Neuron Not Connecting to Mainnet After Data Migration

[NonSubscribed]

Hello,

I recently launched Neuron and got a message about "storage place changing" or something like that was told that a Neuron needed to perform a data migration that would last between 20-60 minutes. I started the process, however, it never ended and I believe something went wrong as Neuron now wont connect to mainnet. I'm attaching the debug log in the hope that someone will be able to figure out what went wrong and how to fix it.

Thanks

neuron_debug_1673464604229.zip

[Keith-CY]

Hi @NonSubscribed , we just got the same issue yesterday and are working on it. I'll reply here once we find the solution.

[NonSubscribed]

Thanks

[NonSubscribed]

Would uninstalling/re-installing Neuron solve this issue?

[Keith-CY]

Would uninstalling/re-installing Neuron solve this issue?

The data migration could be avoided by clearing all synchronized data and starting it from scratch.

The clear all synchronized data button is in Menu => Tools => Clear all synchronized data

---

Uninstalling Neuron doesn’t remove data from the built-in CKB node, so migration is still inevitable.

[S0rci3r]

Hello, following the new update (0.106.0) the migration failed. So I deleted all the .config/Neuron directory to start again on a healthy basis. Despite this, the synchronization takes a long time, after 4 days, it is about 30%. I have a very good throughput, and I had no problem on previous versions.
I am running Manjaro Linux (Arch Base) with a 6.1.1 kernel.
Thx

neuron_debug_1673792332898.zip

[Keith-CY]

Hello, following the new update (0.106.0) the migration failed. So I deleted all the .config/Neuron directory to start again on a healthy basis. Despite this, the synchronization takes a long time, after 4 days, it is about 30%. I have a very good throughput, and I had no problem on previous versions. I am running Manjaro Linux (Arch Base) with a 6.1.1 kernel. Thx

neuron_debug_1673792332898.zip

Hi @S0rci3r

From the bundled-ckb.log we can see

2023-01-15 14:18:22.587 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7513008, hash: Byte32(0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait
2023-01-15 14:18:26.489 +00:00 GlobalRt-3 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(1) because node is in initial block download
2023-01-15 14:18:27.182 +00:00 GlobalRt-4 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(2) because node is in initial block download
2023-01-15 14:18:27.242 +00:00 GlobalRt-2 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(3) because node is in initial block download
2023-01-15 14:18:27.360 +00:00 GlobalRt-0 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(8) because node is in initial block download

It means the built-in CKB node is still at the early stage of synchronization, which is looking for an assumed valid target block.

The log said it reached block 7513008, block hash is 0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333, and can be found on explorer https://explorer.nervos.org/block/0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333

The target set in Neuron@v0.106.0 is block hash 0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b (https://github.com/nervosnetwork/neuron/blob/develop/packages/neuron-wallet/.env#L120), its block number is 8826940, and can be found on explorer https://explorer.nervos.org/block/0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b

So it requires 1313932 more blocks to reach the assumed valid target block. Once the assumed valid target block is recognized, CKB node and Neuron will start actual synchronization.

The early stage(looking for assumed valid target block) is introduced to boost synchronization. Once the assumed valid target is found, CKB node will skip verification of blocks before the assumed valid target block and download data directly. Verification does take time.

Also from the log we can see

2023-01-15 14:04:26.608 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7493008, hash: Byte32(0xc95b021091462f1d40ed9f0b49507d68c89fadb05ecabd641b189264d6480382), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait


2023-01-15 14:11:57.060 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7503008, hash: Byte32(0x50a886a36ba434d7bf624633de435213b1d4a1c9df8b2d107a9393e584392e00), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait


2023-01-15 14:18:22.587 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7513008, hash: Byte32(0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait

It keeps checking the assumed valid target every 10000 blocks(about 7 mins), so please be patient and wait for it reaches block number 8826940. The remaining 1313932 blocks take about 1313932 / 10000 * 7 mins, namely 15 hours.

[Keith-CY]

Hello, following the new update (0.106.0) the migration failed. So I deleted all the .config/Neuron directory to start again on a healthy basis. Despite this, the synchronization takes a long time, after 4 days, it is about 30%. I have a very good throughput, and I had no problem on previous versions. I am running Manjaro Linux (Arch Base) with a 6.1.1 kernel. Thx
neuron_debug_1673792332898.zip

Hi @S0rci3r

From the bundled-ckb.log we can see

2023-01-15 14:18:22.587 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7513008, hash: Byte32(0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait
2023-01-15 14:18:26.489 +00:00 GlobalRt-3 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(1) because node is in initial block download
2023-01-15 14:18:27.182 +00:00 GlobalRt-4 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(2) because node is in initial block download
2023-01-15 14:18:27.242 +00:00 GlobalRt-2 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(3) because node is in initial block download
2023-01-15 14:18:27.360 +00:00 GlobalRt-0 INFO ckb_sync::synchronizer::get_headers_process  Ignoring getheaders from peer=SessionId(8) because node is in initial block download

It means the built-in CKB node is still at the early stage of synchronization, which is looking for an assumed valid target block.

The log said it reached block 7513008, block hash is 0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333, and can be found on explorer explorer.nervos.org/block/0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333

The target set in Neuron@v0.106.0 is block hash 0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b (develop/packages/neuron-wallet/.env#L120), its block number is 8826940, and can be found on explorer explorer.nervos.org/block/0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b

So it requires 1313932 more blocks to reach the assumed valid target block. Once the assumed valid target block is recognized, CKB node and Neuron will start actual synchronization.

The early stage(looking for assumed valid target block) is introduced to boost synchronization. Once the assumed valid target is found, CKB node will skip verification of blocks before the assumed valid target block and download data directly. Verification does take time.

Also from the log we can see

2023-01-15 14:04:26.608 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7493008, hash: Byte32(0xc95b021091462f1d40ed9f0b49507d68c89fadb05ecabd641b189264d6480382), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait


2023-01-15 14:11:57.060 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7503008, hash: Byte32(0x50a886a36ba434d7bf624633de435213b1d4a1c9df8b2d107a9393e584392e00), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait


2023-01-15 14:18:22.587 +00:00 BlockDownload INFO ckb_sync::synchronizer  best known header number: 7513008, hash: Byte32(0x5f5bec6e0936b8fb10b71eb89b023cd41509299fc29c052b72eb28b035ea3333), can't find assume valid target temporarily, hash: Byte32(0x076ffa81f8c6f2e1f22a778a1aa5d48531cc33bfd9b302e0d73fb2ce6fa4de3b) please wait

It keeps checking the assumed valid target every 10000 blocks(about 7 mins), so please be patient and wait for it reaches block number 8826940. The remaining 1313932 blocks take about 1313932 / 10000 * 7 mins, namely 15 hours.

The synced block number won't grow until the early stage is finished because at the early stage, no blocks are downloaded. It's just checking if the assumed valid target block is accessible.

[virtueorvice]

I am having the same issue.

It was better if you had warned us about backing up data before going through the migration.

This is getting ridiculous. We have to go through a lot of painful data synchronization again just because you guys cannot make a stable/working version of Neuron wallet.

[Keith-CY]

I am having the same issue.

It was better if you had warned us about backing up data before going through the migration.

This is getting ridiculous. We have to go through a lot of painful data synchronization again just because you guys cannot make a stable/working version of Neuron wallet.

Hi @virtueorvice

The point is absolutely correct that we should make the warning conspicuous and ahead of the action instead of simply asking users whether to migrate or not.

During the development of Neuron@v0.106.0, our PM was temporarily out, so these details were made from we developers' perspective, which makes the user experience terrible for normal users. I'm so sorry for that. Now the PM is back, and we'll try to correct these unprofessional behaviors.

[virtueorvice]

Since this morning I have updated to version 0.106.0 but I am stuck on "migrating..." and nothing seems to work to start the sync.

[NonSubscribed]

@virtueorvice ,

My guess is that the same thing that happened to me it's happening to you. Migration probably stopped/failed at some point and now Neuron is in a loop where its waiting for migration to end before connecting to mainnet to sync. I followed @Keith-CY 's suggestion and went to Tools=>Clear all synchronized data, restarted Neuron and now it's syncing...currently at 70%. Good luck,

[virtueorvice]

I don't have enough data and memory to resync all over again. I will try to use CKBULL to restore my wallets.

[yanguoyu]

I don't have enough data and memory to resync all over again. I will try to use CKBULL to restore my wallets.

From your uploaded image the migrating process runs failed. You can check if the LOCK file exists, if exist, delete it.
Close your Neuron and check there is no ckb process and restart Neuron to migrate again.
If it's still run failed, maybe you need to click OK to delete old data and resync from the beginning.