BTSX 0.4.5 Bug Fix Status Update

[mbaeichapareiko]

mac OS X  10.9  operating system  not syncing,   

[puppies]

I have reported this to Dan & Eric who have this issue as their top priority.

Thanks.  I will post any further crashes directly at github

[bytemaster]

Will there be a working wallet before Bter gets "hacked" again?

_lulz

The wallet works for many users, but clearly there are some variations in OS and system configuration that are exposing hard to replicate bugs for some individuals.   

[bytemaster]

another crash on my seed node.

Code: [Select]

--- there are now 31 active connections to the p2p network
(wallet closed) >>>
Program received signal SIGSEGV, Segmentation fault.
fc::tcp_socket::close (this=0x3a2265646f6322e3) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
64              if( is_open() )
(gdb) bt
#0  fc::tcp_socket::close (this=0x3a2265646f6322e3) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
#1  0x00000000007d6f9a in bts::net::stcp_socket::close (this=<optimized out>) at /home/user/bitsharesx/libraries/net/stcp_socket.cpp:107
#2  0x00000000007b888c in close_connection (this=<optimized out>) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:222
#3  bts::net::message_oriented_connection::close_connection (this=this@entry=0xd69eb38) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:289
#4  0x00000000007ad5be in bts::net::peer_connection::close_connection (this=this@entry=0xd69ead0) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:231
#5  0x00000000007ae3c9 in bts::net::peer_connection::send_queued_messages_task (this=0xd69ead0) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:186
#6  0x00000000007ae82c in operator() (__closure=<optimized out>) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:223
#7  fc::detail::void_functor_run<bts::net::peer_connection::send_message(const bts::net::message&)::__lambda2>::run(void *, void *) (functor=<optimized out>, prom=0xd2fd6e0)
    at /home/user/bitsharesx/libraries/fc/include/fc/thread/task.hpp:56
#8  0x000000000066ea13 in fc::task_base::run_impl (this=this@entry=0xd2fd6f0) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:42
#9  0x000000000066f0c5 in fc::task_base::run (this=this@entry=0xd2fd6f0) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:31
#10 0x000000000066d24b in run_next_task (this=0x185a350) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:379
#11 fc::thread_d::process_tasks (this=this@entry=0x185a350) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:402
#12 0x000000000066d496 in fc::thread_d::start_process_tasks (my=25535312) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:359
#13 0x0000000000d53ace in make_fcontext ()
#14 0x000000000185a350 in ?? ()
#15 0x0000000000000000 in ?? ()

p2p log https://www.dropbox.com/s/ij8drowc8kv24zd/p2p.log.20140822T230000?dl=0

I have reported this to Dan & Eric who have this issue as their top priority. 

[Tuck Fheman]

Will there be a working wallet [ EDIT : for me personally  =) ] before Bter gets "hacked" again?

_lulz

[puppies]

another crash on my seed node.

Code: [Select]

--- there are now 31 active connections to the p2p network
(wallet closed) >>>
Program received signal SIGSEGV, Segmentation fault.
fc::tcp_socket::close (this=0x3a2265646f6322e3) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
64              if( is_open() )
(gdb) bt
#0  fc::tcp_socket::close (this=0x3a2265646f6322e3) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
#1  0x00000000007d6f9a in bts::net::stcp_socket::close (this=<optimized out>) at /home/user/bitsharesx/libraries/net/stcp_socket.cpp:107
#2  0x00000000007b888c in close_connection (this=<optimized out>) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:222
#3  bts::net::message_oriented_connection::close_connection (this=this@entry=0xd69eb38) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:289
#4  0x00000000007ad5be in bts::net::peer_connection::close_connection (this=this@entry=0xd69ead0) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:231
#5  0x00000000007ae3c9 in bts::net::peer_connection::send_queued_messages_task (this=0xd69ead0) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:186
#6  0x00000000007ae82c in operator() (__closure=<optimized out>) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:223
#7  fc::detail::void_functor_run<bts::net::peer_connection::send_message(const bts::net::message&)::__lambda2>::run(void *, void *) (functor=<optimized out>, prom=0xd2fd6e0)
    at /home/user/bitsharesx/libraries/fc/include/fc/thread/task.hpp:56
#8  0x000000000066ea13 in fc::task_base::run_impl (this=this@entry=0xd2fd6f0) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:42
#9  0x000000000066f0c5 in fc::task_base::run (this=this@entry=0xd2fd6f0) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:31
#10 0x000000000066d24b in run_next_task (this=0x185a350) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:379
#11 fc::thread_d::process_tasks (this=this@entry=0x185a350) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:402
#12 0x000000000066d496 in fc::thread_d::start_process_tasks (my=25535312) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:359
#13 0x0000000000d53ace in make_fcontext ()
#14 0x000000000185a350 in ?? ()
#15 0x0000000000000000 in ?? ()

p2p log https://www.dropbox.com/s/ij8drowc8kv24zd/p2p.log.20140822T230000?dl=0

[CalabiYau]

Ok the gui windows started to synchronize and then it crashes... When should we expect a fix for that?

same with me for the windows gui 0.4.5

config win7 32bit

Same here on win7/64. Crashing.
 Linux  soso~lala

[liondani]

Ok the gui windows started to synchronize and then it crashes... When should we expect a fix for that?

same with me for the windows gui 0.4.5

config win7 32bit

[mf-tzo]

Ok the gui windows started to synchronize and then it crashes... When should we expect a fix for that?

[puppies]

my seed node just crashed on the dacsunlimited master branch.

Code: [Select]

--- there are now 30 active connections to the p2p network
(wallet closed) >>>
Program received signal SIGSEGV, Segmentation fault.
fc::tcp_socket::close (this=0x68) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
64              if( is_open() )
(gdb) bt
#0  fc::tcp_socket::close (this=0x68) at /home/user/bitsharesx/libraries/fc/src/network/tcp_socket.cpp:64
#1  0x00000000007d3e2a in bts::net::stcp_socket::close (this=<optimized out>) at /home/user/bitsharesx/libraries/net/stcp_socket.cpp:107
#2  0x00000000007b571c in close_connection (this=<optimized out>) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:222
#3  bts::net::message_oriented_connection::close_connection (this=this@entry=0x10578098) at /home/user/bitsharesx/libraries/net/message_oriented_connection.cpp:289
#4  0x00000000007aa3ce in bts::net::peer_connection::close_connection (this=this@entry=0x10578030) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:230
#5  0x00000000007ab1d9 in bts::net::peer_connection::send_queued_messages_task (this=0x10578030) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:185
#6  0x00000000007ab63c in operator() (__closure=<optimized out>) at /home/user/bitsharesx/libraries/net/peer_connection.cpp:222
#7  fc::detail::void_functor_run<bts::net::peer_connection::send_message(const bts::net::message&)::__lambda2>::run(void *, void *) (functor=<optimized out>, prom=0x12e7a850)
    at /home/user/bitsharesx/libraries/fc/include/fc/thread/task.hpp:56
#8  0x000000000066c2b3 in fc::task_base::run_impl (this=this@entry=0x12e7a860) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:42
#9  0x000000000066c965 in fc::task_base::run (this=this@entry=0x12e7a860) at /home/user/bitsharesx/libraries/fc/src/thread/task.cpp:31
#10 0x000000000066aaeb in run_next_task (this=0x18533b0) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:379
#11 fc::thread_d::process_tasks (this=this@entry=0x18533b0) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:402
#12 0x000000000066ad36 in fc::thread_d::start_process_tasks (my=25506736) at /home/user/bitsharesx/libraries/fc/src/thread/thread_d.hpp:359
#13 0x0000000000d4e15e in make_fcontext ()
#14 0x00000000018533b0 in ?? ()
#15 0x0000000000000000 in ?? ()
(gdb)

p2p log https://www.dropbox.com/s/ss4l624pewfq0z8/p2p.log.20140822T170000?dl=0

[mf-tzo]

Guys,

I am having all kind of problems with 0.4.5 version on windows and I am wondering if its only me. I have installed and reinstalled 3 times, i am trying to import my wallet and I get a lot of errors such as restored wallet failed...
I see no transactions
I have installed again after deleting blockchains as well and now I have zero connections...
what is happening?

[bytemaster]

Some general observations for delegates:
3) We recommend delegates produce blocks from a node with no incoming connections and perhaps 20 outgoing connections.

Are these generally true going forward or just now until the patches are fixed?

I'd like to set up two more machines - one is another delegate node separate from my existing delegate, and another will be a simple serving node like you suggested.
bytemaster, can you vote for delegate:
delegate1.maqifrnswa
in addition to the existing maqifrnswa, please? thanks

and my
delegate.liondani
in addition to the  existing liondani ?

It is best practice overall to prevent DDOS.

[liondani]

Some general observations for delegates:
3) We recommend delegates produce blocks from a node with no incoming connections and perhaps 20 outgoing connections.

Are these generally true going forward or just now until the patches are fixed?

I'd like to set up two more machines - one is another delegate node separate from my existing delegate, and another will be a simple serving node like you suggested.
bytemaster, can you vote for delegate:
delegate1.maqifrnswa
in addition to the existing maqifrnswa, please? thanks

and my
delegate.liondani
in addition to the  existing liondani ?

[maqifrnswa]

Some general observations for delegates:
3) We recommend delegates produce blocks from a node with no incoming connections and perhaps 20 outgoing connections.

Are these generally true going forward or just now until the patches are fixed?

I'd like to set up two more machines - one is another delegate node separate from my existing delegate, and another will be a simple serving node like you suggested.
bytemaster, can you vote for delegate:
delegate1.maqifrnswa
in addition to the existing maqifrnswa, please? thanks

[alt]

It's AWESOME how quickly you can resolve those bugs!!

 

[Riverhead]

[dxtr]

http://hackertyper.com/

[liondani]

It's AWESOME how quickly you can resolve those bugs!!

they are typing almost quicker than thinking...

[xeroc]

It's AWESOME how quickly you can resolve those bugs!!

[bytemaster]

We think we have found a fix for the transaction ledger errors reported here:

https://github.com/BitShares/bitshares_toolkit/issues/657

We are testing that now. 

[bytemaster]

We have identified the cause of syncing issues... a race condition on which nodes can serve you the chain first.  If you happen to connect to an un-upgraded node and fetch their chain first then you will be unable to finish syncing.   The chain server should fix this in the current version and the next version is putting in extra code to handle this case.

https://github.com/BitShares/bitshares_toolkit/wiki/Using-Chain-Servers

[bytemaster]

We are preparing a post about using a Chain Server for rapid download from a trusted source.  Nathan is preparing a How To.  For those having trouble syncing directly from P2P clients this alternative may work for you.

Good news. Are we talking Electrum like functionality or a pre-indexed blockchain?

Single source fire-hose of all blocks... rather than the much more complex algorithm required for P2P syncing.  Performance is significantly better for "catching up".

We do have Electrum-like functionality on the roadmap and partly designed. 

[Riverhead]

We are preparing a post about using a Chain Server for rapid download from a trusted source.  Nathan is preparing a How To.  For those having trouble syncing directly from P2P clients this alternative may work for you.

Good news. Are we talking Electrum like functionality or a pre-indexed blockchain?

[bytemaster]

We have a mitigating fix for the network crash pending in the toolkit repository.  We will have to wait for DAC Sun Limited to update their master branch.

Some general observations for delegates:
1) Your client is more likely to crash if you have more connections.
2) We are preparing a new command line argument that will disable incoming connections
3) We recommend delegates produce blocks from a node with no incoming connections and perhaps 20 outgoing connections.
4) We recommend delegates ALSO host a public node with a lot of connections to support the network.
5) If the computer has enough RAM running both on the same machine is possible.

We are preparing a post about using a Chain Server for rapid download from a trusted source.  Nathan is preparing a How To.  For those having trouble syncing directly from P2P clients this alternative may work for you. 

[liondani]

 

[Riverhead]

Wait, you created a bug against the tool kit and I created it against DACSunLimited. I guess it is the tool kit and their build is on top of that. I'll delete my bug and we'll use yours.

I've never logged a bug before, just created my Git Hub account today. Such a noob .

[Riverhead]

Where do you want us to send/post the P2P logs?

Post links to the github issues.

i have forwarded your link to github 
https://github.com/BitShares/bitshares_toolkit/issues/649

I posted the wrong link


https://github.com/dacsunlimited/bitsharesx/issues/22

[liondani]

Where do you want us to send/post the P2P logs?

Post links to the github issues.

https://www.dropbox.com/s/ybrli5ruc2xdn3y/riverhead-del-server-1.p2p_logs.tar.gz

i have forwarded your link to github 
https://github.com/BitShares/bitshares_toolkit/issues/649

[Riverhead]

Where do you want us to send/post the P2P logs?

Post links to the github issues.

https://github.com/dacsunlimited/bitsharesx/issues/22

[bytemaster]

Where do you want us to send/post the P2P logs?

Post links to the github issues.

[lucky331]

nice. thanks.

[Riverhead]

Where do you want us to send/post the P2P logs?

[bytemaster]

We have identified some known issues that we are attempting to resolve:

Code: [Select]

         Peer 76.74.170.79:1700 disconnected us: You offered us a block that we reject as invalid
         Peer 106.185.26.162:1976 disconnected us: You offered us a block that we reject as invalid
         Peer 106.185.26.162:1776 disconnected us: You offered us a block that we reject as invalid
         Peer 80.240.133.79:1777 disconnected us: You offered us a block that we reject as invalid
         Peer 180.153.142.120:1777 disconnected us: You offered us a block that we reject as invalid
         Peer 104.131.134.181:36856 disconnected us: You offered us a block that we reject as invalid
         Peer 162.243.220.142:1776 disconnected us: You offered us a block that we reject as invalid

The last time I did a complete resync with the network I had this message from a few peers.   It all stems from a single assert I added to make sure we didn't download blocks from forks that were "too old".   I need to gather more info on this one.

2) We have a random crash for some delegates in the networking code.   This is an issue we have been tracking for a few days.  It happens enough in the wild, but is very difficult to reproduce on demand.   We have some ideas but nothing conclusive.  Delegates, please be vigilant and run your client in GDB and capture the P2P log file if you happen to catch this crash in progress.
https://github.com/BitShares/bitshares_toolkit/issues/649

3) Some people are reporting syncing issues. 

96% of the delegates are online and synced so the network is healthy.  I have downloaded the chain from scratch with a fresh build 2x now on OS X (command line client).  I am not sure why some people end up stuck on the same block time and time again.   Please of patient as we work to resolve these bugs that are difficult to reproduce. 

4) Sometimes when a large transaction is made the transaction ledger reports the wrong amount as being sent (receiver sees the right amount).
https://github.com/BitShares/bitshares_toolkit/issues/657