After recovering the node, some channels are still being closed

The existing node has low CPU performance and operating in a VM environment seems to cause performance degradation, so I want to move to a native server.

However, due to lack of understanding of the recovery process, all channels were closing after recovery on the new node.

And I think I saw that you have to wait at least 14 days. Even after all this time, some channels are still being closed.

previous node spec
Esxi 6.7, cpu J4105 , ram 6GB , Base OS ubuntu 22.04.3 LTS
umbrel ( 0.5.4 ) install : ‘curl -L | bash’ command
install app : Bitcoin Node, BTC RPC Explorer, Electrs, Lightning Node, mempool, Ride The Lightning, Torq

New node : i3-6500, ECC 16GB RAM , SSD 1TB , ubuntu 22.04.3 LTS
umbrel install : ‘curl -L | bash’ command
install app : Bitcoin Node, BTC RPC Explorer, Electrs, Lightning Node, mempool, Ride The Lightning, Torq

Before inserting the wallet restoration seed into the new node, back up the channel on the existing node and turn off the existing node.
Enter the recovery phrase in the new node, upload the last backed up channel file, and restart.

Contrary to what I thought (I was wrong), all channels were forced to close and remained in a pending state.

The funds started coming back after 2 hours. In the case of many channels, the channel termination does not complete,
so you have to wait for 14 days.
After 14 days, 21 of the 25 channels I opened were terminated, but the remaining 4 are still pending.

enter the command below

sudo ./scripts/app compose lightning exec lnd lncli pendingchannels

Output was below

“channel”: {
“remote_node_pub”: “03864ef025fde8fb587d989186ce6a4a186895ee44a9 26bfc370e2c366597a3f8f”,
“channel_point”: “41f1034ff5df04e4ef247eae3dd90e2f44a27cdcbb05c3 4ec9d057156687156e:0”,
“capacity”: “1000000”,
“local_balance”: “0”,
“remote_balance”: “0”,
“local_chan_reserve_sat”: “0”,
“remote_chan_reserve_sat”: “0”,
“initiator”: “INITIATOR_LOCAL”,
“commitment_type”: “ANCHORS”,
“num_forwarding_packages”: “0”,
“chan_status_flags”: "ChanStatusLocalDataLoss|ChanStatusRestored ",
“private”: true
“limbo_balance”: “0”,
“commitments”: {
“local_txid”: “”,
“remote_txid”: “”,
“remote_pending_txid”: “”,
“local_commit_fee_sat”: “0”,
“remote_commit_fee_sat”: “0”,
“remote_pending_commit_fee_sat”: “0”
“closing_txid”: “”

As mentioned in some howtos, the forced shutdown command is as follows.

sudo ./scripts/app compose lightning exec lnd lncli closechannel --force 41f1034ff5df04e4ef247eae3dd90e2f44a27cdcbb05c34ec9d057156687156e 0
[lncli] rpc error: code = Unknown desc = cannot close channel with state: ChanStatusLocalDataLoss|ChanStatusRestored

There is no way to contact the 4 pending node operators in 1ML or amboss,
so I searched and left a message requesting to force the channel to be terminated, but there was no reply.

Summary of questions.

  1. I would like to start anew after recovering my funds, but 4 channels are not being terminated due to the above problem.
    The same applies if you wait 14 days. I heard that the existing node and the new node should not overlap,
    so the existing node continues to be in a terminated state. Is it possible to terminate the channel using the existing node?

  2. Although funds came into the wallet in the meantime, most of the funds returned are in an unconfirmed state and cannot be transferred.
    If you look at the onchain menu in the ride the lightning app, most of the UTXOs field is labeled as 0.sweep.
    Will I be able to withdraw money if I Sweep all?

  3. a. In the case of unconfirmed transaction number 2 above, there are some whose TXID cannot be searched.
    I understand that low-fee transactions are purged from Mempool. Can I continue to leave them as is?

txid : 7013f0116081c1edce9370b554f91e60a9f887c5c0f0950e34c019affa6a0e7b

b. Some transactions are being BRF with the addition of a lightning swept anchor as shown below. Can I wait for this as well?

Please help me safely retrieve my funds.
Thanks in advanced.

We could possibly make this clearer for users, as I believe if you start up a new LND instance with the same key while the old one is still live, LND rules will attempt to close your channels. There are other causes too, for the moment we generally recommend closing channels before migrating servers. Reference: Migrating LND - Builder's Guide

If you are unable to contact the other nodes, you will likely need to use Chantools as per the below.

Please give me a moment, and I will try to write this up into a concise guide.

Thanks for your reply.

Before recovering using the new node, I turned off the original node an hour ago.

anyway, yesterday, 21 out of 25 channels were closed in the new node passed 16 days, and the returned onchain bitcoin was transferred to a hardware wallet and the node was shut down.

I tried restarting the previous original node.
It took some time to synchronize blocks that had been off for 15 days, but it was successfully restarted. The four channels that were in a pending state on the new node remain open and changed to the active state after synchronization was completed.

After individually backing up the four channels with balance information, I terminated the original node and restored the channel (which was previously in a pending state) to another test node. still
The message " ChanStatusLocalDataLoss|ChanStatusRestored " appears, and it has not been properly restored and is in a pending state.
In the end, the new test node shuts down again.

I rebooted the original node, waited for the channel to become active, and executed a channel closing command by applying a transmission fee of 13 sat/vbyte.

The four channels that had been in a pending state for over 15 days in the new node were open and active in the original node.

when the channel closing command was issued, they now entered a pending state.

The status of channels being terminated in the Ride the Lightning app is as follows. ( one of them )

Wait Closing Channel Information
Transaction ID
Channel Point
Peer Alias
Peer Node Pubkey
Limbo Balance
Local Balance
Remote Balance

I don’t know yet whether it will be restored normally this way.
And no transaction has yet been viewed in the mempool with the TXID shown above.

I’ll wait for a day and if it doesn’t work, I’ll try the chantool you mentioned above.

Thank you again.