upgrading - Rejoin
Re-join the Network
The configured layer0 profile will rejoin the network. In this case the profile dag-l0 is configured as the layer0 and will attempt to join.
In this upgrade documentation, we did not choose the -w (watch) mode.
Unless needed, the watch mode is not recommended because it will slow down our upgrade.
Without the -w we create an upgrade that is less verbose. We save time by not forcing the Node Operator to wait for all peer to peer connections to be established. Instead, once the Node reaches a state where it is able to participate on the network, nodectl will stop watching the peer connections. While the remaining peers connect in the background, we can safely continue the upgrade process.
DownloadInProgress
Most likely, your Node will reach the DownloadInProgress state, and we may want to continue the upgrade prior to moving out of this state.
Note
This documentation uses the word stage and state interchangeably.
This is perfectly normal situation and this stage may take longer than most other stages of the join process.
DownloadInProgress is the state (stage) of the Node's joining process where it will download your copy of the blockchain. This is required to participate on the cluster, and this process can take a long time.
First time connections
Upon the initial join process to the cluster, since this is the first time your Node is joining the cluster, it may take a few hours to complete the DownloadInProgress stage. Please practice patience during this time.
You may not reach the next stage shown below, the Node may not properly connect to the layer1. This is because layer1 must wait for layer0 to complete its join process before joining the layer1 cluster.
You will be required to come back to your Node after a few hours to complete the join process and bring your layer1 online.
Alternatively, you can engage the auto_restart feature. When engaged, nodectl will wait for layer0 to move into Ready state and automatically join your Node to the layer1 cluster for you.
Continuing the upgrade
Below we will see that our Node reached DownloadInProgress while it was connected to 36 out of 240 known Nodes on the network.
At ~49( the Node reached a timer threshold ~40 ) and nodectl continued on with the upgrade process...
Reviewing [dag-l0] ............................ ReadyToJoin
Joining with peer [601f8f1d...f2163467] ....... 1.1.1.1
Join cluster status [dag-l0] .................. Preparing
Max Timer 300 seconds
Peers: 240 Connected: 36 State: DownloadInProgress Timer: 49
nodectl has detected DownloadInProgress state.
This is not an issue; however, Nodes may take longer than expected time to complete this process. nodectl will terminate the watching for peers process during this join in order to avoid undesirable wait times.
Join process complete ......................... done
IMPORTANT It is ok that the peer count < cluster peer count because watch mode was not chosen by upgrade.
Checking status [dag-l0] ......................
Show Status Again
dag-l0 active (running) DownloadInProgress
PUBLIC API TCP P2P API TCP CLI API TCP
9000 9001 9002
LATEST ORDINAL LAST DLed BLK EXP ORDINAL
1010472 1010286 1010472
CURRENT SESSION FOUND SESSION ON NETWORK
1702306944446 1702306944446 True
CLUSTER START NODE START SYSTEM START
2023-12-11-15:02:24Z ~1M 2S 2023-12-06 11:46:32
CLUSTER UPTIME NODE UPTIME SYSTEM UPTIME
~9D 3H 38M 9S n/a ~14D 6H 59M 31S
NODE ID IN CONSENSUS
12345...12345 True
Once nodectl reaches Ready on the primary layer0, it will then attempt to join layer1.
In our example, the Node has continued to remain in DownloadInProgress while it was in the process of completing this stage of joining the network.