This is probably fixed in commit 484c860
No, no, this is awesome. I copy-pasted your post to GPT-5, and it fixed it in commit 723ff58
In the meantime I found GPT5 to be the mosst comprehensive, thorough, and require less handholding than any other model my huge margin, and it uses tokens sparingly.
I attempted to have Gemini Pro fix the issue, it wasted $20 worth of tokens and did not fix anything. GPT-5 found and fix the problem for $0.35
I"m officially done with Google.
4 Likes
After last changes server works well. However I have question/doubt about meaning of some values now 
Traffic by Satellite (10:28 PM - 11:28 PM),Traffic by Satellite (10:32 PM - 11:32 PM)To me this name suggest that those are numbers (counts, averages) for last hour. It looks ~OK for nodes (my estimation), speed looks ~OK for aggregare, but other values (Upload/download success, satelite values) are just growing and are >> than sum of nodes.Operation Latency Analytics (11:12 PM - 11:32 PM)- on range 1h it’s quite off. Start of interval == top time on slowest operations (checked several times. For other ranges. Values (timestamps) inslowest operationsare as well strange (I somehow cannot believe that having 24h range all slow times are in 11:25PM-11:33PM)
You inspired me to gather my courage and start active migration of one node. Now my times are not all 0 
Hey, repo isn’t available anymore?
Yeah I don’t know why it’s down - I can’t access it either @arrogantrabbit ?
I wanted to install the dashboard right now, where is the giuthub repo? It’s gone 
It seems @arrogantrabbit went incognito and took the repo along for the ride. He’s probably just busy with other stuff and will return shortly
Knowing mr. opinionated leporidae, there is a non-zero chance that he built a tracker into the repository, and is typing up a lengthy post about not running stuff from the interwebs nilly willy, without going through the source code
1 Like
We will see, if there is a post on his blog one day
Can you re-upload the files for this project to github so we can get them?
Maybe there was a reason why it got taken down? Will never know it, if he doesn’t write it here. But quite sad to see it disappear.
Sorry to say, but I’m afraid it will not be fair. Even arrogantrabbit always refuse to be author, saying it was fully AI, he was the person that invested time, money and effort to create it and publish. And for whatever reason he decided to withdraw it. It is his right and I’ll honor it.
2 Likes
It looks like, he also disappeared. Github is private, Storj account is private, looks like the forum “lost” one of it’s “rabbits”.
A great shame. I do hope he is well.
I agree. That’s the reason I told him, that he should put his name and/or GitHub in the fooder. He spent a great amount of money to make the dashboard.
And it would not be fair to claim his work as ours. And that would be the effect if someone would put a clone online
2 Likes
>:(
Come back, boastfulbunny. My world needs a contender with a firm look on the world, valid points and the technical prowess to back up his arguments to discuss things with.
5 Likes
I rarely weigh in on meta-topics here, but I’m stepping in because some replies feel like they are over-justifying the author’s choices. There are a few plain good-conduct points that matter when a project invites public participation, and they shouldn’t be brushed aside.
First, credit where it’s due: the dashboard had real value. Many of us tested it, filed bugs, submitted PRs, and integrated it locally. That collaboration produced tangible improvements.
The core issue is different: the repository was made private without any status note after public promotion and after the community invested time (issues, PRs, setups). This isn’t a question of legality - of course the author can take the repo private. It’s about maintainer hygiene and respect for contributors time. Once you solicit feedback and merge community effort, you take on a minimal duty of care when stepping away: tell people what changed and what they should do next. That’s not piling on - it’s the baseline that keeps collaborative work healthy.
Common reactions in the thread don’t change that point:
- Maybe he’ll be back. Maybe - but users are blocked right now.
- It’s his work and his decision. True - rights aren’t in dispute; the communication gap is.
- Hope he’s okay. Same here - and a two-sentence update is fully compatible with that concern.
This isn’t a personal attack. It’s calling out the behavior: exiting like this is unprofessional and undermines trust in the project. Nobody expects lifetime support; people do expect a minimal wrap-up.
What would have prevented the frustration (and would still help now):
- A short status note such as “on hold”, “discontinued”, or “coming back later”.
- A read-only tag or release of the last public state, so existing setups have a reference point.
- If there’s a specific reason not to republish (security, policy, personal), say it plainly - people will respect a clear boundary.
What the community can do now:
- Ask once, clearly and respectfully for a status update.
- If no snapshot appears, avoid reposting old code.
- Support and improve alternative, openly available dashboards so monitoring doesn’t hinge on the fate of one project.
In the open-source project community, good manners are simple: set expectations, manage changes, and exit cleanly when others have already invested time. A minimal update along the lines above would turn a frustrating situation into a respectful hand-off. Either way - thanks for the initial push and the ideas it sparked.
1 Like
I have a private fork of his project. He used the “CC0 1.0 Universal” license for his code. So this would be legit to continue to use his code. But it feels for me a bit like stealing his code. I dont know what others think of it.
Ps: for anyone using it currently, please pay attention to memory usage, because this will grow over time and amount of data.
Indeed, there’s a memory leak of unclosed allocations. Which, in my case, amounts to about 12GB/node per week.
So re-start at your leisure.
2 cents,
Julio
Fellas, thank you for the kind words. I had to abruptly step away from tinkering with hardware, scripting, and arguing the obvious on the forum, among other things, for a while, for personal reasons.
@kosti11, I did not expect that merely sharing a curious abomination could be construed by some as “soliciting feedback”, let alone perceived as an “open source project", or worse, wasting anyone’s time, let alone emcombering myself with any obligations. It’s not a “project”. It’s an amusing experiment. It’s not somethign you expect anyone to maintain for any period of time. It’s a one-shot disposable crap. Anyone can feed the Stoagenode log to LLM and ask it to create a parser and a web page. And it will be better than the last one, becase LLMs and tools evolve all the time. But sure, here is the last version from a few weeks ago:
https://link.storjshare.io/s/jwrgii34zri55urbbkiontfkgsja/pile
Feel free to adopt, enhance, abandon. Tell another ai – hey, look at this, rewrite it, make it better.
Keep this in mind:
IMPORTANT WARNING
- This repository is 100% AI-generated.
- No human reviewed, edited, or otherwise contributed to the code or documentation.
- Each feature and iteration was produced via one-shot prompts without manual refinement or multi-pass editing.
- Use at your own risk.
(yes, this warning from readme file is also AI generated)
To @Ottetal point, no, I did not ask to embed any sheniningans into the code – but also I don’t know if there are any. I could ask another AI to verify – but what value woudl be in that? I run it under a very restrictd user on a very restricted system, because I don’t trust it either.
@MarviBiene, @mike, @hwm.land, @gingerbread233: I appreciate the sentiment, but this is truly public domain. I just thew a few prompts at the model(s), and complained when it did not work. I also copy-pasted forum comments to get fixes on the reported issues. It is an experiment on how far can one-shotting requests go. It’s probably not how good code shall be structured, designed, or written – but I have zero experience in JavaScript, web, and very little skills in Python; so I would not be able to assess, let alone meaningfully contribute to, any of that, even if I wanted to.
For example, it decided to use storagenode API and figured out how to do so on its own, without me telling it. It wasn’t even my idea. It volunteerd it (“user provided IP on the command line – i must check if the storagenode API is accessible on that IP”) . Where did the information come from? Probably Storj documentation and this forum. So who is the real author here? See my point? At best, I commissioned this app, and therefore it’s Anthropic, OpenAI, and Google who should be credited, once our copyright laws catch up with how to treat AI vomit modern times. For now - it’s public domain.
To @MarviBiene point, if I go spend money to see a movie - I don’t own the movie or experience. This is the same thing. I spent money to amuse myself. It was just an entertainment. I don’t own this. It’s like rain. It just happend.
I’ll keep my nodes running, they have been hands off for years. Hope to see you all in a bit. Best of luck.
10 Likes