Peerplays_Blockchain/peerplays_migrated
1
0
Fork 0
Code Issues 35 Pull requests 4 Projects Releases 20 Packages Wiki Activity Actions
1447 commits 435 branches 47 tags 121 MiB
Commit graph

50 commits

Author SHA1 Message Date
abitmore
5ad763b0bf Catch all exception when connecting to other peers 2016-09-06 20:57:01 +02:00
abitmore
aa844f56fa Fix seed node stuck issue #654 2016-05-16 22:03:08 +02:00
Nathan Hourt
b2835bc02d Add headers to install targets 2016-03-17 17:16:13 -05:00
theoreticalbts
2666350978 Fix iteration logic in _handle_message_calls_in_progress shutdown loop to handle concurrent modification #598 2016-02-25 11:31:08 -05:00
Fabian Schuh
5111084730 [License] Modify cpp and hpp headers to MIT #496 2016-01-07 11:44:52 -05:00
theoreticalbts
d0609e8e74 Update copyright notice for BitShares 2015-10-12 13:48:40 -04:00
theoreticalbts
60f0812de2 Update copyright notice 2015-10-12 13:32:47 -04:00
Eric Frias
7bc7004ef9 Remove items from p2p's list of items to fetch when we expect them to become unfetchable 2015-10-08 15:17:01 -04:00
Daniel Larimer
bb0ea77e80 throttling P2P code 2015-10-05 08:44:59 -04:00
Eric Frias
74bbde785b In the p2p networking code, request multiple transactions at a time to improve performance during flooding 2015-09-29 17:37:32 -04:00
Eric Frias
405a9e71b2 When the p2p code pushes a transaction that fails to apply, remember that transaction so we don't fetch it again as soon as a different peer tells us about it. 2015-09-22 17:13:35 -04:00
Eric Frias
d92736617e Improve logging of exceptions in p2p code (log via reflection instead of converting to strings) 2015-09-22 17:13:35 -04:00
Eric Frias
5bfbe0df86 When a we're in normal (in-sync) mode and we get a block that doesn't link to the chain, restart sync with that peer to try to get the missing block 2015-09-21 16:15:51 -04:00
Eric Frias
9536350a50 Reduce the amount of time we'll wait for a peer to provide a requested block to 
one second (down from one full block interval), to reduce the chance of us getting
blocks out of order.
Remove the variable we use to keep track of the last block number a peer has;
This was an optimization in BitShares 0.x where looking up the number was expensive,
and maintaining it is error-prone.
 2015-09-21 11:48:58 -04:00
theoreticalbts
a2b6f1a741 Change assert to elog #334 2015-09-21 11:01:11 -04:00
Daniel Larimer
a5c1bc7faf Merge branch 'master' of https://github.com/cryptonomex/graphene 2015-09-21 08:46:57 -04:00
Daniel Larimer
6b915054d2 potential fix for peer not disconnecting after invalid block 2015-09-21 08:46:55 -04:00
Eric Frias
cdd5cae385 Add extra error reporting when we get a list of sync blocks that don't make sense, and replace the assert 
with a log+disconnect.
 2015-09-19 17:08:57 -04:00
Eric Frias
e5777faa7d Reduce the disconnection timeout in the p2p code for when a peer offers us a block and doesn't respond, now it's one block interval, down from three. This is intended to reduce the chance of a slow peer causing us to push blocks out of order. 2015-09-18 09:39:43 -04:00
Eric Frias
0046cc0698 Fix error messages generated when peer is syncing with empty blockchain 2015-09-11 15:26:55 -04:00
Eric Frias
1141d1ce89 Disable log messages that logged every step in the process of queueing and sending messages to peers, they were seldom useful. 2015-09-11 11:17:52 -04:00
Eric Frias
f9a27059e5 Fix incorrect error message generated when syncing with peer that has no blocks 2015-09-11 10:09:42 -04:00
Eric Frias
8e9bd890a8 Fix bugs, improve logging in p2p sync 2015-09-10 19:33:11 -04:00
Eric Frias
928b6934e0 Merge branch 'master' into p2p_sync_fixes 
Conflicts:
	libraries/chain/include/graphene/chain/database.hpp
 2015-09-10 17:34:20 -04:00
Eric Frias
b6bba74301 Restore much of the syncing code from BitShares 0.x, and add optimizations that prevent us from attempting to sync with peers those blockchain diverged from ours many blocks ago, which also reduces the size of sync messages. Add logging. 2015-09-10 08:56:29 -04:00
theoreticalbts
8ebc1cfc43 node.cpp: Don't dump empty iterators 2015-09-08 16:08:11 -04:00
Eric Frias
a943de8b76 When rolling back on save, remove the rolled back blocks from our 
databases so when we restart and begin syncing, we re-download and
push those rolled back blocks  (fixes bug introduced in
a5071f2568).
Fix logging of incorrect block numbers in p2p log.
 2015-09-04 15:22:12 -04:00
Eric Frias
5662a8423c Increase duration of message cache from two blocks (which in practice meant anything > 1 block) to 30 blocks, and move the constant to config.hpp 2015-09-02 17:26:26 -04:00
Eric Frias
d35e2d6d4e reorder the p2p network's terminate_inactive_connections_loop() to prevent us 
from being interrupted before we delete an unresponsive peer #288
 2015-09-02 12:15:02 -04:00
Daniel Larimer
2464b788ad improve reindexing performance 2015-08-24 17:57:44 -04:00
Eric Frias
c0b9af9a99 Greatly reduce the amount of time the p2p network code will wait for a peer to 
return a requested block/transaction.  Make this time dependent on the actual block
interval.  This should allow the the node to give up and request the block from
another peer before the ~30 second undo interval has passed.
Fix the merkle root calculation to avoid reading
past the end of a vector.  Modify the algorithm to do what was likely intended
(this modification is currently disabled because it will yield different results
than the currently-running testnet)
Fix windows build errors.
 2015-08-24 09:57:47 -04:00
theoreticalbts
2d1e76aed0 Get rid of chain ID global, move chain_id_type to types.hpp 2015-08-06 12:56:40 -04:00
theoreticalbts
2e9876b557 Implement chain-locked transactions 2015-08-06 12:56:37 -04:00
theoreticalbts
c7444f8d4d Implement minimal "make install" command #219 2015-08-06 12:40:53 -04:00
Vikram Rajkumar
d0b8c66aad Fix Linux Clang build with missing includes; resolve #174 2015-07-20 14:57:08 -04:00
Eric Frias
11a5d2b620 When the p2p code processes a block that contains transactions 
we haven't seen, avoid fetching those transactions separately
 2015-07-16 15:31:52 -04:00
Vikram Rajkumar
d9855f9023 Clean up some TODOs; #77 2015-07-15 18:15:41 -04:00
Vikram Rajkumar
6043586d15 Rename bad references to delegate -> witness; #147 2015-07-13 16:08:17 -04:00
Vikram Rajkumar
12d8f3732e Minor 2015-07-13 16:08:16 -04:00
theoreticalbts
e201e59b3d Define GRAPHENE_MIN_BLOCK_INTERVAL and use it to fix minimum_time_of_last_offered_block logic #154 2015-07-13 12:33:52 -04:00
Daniel Larimer
04b8820b3b Merge branch 'master' into fee_refactor 2015-07-09 16:34:56 -04:00
theoreticalbts
ae7b904e4a Clean up some unused files 2015-07-09 11:59:06 -04:00
Daniel Larimer
fdcf821c41 Progress on #144 - refactoring fee schedule 2015-07-08 18:45:53 -04:00
Eric Frias
609f5a746e Quiet harmless compiler warnings 2015-07-01 11:16:25 -04:00
Eric Frias
49acfb3bd6 Fix up application_impl's has_item() which was throwing to indicate a 
missing item instead of returning false.  #55
 2015-06-25 10:36:28 -04:00
Eric Frias
37cadb2be8 Create directory for storing p2p config if it doesn't already exist 2015-06-24 18:21:20 -04:00
Eric Frias
11b357494b Remove leveldb code from peer database, replace with writing to json 
file.  Still need to add code to periodically write the file out to
disk.
 2015-06-24 18:21:20 -04:00
Eric Frias
1b9ca9ffe8 Add wallet command for whitelisting accounts. 
Re-enable shutdown in node code which was causing some (not all) of the
errors in two_node_network test #55, and likely occasional crashes at
shutdown for the witness node
 2015-06-20 18:36:02 -04:00
Daniel Larimer
ca89d5057c remove dependency on level db, bugs left to fix 2015-06-16 15:56:13 -04:00
Nathan Hourt
0a945b0257 Initial commit 2015-06-08 12:36:37 -04:00