I have been using the leaderLogs.py script PoolTool.io released some time back to track when I might expect to be elected as Slot leader for maintenance purposes.
Recently I was elected for a Slot and this block was not created by my pool, a different slot produced the block altogether. The slot number I was supposedly elected for was between slots 168500 and 168416 which were both created by ShelleyGenesis pools.
I assumed that my pool failed to create the block but I have scoured my logs and I cannot find hide nor hair that my producer node was elected as slot leader during this time. I can confirm that my producer has been started with the following startup script:
The only difference between this and the example for starting a producer in the official documentation is that I have not provided a host-addr. This I am aware of but as I understand it my node has bound to 0.0.0.0 because this setting is for specifying an address to bind to.
So my questions are:
Is the leaderLogs.py script flawless in determining leadership?
What should I be looking for in the logs of my producer to determine whether or not it was elected to create a block?
Is there any other diagnosis I can perform to determine that my producer is functioning correctly? I just yesterday upgraded to version 1.25.1 I guess I could have broken something.
Any other help or suggestions would be greatly appreciated.
I’d like to post how displeased I am with this situation and I don’t think I am being totally unreasonable here.
Firstly LiveView was removed which allowed some method of easily and quickly determining missed blocks.
Secondly the replacement RT-View is not listed as compatible with 1.25.1 and despite following the instructions I have not been able to get this working. I confess this could be my own failing.
Third the Prometheus dashboard has been modified and also no longer informs Stake Pool owners of missed blocks.
Fourth the logs which I have written out using a scribe in the mainnet-config.json also do not mention missed blocks.
Now whether all of the above is because of failings on my part I have yet to come across a SPO who can definitively answer my questions but the lack of available information is something we must rectify! How can small SPOs prove that their pool is working or even if it is not working with the nodes in their current state? It’s not quite so bad for a large pool that makes many blocks and will know shortly after an upgrade that everything is in good health.
Please can I get some advice from the community about whatever it is that I must be missing? These are far from unfair questions and have me very concerned about the entire project at this time.
gliveview is accesible on a fancy format, but if u are not using glive u have an alternative to liveview
I didn’t heard about any issues with RTview after 1.25.1 update
all metrics now reside in cardano.node.metrics and u will need to update the metrics in ur grafana dashboard
U can use leaderlog script to check if ur Producer has blocks assigned for the next epoch (u can check with 1.5 day before the next epoch)
How u can check if ur BP is ready to produce blocks? First u should check if it is started as a producer not as a relay, then check if is processing transactions, and check ur KES, certifications (should be vallid) etc