Dry Run 6: Great Scott!

[bytemaster]

Based upon the log file, it appears that at some point in the life of your delegate ntp_error_seconds was different?   

I just checked in an update to the NTP code that moves it to its own dedicated thread and checks for any major errors due to CPU load and network latency.  IE: I filter all NTP requests that take over 1 second to execute.

This error msg is not right?

Code: [Select]

20140625T221020.467325       th_a            read_loop ] disconnected 0 exception: unspecified
Bad file descriptor
    {"message":"Bad file descriptor"}
    asio  asio.cpp:28 read_write_handler

    {"len":16}
    th_a  stcp_socket.cpp:73 readsome                   message_oriented_connection.cpp:144
It just loop again and again

It isn't pretty, but is harmless. 

[alt]

Untested fix: https://github.com/BitShares/bitshares_toolkit/commit/c5a4d1cf708829dd30fa3112dfbe9e3e184a7b2c

thx, I'll try after lunch

[alt]

Based upon the log file, it appears that at some point in the life of your delegate ntp_error_seconds was different?   

I just checked in an update to the NTP code that moves it to its own dedicated thread and checks for any major errors due to CPU load and network latency.  IE: I filter all NTP requests that take over 1 second to execute.

This error msg is not right?

Code: [Select]

20140625T221020.467325       th_a            read_loop ] disconnected 0 exception: unspecified
Bad file descriptor
    {"message":"Bad file descriptor"}
    asio  asio.cpp:28 read_write_handler

    {"len":16}
    th_a  stcp_socket.cpp:73 readsome                   message_oriented_connection.cpp:144
It just loop again and again

[vikram]

how many block produced by delegate-baozi?

Code: [Select]

delegate (unlocked) >>> blockchain_get_account_record delegate-baozi
Record for 'delegate-baozi' -- Registered on 2014-Jun-25 21:54:00, last update was 4 hours ago
Owner's key: XTS6UxJfhUpHVRCUC5Xk5EVtmiCjh8ZoNFaAkaWunLKqAxKEHi76z
NET VOTES           BLOCKS PRODUCED BLOCKS MISSED   PRODUCTION RATIO    LAST BLOCK #    TOTAL PAY
----------------------------------------------------------------------------------------------------------------------------------------------------
0.00604399%         0               1               0                   N/A             0.00000 XTS
Public data:
""
delegate (unlocked) >>> blockchain_get_delegate_block_stats delegate-baozi
[[
    142,{
      "missed": true,
      "latency": null
    }
  ],[
    143,{
      "missed": false,
      "latency": 1
    }
  ]
]


Good catch... this is what the seed node reports:

blockchain_get_delegate_block_stats delegate-baozi
[[
    142,{
      "missed": true,
      "latency": null
    }
  ]
]

This is part of a known bug that Vikram is looking into.

Untested fix: https://github.com/BitShares/bitshares_toolkit/commit/c5a4d1cf708829dd30fa3112dfbe9e3e184a7b2c

[bytemaster]

20140626T014551.698048 (23 minutes ago )
30018 time_in_future: time is in the future

    {"block_data.timestamp":"20140626T014200","now":"20140626T013613","delta_seconds":347}

 
   "ntp_time": "20140626T021000.433953",
  "ntp_error_seconds": -0.0010579999999999999,
  "wallet_local_time": "20140626T021000"

{"previous":"5e77eec35d560bb577ab8a6b8c933c52ec329626","block_num":162

Very curious...  block_num 162 was produce by init27 which is running on the seed node server which is reporting:

"ntp_error_seconds": -0.020045

So some how we got a ~6 minute error in when the block was produced.

Based upon the log file, it appears that at some point in the life of your delegate ntp_error_seconds was different?   

I just checked in an update to the NTP code that moves it to its own dedicated thread and checks for any major errors due to CPU load and network latency.  IE: I filter all NTP requests that take over 1 second to execute.

[Snail]

@snail

You're talking about the one you sent to your delegate?
7) Wallets with active delegates do not scan transactions

Thank you for your answer

[vikram]

Export fork graphs using command 'blockchain_export_fork_graph'. You can see usage by typing 'help blockchain_export_fork_graph'.

[toast]

@snail

You're talking about the one you sent to your delegate?
7) Wallets with active delegates do not scan transactions

[vikram]

sexy missed 163
 by the way , how to export fork graph ?

blockchain_get_delegate_block_stats sexy
[[
    163,{
      "missed": false,
      "latency": 0
    }
  ]
]

delegate-dorian  is the one who apparently skipped block 164 produced by init4

Are you around dorian to report the issue you saw?

I think the forks are all created by my node?
I have these account: delegate-alt, delegate-dorian, delegate-baozi, delegate-watchman, delegate-dragon
delegate-watchman product block 111 success.
delegate-dorian miss 165
delegate-alt miss 136
delegate-dragon miss 113
delegate-baozi miss 142

here is my version info

Code: [Select]

delegate (unlocked) >>> about
{
  "bitshares_toolkit_revision": "18edd548de8a55c432e940afa04a7050919bafc4",
  "bitshares_toolkit_revision_age": "5 hours ago",
  "fc_revision": "3914383cfcf8be4e57f60f5d5eb7e4e83de2e2b2",
  "fc_revision_age": "23 hours ago",
  "compile_date": "compiled on Jun 25 2014 at 21:45:11"
}


"missed": false means the block was NOT missed.

[Snail]

Win-0.0.6-Wallets

System:Windows 7 ultimate

Snail to mybitshares to 1000XTS wallet, block there, but GUI does not display, can not transfer

Accounts does not display

But the account display

Once again the transfer error，insufficient funds，I need 1000.1 XTS, but I only have 989999.

[bitcoinerS]

Code: [Select]

>>> wallet_list_unspent_balances
           BALANCE                                   OWNER                     VOTE
--------------------------------------------------------------------------------------------------------------------------
      62.42148 XTS    XTSEWu4aQPJpDyABZtgtcRzvvJnYpp7jhDVv10 assert_exception: Assert Exception
!"unimplemented condition type":
    {}
    th_a  cli.cpp:1172 print_unspent_balances


[bytemaster]

I install BitSharesXT-0.0.6.exe .but find ntptime error .
how to solve the ntp problem
>> get_info

{
  "blockchain_head_block_num": 199,
  "blockchain_head_block_time": "20140626T030800",
  "blockchain_head_block_time_rel": "33 seconds in the future",
  "blockchain_confirmation_requirement": 47,
  "blockchain_average_delegate_participation": 90.867579908675793,
  "network_num_connections": 6,
  "ntp_time": "NTP time not available",
  "ntp_error_seconds": "NTP time not available",
  "wallet_unlocked_seconds_remaining": 59982,
  "wallet_next_block_production_time": null,
  "wallet_seconds_until_next_block_production": null,
  "wallet_local_time": "20140626T030727",
  "blockchain_random_seed": "d0acbe6a58aa8fd213a6cb64a78965873ed1c5e7",
  "blockchain_shares": 199999897443990,
  "network_num_connections_max": 200,
  "network_protocol_version": 104,
  "wallet_open": true,
  "wallet_unlocked_until": "20140626T194709",
  "wallet_version": 100
}

The problem should resolve itself within an hour.  It just means the UDP packet to the NTP server was dropped.  I am fixing a bug with this code now for Alt.

[yangsbo]

I install BitSharesXT-0.0.6.exe .but find ntptime error .
how to solve the ntp problem
>> get_info

{
  "blockchain_head_block_num": 199,
  "blockchain_head_block_time": "20140626T030800",
  "blockchain_head_block_time_rel": "33 seconds in the future",
  "blockchain_confirmation_requirement": 47,
  "blockchain_average_delegate_participation": 90.867579908675793,
  "network_num_connections": 6,
  "ntp_time": "NTP time not available",
  "ntp_error_seconds": "NTP time not available",
  "wallet_unlocked_seconds_remaining": 59982,
  "wallet_next_block_production_time": null,
  "wallet_seconds_until_next_block_production": null,
  "wallet_local_time": "20140626T030727",
  "blockchain_random_seed": "d0acbe6a58aa8fd213a6cb64a78965873ed1c5e7",
  "blockchain_shares": 199999897443990,
  "network_num_connections_max": 200,
  "network_protocol_version": 104,
  "wallet_open": true,
  "wallet_unlocked_until": "20140626T194709",
  "wallet_version": 100
}

[bytemaster]

Alt... I think I found a major hint as to the problem:

1) Your local clock reported the "time in future error" at  01:45:51 according to your local clock,  the time on the block was 01:42:00, and the NTP adjusted time internal to bitshares was 01:36:13

So according to your system clock it wasn't in the future.... it was 3 minutes in the past...but according to NTP adjusted time it was 6 minutes in the future...

If I had to guess I would blame our NTP time sync code.

[alt]

I saw this node sync very strange.
from 02:03 - 02:09 sync nothing
then sync 3 blocks suddenly

May be  the network is off-line sometimes?

no, I can remote control without latency at the same time.