Harvester not working

Support
41

I just migrated over from my WinTel box to the Ubuntu64/RPi box, so the first version on the RPi was 1.2.11 and it was working for weeks… Strange for it to go weird like that…

Update today is that the wallet is still syncing, and actually shows a wellet height. The size is aroun 6.5GB, and wallet height is reporting as 1153770 - no idea how high it’s supposed to be (new feature request maybe?).

As it’s going through the wallet chain, Chiadog is reporting XCH payments like they are new, but they’re really just old payments repeating themselves… my SpacePool doesn’t show any payments recently.

The Spacepool page also shows that I have no plots working on the pool, so after the wallet is finished syncing, I’m going to leave the SpacePool via ‘chia plotnft -leave’ then rejoin up.

Anyone have any idea how large the wallet db or wallet height is supposed to be?

43

My wallet size is right now at ~9GB, and the height is the same as the main blockchain.

Before you leave and re-join Space pool, check your harvester / farmer logs:

grep harvester debug.log

2022-01-07T11:47:56.980 harvester chia.harvester.harvester: INFO     refresh_batch: event started, loaded 0, removed 0, processed 0, remaining 1182, duration: 0.00 seconds
2022-01-07T11:47:57.056 harvester chia.harvester.harvester: INFO     refresh_batch: event batch_processed, loaded 0, removed 0, processed 300, remaining 882, duration: 0.02 seconds
...
2022-01-07T11:47:57.136 harvester chia.harvester.harvester: INFO     refresh_batch: event done, loaded 0, removed 0, processed 1182, remaining 0, duration: 0.02 seconds
2022-01-07T11:48:00.496 harvester chia.harvester.harvester: INFO     0 plots were eligible for farming d91ba98e82... Found 4 proofs. Time: 1.89092 s. Total 1182 plots
2022-01-07T11:48:08.761 harvester chia.harvester.harvester: INFO     4 plots were eligible for farming d91ba98e82... Found 3 proofs. Time: 1.43725 s. Total 1182 plots

grep farmer debug.log

2022-01-07T11:51:51.449 farmer chia.farmer.farmer         : INFO     Submitting partial for 123456789012345678901234567890123456789012345678901234567890  to https://xch-us-west.flexpool.io
2022-01-07T11:51:51.925 farmer chia.farmer.farmer         : INFO     Pool response: {'new_difficulty': 1}
...
2022-01-07T11:53:59.403 farmer chia.farmer.farmer         : INFO     Pool response: {'error_code': 2, 'error_message': 'The partial is too late. Make sure your proof of space lookups are fast, and network connectivity is good. Response must happen in less than 25 seconds, but the partial was received in 28 seconds. NAS or network farming can be an issue'}
2022-01-07T11:53:59.403 farmer chia.farmer.farmer         : ERROR    Error in pooling: (2, 'The partial is too late. Make sure your proof of space lookups are fast, and network connectivity is good. Response must happen in less than 25 seconds, but the partial was received in 28 seconds. NAS or network farming can be an issue')

2022-01-03T01:38:39.374 farmer chia.farmer.farmer         : INFO     GET /pool_info response: {'name': 'Space Pool', 'logo_url': 'https://pool.space/logo-250.png', 'minimum_difficulty': 1, 'relative_lock_height': 64, 'protocol_version': 1, 'fee': 0.01, 'description': 'Welcome to Space Pool. By creating this Plot NFT, you agree to the following terms https://pool.space/terms', 'target_puzzle_hash': '123456789012345678901234567890', 'authentication_token_timeout': 5}
2022-01-03T01:42:48.217 farmer chia.farmer.farmer         : ERROR    Failed to update farmer information on the pool https://asia1.pool.space

Those harvester lines indicate that your harvester is searching (every 2 mins) for new/removed plots, and also that is receiving challenges from the farmer, and either finds something or not based on those challenges.

And those farmer responses show that your farmer is at least trying to submit something to your pool (flex in my case), and that from time to time, your pool is providing some kind of status (Pool response:).

I also have one NFT connected to Space pool, but with zero plots, so am getting that error responses from it.

Yeah, I think that I am getting those stales after I upgraded to v1.2.11.

1 Like
44

Same height as blockchain, so latest block no.

45

I’m getting a lot of these errors:
2022-01-07T16:14:51.663 full_node chia.full_node.full_node: INFO Added unfinished_block 79900a465baa6ed24c58a0b36677e03dcffcd7f61acd3a8602eae5ffe1483f7f, not farmed by us, SP: 45 farmer response time: 5.2681, Pool pk xch18zkfw3wt6w9ns002zxdm89z30rc6m0lg79g9dj36jyttkpc3pmcs0s26ca, validation time: 0.0842 seconds, pre_validation time 0.7779, cost: 440851386, percent full: 4.008%
2022-01-07T16:14:51.950 full_node chia.full_node.full_node: INFO Added unfinished_block c82637c00101fedd16a81e05fcf01e6c9ddecef28483d2e4a5ef633db904f2a7, not farmed by us, SP: 45 farmer response time: 5.9408, Pool pk xch1f0ryxk6qn096hefcwrdwpuph2hm24w69jnzezhkfswk0z2jar7aq5zzpfj, validation time: 0.0900 seconds, pre_validation time 0.3860, cost: 440851386, percent full: 4.008%

46

They’re not errors, they don’t use the word error.
They are just bits of info your not used to seeing as you’ve not had your logs set to info level.

1 Like
47

Ok, success! I’m showing partials on FlexPool now and rewards (real ones) have resumed as before.

I made four changes and it’s fixed everything.
Change log level to INFO so ChiaDog stops reporting nothing is happening (I probably did this to myself during my reinstall(s) of Chia.)
Increased the pagefile from 1GB to 4GB - this seems to have reduced thread and system interrupt queuing.
Decreased the number of peers from 80 to 10.
Deleted the wallet db folder and resync’d everything back up (I think this is the primary solution right here…)

Thank you everyone! No stale partials. No slow partials. Everything 100% on the pool, and very low CPU usage after the wallet finished resyncing.

3 Likes
48

Great to see you finally running!

Yeah, those changes (INFO / peers: 10) should be done by chia installer. There is no excuse to not check for the system that chia installer is running on, and doing appropriate changes.

The swap file change should be on chia’s install page for RPis (maybe it is).

It would take just one chia engineer 30 mins to do it, vs. plenty of farmers being screwed (especially during those dust storms).