? I just created them from windows throug samba, I will try. Quote Link to comment
Jcloud Posted March 23, 2018 Author Share Posted March 23, 2018 7 minutes ago, L0rdRaiden said: Do I have to create the folders in CLI? I just created them from windows throug samba, I will try. Hrmm perhaps run the Fix file permissions on webui (under tools I think), perhaps permissions are off because of using samba versus what Docker container wants. Couldn't hurt (making a pot-shot in the dark here) Quote Link to comment
L0rdRaiden Posted March 29, 2018 Share Posted March 29, 2018 If I run the comand below in the screenshot I get the output above the command Every time I run the command it writes a new Json file but I don't know where it is stored Every time I run the command to create a new node I get this error The path are fine, is case sensitive, I don't now what could be failing, I have found this https://github.com/Storj/storjshare-daemon/issues/122 but I don't know how to use it to fix this issue Any idea? Quote Link to comment
NewDisplayName Posted March 29, 2018 Share Posted March 29, 2018 Cool, now i have full reputation (5000) and already 3gb. First it was 100mb, then some time 300 then sometime 900, now one day, 3gb... :=) I wonder how much they pay. reputation seems to be important (24/7 online, good connection, 5000 is max) 1 Quote Link to comment
Prokie Posted March 29, 2018 Share Posted March 29, 2018 Should I only portfordard 4000-4003 in my router or should I do it in unraid aswell? Quote Link to comment
NewDisplayName Posted March 29, 2018 Share Posted March 29, 2018 (edited) u port forward from router to unraid ip and in unraid to docker, or if u use standalone ip for docker, then to the standalone ip, depends on your setup. With https://storjstat.com/index.asp you can check if port is open (click on your ip or DNS) and you can see some nice stats. You diont need to install anything for it just search for your node id. Edited March 29, 2018 by nuhll 1 Quote Link to comment
Prokie Posted March 29, 2018 Share Posted March 29, 2018 24 minutes ago, nuhll said: u port forward from router to unraid ip and in unraid to docker, or if u use standalone ip for docker, then to the standalone ip, depends on your setup. With https://storjstat.com/index.asp you can check if port is open (click on your ip or DNS) and you can see some nice stats. You diont need to install anything for it just search for your node id. Okay thanks it seems to be working. Is it worth it trying to use multiple nodes or should I stick with one? Quote Link to comment
MrChunky Posted March 29, 2018 Share Posted March 29, 2018 @nuhll How long did you run for in total to get to 5000? Do you guys think auto update should be off, since the server will shut down for the update and hence will not be up 24/7? How would that affect updates in general? Quote Link to comment
NewDisplayName Posted March 29, 2018 Share Posted March 29, 2018 (edited) No, update is important! Docker gets updated like every day, np. (they also count who use old images, so i guess, its a quality factor!? Or might be at some time?!) I would go with one node and when its easy to do sometime, then add more, or if one node is full. Currently iam nearly at 4gb, so i guess a good day 24.2 is my first post in this thread, so i guess around that date. You get some points for every X online. (i dont know X) Edited March 29, 2018 by nuhll 1 Quote Link to comment
Prokie Posted March 29, 2018 Share Posted March 29, 2018 I am trying out multiple nodes, is there a way I can configure them to automatically start after an auto update to the storj docker? Quote Link to comment
Jcloud Posted March 29, 2018 Author Share Posted March 29, 2018 (edited) 1 hour ago, nuhll said: 7 hours ago, L0rdRaiden said: Every time I run the command to create a new node I get this error adfsafds Arg forums and cutting/pasting. @L0rdRaiden check that command line, Linux is case sensitive (Sorry, you know that) also it's the the path with respect to *inside* the container so try changing: --storage /mnt/user/Storj/Node_1 to: --storage /storj/Node_1 This will make it the config file go to (in unRAID host) to: /mnt/user/storj/Node_1 I know, sorry, it's command-line version of "Inception" (the movie; or 13th Floor for you purists ) Edited March 29, 2018 by Jcloud Quote Link to comment
Jcloud Posted March 29, 2018 Author Share Posted March 29, 2018 28 minutes ago, Prokie said: I am trying out multiple nodes, is there a way I can configure them to automatically start after an auto update to the storj docker? I've been trying to work on this too, I don't have a great working solution yet. My first attempt at doing this (seems to work about 50%) is outlined here 1 Quote Link to comment
Prokie Posted March 29, 2018 Share Posted March 29, 2018 2 minutes ago, Jcloud said: I've been trying to work on this too, I don't have a great working solution yet. My first attempt at doing this (seems to work about 50%) is outlined here Yeah I saw it earlier after I wrote. I will try your solution, thanks! 1 Quote Link to comment
Jcloud Posted March 29, 2018 Author Share Posted March 29, 2018 (edited) 1 hour ago, Prokie said: Yeah I saw it earlier after I wrote. I will try your solution, thanks! Thinking about this at work, since I've been away from box for a few days, I think this works for when one updates the template, but doesn't actually work for the automatic updates. Which I think is why I thought, "it worked 50% of the time." So, my mistake. I think I was looking at CA userScript plugin, making a "stupid" scheduled task, about 10 minutes after the update Window, which would kick off loading the multiple nodes. Edited March 29, 2018 by Jcloud Quote Link to comment
NewDisplayName Posted March 30, 2018 Share Posted March 30, 2018 There are news, nearly 5gb (4,95). And i got my first payment (1,24usd worth) - but this must be the payment for i dont know 300mb, maybe. Quote Link to comment
Prokie Posted March 30, 2018 Share Posted March 30, 2018 (edited) 14 hours ago, Jcloud said: Thinking about this at work, since I've been away from box for a few days, I think this works for when one updates the template, but doesn't actually work for the automatic updates. Which I think is why I thought, "it worked 50% of the time." So, my mistake. I think I was looking at CA userScript plugin, making a "stupid" scheduled task, about 10 minutes after the update Window, which would kick off loading the multiple nodes. Yeah I am doing the "stupid" script now aswell, but there has to be a better solution.I will continue to research. Edited March 30, 2018 by Prokie Quote Link to comment
NewDisplayName Posted March 30, 2018 Share Posted March 30, 2018 I dont get it, why its not possible to make a docker per node? Quote Link to comment
L0rdRaiden Posted March 30, 2018 Share Posted March 30, 2018 (edited) I hope you can get this work for dummies, like everything configurable from the docker template. I have tried to troubleshoot the folder error but nothing works. Just running 1 node. Edited March 30, 2018 by L0rdRaiden Quote Link to comment
NewDisplayName Posted March 30, 2018 Share Posted March 30, 2018 (edited) Cant we just add another docker container with a different name/folder? Edit: tried it, works just perfect, didnt changed ports, will monitor if that works, but besides this... i changed my one node (1TB) to 2 nodes with 100GB each. root@Unraid-Server:~# docker exec Storj1 storjshare status ┌─────────────────────────────────────────────┬─────────┬──────────┬──────────┬─────────┬───────────────┬─────────┬──────────┬───────────┬──────────────┐ │ Node │ Status │ Uptime │ Restarts │ Peers │ Allocs │ Delta │ Port │ Shared │ Bridges │ ├─────────────────────────────────────────────┼─────────┼──────────┼──────────┼─────────┼───────────────┼─────────┼──────────┼───────────┼──────────────┤ │ 5e2ce75d9e5d679d125eb2a2228c987f2f236a61 │ running │ 36s │ 0 │ 95 │ 0 │ 15ms │ 4000 │ ... │ connected │ │ → /storj/share │ │ │ │ │ 0 received │ │ (TCP) │ (...%) │ │ └─────────────────────────────────────────────┴─────────┴──────────┴──────────┴─────────┴───────────────┴─────────┴──────────┴───────────┴──────────────┘ root@Unraid-Server:~# docker exec Storj storjshare status ┌─────────────────────────────────────────────┬─────────┬──────────┬──────────┬─────────┬───────────────┬─────────┬──────────┬───────────┬──────────────┐ │ Node │ Status │ Uptime │ Restarts │ Peers │ Allocs │ Delta │ Port │ Shared │ Bridges │ ├─────────────────────────────────────────────┼─────────┼──────────┼──────────┼─────────┼───────────────┼─────────┼──────────┼───────────┼──────────────┤ │ 1d065de2cb7e749bcb6415f4b4f555e8b475a5d0 │ running │ 8m 7s │ 0 │ 161 │ 1 │ 21ms │ 4000 │ 4.96GB │ connected │ │ → /storj/share │ │ │ │ │ 0 received │ │ (TCP) │ (5%) │ │ └─────────────────────────────────────────────┴─────────┴──────────┴──────────┴─────────┴──────────── Edited March 30, 2018 by nuhll 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.