I personally feel that the best long term is incorporating it into launcher. Running multiple apps to monitor your nodes is not best. Especially for non-technical people who still want a speedo & fuel gauge & temp gauge for their machine.
10 Likes
We’re hoping that some updates to the Launchpad will be useful in diagnosis of node issues.
But it likely won’t be quite as granular as vDash… There’s place for both!
Screengrab in the OP shows the direction. Happy to discuss!
5 Likes
I agree unless launchpad can cover all users, but I don’t expect that. If nobody needs vdash becausethey are all happy with launchpad, it will be time to let vdash go. I’ve not tried NTracking but that plus launchpad is also a possible replacement.
@dirvine I would be sad if vdash became redundant at but I’d rather that then keep it alive as a kind of thank you to me. So I agree with the tweaks because I think that will be useful for beta now. Switching to metrics can be considered later when we see better what’s useful post beta, but I have written stuff many times and seen it outlive it’s usefulness, so that would be fine too.
13 Likes
Sorry @happybeing I have already replaced vdash with a bash script using /metrics, even with the # of records missing the active count
The metrics I show is
store cost
wallet balance
forwarded balance (yea still got the bug affecting this)
uptime
number of records
memory used
cpu used
(I don’t use but could have puts/gets)
peer count (and connection count if I wanted jack in the box figures)
And for the local network test I could update for 400 nodes in less than a second and its only a bash script. And the cpu load is unmeasurable on my 24 core machine
But even on a 2 core machine the /metrics reading is so fast even with bash scripting
10 Likes
Launchpad is targeted at getting non-technical users up and running in a low friction way, but with a rapid dev turnaround, and within the constriants of a ramp to launch that we are currently on.
We are starting to add in some feedback that does err on the more technical in order to help with diagnosis, and allow people to get support from others though the community (sharing stats, screengrabs etc, useful documented metrics) and a way to stop start nodes when things go bad.
So we are walking a fine line in not bloating things up too far… but just enoigh density to be the most useful,the the most people within the constriants.
But… if we aimed for the Launchpad to try and serve everyone that would be a surefire way to either bloat it, or create lengthy timescales, and pull time away from fixing the issues that aim to make the nodes more inherently hands-off.
Long winded way to say: long live vDash!
12 Likes
In future, I love to see the ability to “click” on a node and it bring up deeper metrics for that node - eg show the meta data and the major metrics on earnings/store-cost/etc, RT peers, and put/get type of figures. Not so much the nitty gritty metrics available
4 Likes
Someone could maintain a fork with some extras 
7 Likes
I feel the switch to metrics would be useful/required? Not sure how that feels?. I am too far for it all so I am just using gut feeling.
I could not agree more, at least as a stand alone thing, but I suspect the single launchpad may be the default? Not sure
8 Likes
Like Jim said, for the not so technical user launchpad with an essential set of metrics like the ones proposed will be what those users want. Better than having to use yet another program that is more technical, more “professional” monitoring program just to see whats happening
8 Likes
I’ve been out of the loop for a few weeks.
Does anyone have a valid peer I can use to bootstrap my nodes, please?
2 Likes
Let the node download the latest.
There should be no need to specify a peer. For the beta, there has to be something wrong if you need to use --peer option.
Does it not work if you don’t specify a --peer
6 Likes
Thanks - I removed --peer and it worked!
6 Likes
This still does not work for me, maybe I’m doing something wrong.
I killed all the nodes killall safenode. I removed ~/.local/share/safe. I checked out commit 08b0a492 . i run the network with cargo run --release --bin safenode-manager -- local run --build --clean --interval 2000 --count 6 And still:
[safe_network]$ cargo run --release --bin safenode-manager -- status --details
Finished `release` profile [optimized] target(s) in 0.55s
Running `target/release/safenode-manager status --details`
[safe_network]$ cargo run --release --bin safenode-manager -- local status
Finished `release` profile [optimized] target(s) in 0.54s
Running `target/release/safenode-manager local status`
╔═══════════════════╗
║ Local Network ║
╚═══════════════════╝
Refreshing the node registry...
Service Name Peer ID Status Connected Peers
safenode-local1 12D3KooWKmN6my5KgJd8ikAZTTc5Xng38JUFysFdpkTzXee84Qhp STOPPED -
safenode-local2 12D3KooWStey91n8Sk9FU8RKyjqFNg61CHY4LQ2AeAaRYVxZWHJj STOPPED -
safenode-local3 12D3KooWJuPSsTXHr1kMfVLkwqrvtwbXXJVHpFqaRVWXPKbdxokH STOPPED -
safenode-local4 12D3KooWBuw1LYaSNCGd3n3S5j5LddEs1vTny4sdS4wLMqPmBGXn STOPPED -
safenode-local5 12D3KooWQcf6Xuu5mvizsf9LjPGTeoqPT1679ic4T2UHYQRngtdo STOPPED -
safenode-local6 12D3KooWDPF7SaxPbQTRjPZhALV3FXHepyqxT7rKjaAYGbGuDH9Y STOPPED -
faucet - RUNNING -
3 Likes
Sorry, I just saw this.
For me it works fine on the main branch:
❯ cargo run --release --bin safenode-manager -- local status
Finished `release` profile [optimized] target(s) in 0.26s
Running `target/release/safenode-manager local status`
╔═══════════════════╗
║ Local Network ║
╚═══════════════════╝
Refreshing the node registry...
Service Name Peer ID Status Connected Peers
safenode-local1 12D3KooWBQQSi6C7hc3B8sRdwLhUuYKK5BiJGt3cgHzJgJbGQxvL RUNNING 6
safenode-local2 12D3KooWNMpbU7xWAnCfTU7Js7cGyQfjPKt1BKycXEi2SSw9poYB RUNNING 6
safenode-local3 12D3KooWCfrCLpCTJA42k84VfpZrBsaxBvDAQrPKbssBotDBjs2u RUNNING 6
safenode-local4 12D3KooWEucaGYqRNBY3AMBzdioq1CetjQpAxMcAxsJTEwbVziUe RUNNING 6
safenode-local5 12D3KooWEYwtYyQZPZCHnXQ6uM9AGKRYSdp4QZDrAH3krYJWKPUt RUNNING 6
safenode-local6 12D3KooWLbbS6Hx7UAdBZd3VXCXwVAWgF22CxitgvaUW9TKPnUVL RUNNING 6
faucet - RUNNING -
I used the same command you did to launch the network.
2 Likes
Thanks for looking into it. What could be the problem here? Are there any standard things I can do to reset system state (Linux), so the node-manager would run with fresh environment? Any debugging info I could supply?
Honestly, I am really not sure. The only thing I could think was that you had an old node registry file, but, if you deleted ~/.local/share/safe, that shouldn’t be the case.
So, I honestly can’t account for the difference here.
Edit: btw, why don’t you just try running from main, rather than a specific commit? That commit hash you posted is not the commit that has this change.
I just use commits that are tagged with new versions of crates. I’ll try from main head.
It worked, thanks.