kaspanet / rusty-kaspa

Kaspa full-node and related libraries in the Rust programming language. This is a stable version at the initial rollout phases.
ISC License
371 stars 118 forks source link

Weird issue starting integrated node #478

Closed Jinxsyns closed 1 month ago

Jinxsyns commented 1 month ago

Describe the bug Unsure what may have caused it but I all of a sudden can't get the p2p node to start when using the integrated daemon. If i go to a remote node it works though. it just seems to get stuck in starting (it just says not connected and starting)

To Reproduce Steps to reproduce the behavior: All I did was open the program. it worked fine the last time I ran it? And like i said switching to a remote node to sync off of works but even switching back to integrated daemon doesn't work to start up the integrated daemon Expected behavior It to function? I apologize if that sounds rude not my intention it just isn't functioning at all (the node specifically the rest of the program works)

Screenshots image

Desktop (please complete the following information):

Additional context Add any other context about the problem here.

aspect commented 1 month ago

It is more of a Kaspa NG issue, which should be filed at https://github.com/aspectron/kaspa-ng/issues (albeit it does look like the daemon is having trouble starting up, which can be related to Rusty Kaspa daemon itself).

This looks environmental or state/PC related as people are running KNG on Windows without encountering such issues.

Regardless, I have the following questions and suggestions: 1) Does the machine have enough free disk space? 2) Does switching network type have any effect? (mainnet vs testnet-11) 3) What does the Log tab show when this occurs? (Under certain circumstances, the p2p node can start re-indexing the database on startup. This can take a few minutes on regular machines. However, when this occurs, logs should state that the node is re-indexing data. If this is the case, one should wait for the re-indexing to complete.) 4) Can you please exit Kaspa NG and run the node manually from the command line. You can do this by downloading the latest release from https://github.com/kaspanet/rusty-kaspa/releases and running it as kaspad --utxoindex.

Only after you check 3) and try 4) (as running the node manually can present additional clues), can you please go to C:\Users\<username>\AppData\Local\ and rename rusty-kaspa folder to rusty-kaspa-01 (or you can also delete it). (This will help in case there has been any type of data corruption. This does not occur under normal circumstances but you never know - desktop PCs can exhibit various strange behaviors.)

You should join Kaspa Discord#help-kaspa-ng channel and ask the questions there (there are more people and it is faster to get feedback/ideas in relation to any problems).

LMK if anything I just described helps or has any effect.

Jinxsyns commented 1 month ago

It says it is resyncing the utxo index in the logs (put a screenshot for 3) I will let that cook for awhile and see what happens because 1. definitely enough free space 2. testnet 11 works. 3. image

  1. I will try if the utxo sync doesn't work after like 24 to 48 hours ( probably around 26ish because I will be needing to use my computer around that time and for what I need to use it for having the wallet run might impact performance)
Jinxsyns commented 1 month ago

will update when or if something happens (after I try running it manually from command line first if the utxo sync doesn't happen)

Jinxsyns commented 1 month ago

Well it's working now (i waited almost an hour for it last night and nothing happened) so I guess it just needed to resync thank you for your time and the assistance have a wonderful day!

aspect commented 1 month ago

@Jinxsyns Can you please give me your machine spec (CPU kind, RAM amount, Storage kind)

Jinxsyns commented 1 month ago

it fixed itself I just had to wait for the utxo sync to finish (it just took an unusually long time yesterday) The problem is solved thank you!

aspect commented 1 month ago

I am just curious about your setup - the time it takes is proportional to the computer performance. Like I have a powerful dev machine and it usually takes max 5 min.

We need to add a status in the status bar so that instead of "Starting" it says "Re-indexing". to make this less confusing. Something I overlook as I am "used to this". There is a bit of chicken and the egg problem as the node doesn't connect until it's done re-indexing so this is one condition where the status can't be easily updated... need to think how to handle this internally..

Jinxsyns commented 1 month ago

Oh a 7950x 64gigs and a gen 3 nvme ssd i believe if not gen 4

On Mon, May 27, 2024, 6:01 AM aspect @.***> wrote:

I am just curious about your setup - the time it takes is proportional to the computer performance. Like I have a powerful dev machine and it usually takes max 5 min.

We need to add a status in the status bar so that instead of "Starting" it says "Re-indexing". to make this less confusing. Something I overlook as I am "used to this". There is a bit of chicken and the egg problem as the node doesn't connect until it's done re-indexing so this is one condition where the status can't be easily updated... need to think how to handle this internally..

— Reply to this email directly, view it on GitHub https://github.com/kaspanet/rusty-kaspa/issues/478#issuecomment-2133230679, or unsubscribe https://github.com/notifications/unsubscribe-auth/A7FH4LOMDBXAK3MBPPKMRXDZEMG7ZAVCNFSM6AAAAABIKGLCS6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZTGIZTANRXHE . You are receiving this because you were mentioned.Message ID: @.***>

aspect commented 1 month ago

Ok, that's a powerful machine. Thanks for the info. I'll see how the UX can be improved for this.