Hi,
I see in the .gLview , the leader block =1 and adopt = 0, (see below pic)
Is that mean I am supposed to mint a block but I missed?
How do I find out what is the reason of missing it
thanks,
Connie
Also , I see the slot number on the BP node is way behind from the 2 Relays node
the relay node slot is 114723, the BP slot is 102738. Why is this delay happened?
Do u know the time when the block should be minted? Because 1 it means u have assigned one block for epoch 301
I’d suspect it’s a problem with syncing time.
- Please compare time for all 3 machines by running the
datecommand - Do you have Chrony installed? https://www.coincashew.com/coins/overview-ada/guide-how-to-build-a-haskell-stakepool-node/how-to-setup-chrony
@Alexd1985 ,
I don’t know the time, is that mean I missed? So I can restart the node to sync?
@mcrio ,
The 3 nodes are restarted at the same time when I upgrade to version 1.30, I did not install chrony, is that a must?
What do you suggest I should do? restart the BP node or all 3 nodes?
When u checked with cncli didn’t provided to u all informations about the slot assigned?
You can check again with cncli
@Alexd1985 ,
May I asked how do I check the slot assigned? I only look at the .gLview and notice the leader =1 . In above pic
I mean do I need to first install the script that mciro mentioned?
U are using coincashew guide right?
Yes, that’s right
Step 18.12 from here
Ok, let me go through that. So if I find more details, what is the goal here?
Should I restarts all the nodes or only BP node? Just want to make sure I don’t restart in the middle of being assign
I mean For the BP slot number out of sync issue…
check first with cncli also, do u have chrony installed on your nodes? You already answered … chrony is for a better stnchronization of the nodes with the blockchain; u should install it on all 3 nodes
Ok, will do,
Thanks a lot for both of your help!
Connie
1 Like
The node restart is not related to the date/time set. Enter the command date on all 3 nodes and see if there are noticeable differences. As Alex mentioned, it’s advisable to install Chrony as it’ll make sure the node times are in sync.
That is indeed strange that the BP was fine with the relays being so far behind.
Now, I installed chrony on all nodes, what time zone should I use?
I see they are by default is utc for all 3 nodes, and I changed one of them to PST, not sure if I should change it back to UTC, I don’t know which location in US is using UTC timezone? How do I change the node back to UTC time zone?
It doesn’t matter, it’s fine as long the chrony is running
Out of curiosity, did you compare the dates on each machine before you installed Chrony?
No location in the US uses UTC. UTC equals the GMT timezone that goes straight through London.
As Alex said, it doesn’t matter which timezone you use on your machine.
I did compare, they are about the same time, they are all UTC before I install the chrony.
but before I install, they are in sync already. I don’t know why I missed the mint still
one question is : after install , it is autmatically run as long as I follow the link above?
So how do I find out why I missed the mint?
Now BP and Relay 1 is in UTC and Relay2 in PTS, that doesn’t matter?
If you had to generate new kes keys make sure that you generated the new node.cert using the new kes keys and that you used the previous node.counter file. I have seen people incorrectly generate new node.cert, but using old kes keys, or generating the node.cert using new kes keys and a new node.counter file …
The steps are here at this Link
Also there is about a 1% chance that you will be assigned a slot at the same time as someone else does (purely from own experience, no hard math behind this) and at that point it’s a flip of a coin. Worth checking if the block was actually minted, just by someone else. You can check that on adapools.org by going through the list of all minted blocks
Then lastly in your printscreen your have stats Missed slot leader checks which is low at 0.2% which i would say is about normal, but that says that % of slots you miss as a result of bad connectivity , so your slot could have been in those 0.2%
If you know the exact time when your slot was you can check the logs of the cardano node if it is managed by systemd, by running something like this and changing the dates
journalctl --unit=cardano-node --since='2021-11-15 08:14:00' --until='2021-11-15 08:18:00'