Running a Tezos node on the *alphanet* and the *zeronet* made super-simple (with RPI3)


#41

This is what I see when I run the ls -l command. From what you said it appears I have the correct time stamp for the tezos folder. I also definitely performed the tz-switch.sh command to switch to zero net. It seems to be running, but I’m hoping you can help me confirm that.

Thanks!


#42

It only lets me post one picture per post, so here is what I currently see on my screen:


#43

@snoopyrj7

well done, you are in the game…
CONGRATULATIONS!

If your node was still running last night it should have restarted automatically at 1am, maybe check it out. Type ps aux you should a line like this “/home/tzuser/tezos/tezos-node run” with start time 01:00 (or Apr25)


#44

@maxtez-raspbaker

I got it running with the ./tz-start shell script and it also restarted May 1st at 1:00 so I think that is all going smoothly!

However, I keep getting a bunch of “validator.peer(332)” with a hash after it followed by the error code “Unclassified error: Tezos_error_monad__Error_monad.Canceled” text and then followed by “p2p.maintenance: Too few connections (0)”. These several lines repeat the most often. Every once in a while it seems to actually establish a connection with a few peers. What could this be due to? Could I have an erroneous blockchain data? Is there a way I can verify it?

I was also curious as to how updates will be handled in the future. I believe that right now you create a precompiled version of the updates from the tezos github, add it on drop box, and then use the tz-switch script file to unzip the new version and switch in the appropriate files. Would there be a way to create an update script that would go directly to the tezos github, compile the necessary updates and switch them in? Just thinking about the future of this project and how I could possibly make one of these myself if needed.

Thanks!


#45

@snoopyrj7
I see the “unclassified error” too, I am not particularly worried at the moment. My explanation (maybe completely wrong) is that something has changed in the block format with the recent version for the zeronet and some nodes (not ours) are still running an old version

“too few connections(0)” is normal, as long as it happens only from time to time. If you keep the node running long enough, it will establish more and more connections and the chance of getting (0) becomes less likely. Keep in mind also that there are few nodes running on the zeronet.

Updates, good question.
I’ll try to keep supporting Tezos for RPI3 with the precompiled version through all the stages, alphanet->betanet->mainnet, unless at some point I get a red card from the TF. You are right the zip file is just that, basically I save you the trouble of compiling it by yourself.
But if you are planning to run a node on the mainnet you should definitely give it a try, please do it, I can help you out, and please feel free to contribute to the wiki page (it is open access).

To get started with the compilation, in the wiki page there is a link to old notes [a], have a look at the steps
to compile Tezos (second part of the notes), however because they are somehow outdated, before jumping into the real work, have a look also at page [c] (unpolished instructions) and replace whatever is in the old notes[a] with the updated instructions in [c]. I know it is far from ideal, but keep re-writing detailed instructions that would become outdated quickly is not worth the effort right now.

Anyway if you get stuck somewhere in the process let me know…


#46

Hi all –

100% noob here. I’m embarrassed to say that even the super-simple instructions may have been too much for me. I’m hoping that I’m a good test case because at least some other tezos folks will be pretty ignorant when it comes to the RPI3 and baking. Hopefully they can learn from my mistakes.

Here’s what I did:

  1. Bought a RPI3 and fooled around with it for a bit.
  2. Formatted a 16GB SD card.
  3. Used etcher.io to write the recommended 4GB file onto the SD card.
  4. Popped the SD card into my RPI3 and plugged it in.

Every time I have done this – and I’ve done it about 8 times on 3 different SD cards over a few days – the RPI3 starts running, a lot of text passes across the monitor, eventually (after maybe 30 seconds and maybe 100+ lines of information), I get one blue line that says “[something something] Fedora 26”. Then a little bit more script in white, and the screen goes dark. It remains dark and nothing else ever appears. I can leave it alone for a long time and nothing happens. Typing on the keyboard, using the RPI3 touchscreen, and/or clicking on the mouse also have no effect. I don’t see a command prompt or anything else no matter how long I wait.

As I said, I have re-downloaded the 4GB image file at least 3 times, and etched it onto at least 3 different SD cards, and the same thing happens every time. So I doubt it’s the cards.

Has anyone else experienced this? Is there something obvious I’ve screwed up? My best guess is that I shouldn’t have fully formatted the SD card before etching the image file onto it. (I.e. the card should contain something other than just the .img file when I put it in my RPI3.) Second-best guess is that I screwed up using Etcher, which would be impressive but is totally possible. (The appearance of all that text when I boot up the RPI3 cuts against this explanation.)

Any help would be greatly appreciated. Also, although I’m an intelligent human being in matters unrelated to computers/programming, you’re probably best off explaining what I’ve done wrong as you would to your grandmother. :slight_smile:


#47

Hi @misifecu
no worries, this is work in progress for all of us…
If etcher completed the writing process successfully, then the image on your sd card should be ok. The correct booting of the RPI3 ends with the X11 environment (X-windows) with a basic graphic interface and an open terminal shell. My best guess is that there is something going on with your monitor/display. If you have a spare monitor I would try it and repeat the boot.

But before trying anything you should go to the RPI3 wiki (github.com/demodun/tezos-rpi3/wiki) and grab the latest image which contains Fedora28+precompiled Tezos(zeronet network), file: RPI3-TEZOS-050618.img. The setup is the same (etcher etc.), the image is a bit bigger (~7GB) but F28 will be the OS that we’ll rely on when the official tezos(betanet/mainet) will go live, at least for a year or so.
In summary, (1) get the new image RPI3-TEZOS-050618.img, (2) use etcher to write it on the SD card (3) boot the RPI3 (4) if all goes well, type the 3 commands to get your RPI3 on the zeronet (./resize.sh, ./tz-init.sh and ./tz-start.sh) (5) please let me know the outcome


#48

Thanks much for the quick reply. I followed the steps you listed and I’m up and running. Very grateful for the help!


#49

excellent news and congratulations!