Bitcoin Forum
November 10, 2024, 02:18:30 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Bitcoin Forum > Bitcoin > Development & Technical Discussion > the bs "Satoshi:0.8.99" and 0 byte pings
Pages: [1]
« previous topic next topic »
  Print  
Author Topic: the bs "Satoshi:0.8.99" and 0 byte pings  (Read 1356 times)
zvs (OP)
Legendary
*
Offline Offline

Activity: 1680
Merit: 1000


https://web.archive.org/web/*/nogleg.com


View Profile WWW
November 03, 2013, 02:01:35 PM
 #1
Code:
2013-11-03 05:09:44 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:10:14 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:10:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:11:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:11:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:12:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:12:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:13:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:13:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:14:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:14:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:15:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:15:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:16:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:16:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:17:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:17:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:18:02 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length

<snip>

2013-11-03 11:33:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:34:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:34:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:35:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:35:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:36:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:36:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:37:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:37:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:38:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:38:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:39:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:39:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:40:03 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:40:33 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:41:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:41:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:42:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:42:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:43:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:43:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:44:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:44:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:45:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:45:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:46:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:46:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:47:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:47:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:48:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:48:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:49:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:49:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:50:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:50:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:51:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:51:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:52:04 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 11:52:34 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length

They also all have:

ProcessMessage(ping, 0 bytes) FAILED immediately following that message.

This is where it started:

Code:
2013-11-03 05:11:27 accepted connection 176.9.144.41:32687
2013-11-03 05:11:27 accepted connection 46.4.64.21:56061
2013-11-03 05:11:27 accepted connection 5.9.30.207:42607
2013-11-03 05:11:27 accepted connection 144.76.70.73:54425
2013-11-03 05:11:27 accepted connection 144.76.102.176:30757
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=176.9.144.41:32687, peer=176.9.144.41:32687
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=176.9.144.41:32687
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=46.4.64.21:56061, peer=46.4.64.21:56061
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=46.4.64.21:56061
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=5.9.30.207:42607, peer=5.9.30.207:42607
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=5.9.30.207:42607
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=144.76.70.73:54425, peer=144.76.70.73:54425
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=144.76.70.73:54425
2013-11-03 05:11:27 accepted connection 5.9.110.78:55205
2013-11-03 05:11:27 accepted connection 5.9.245.121:60001
2013-11-03 05:11:27 accepted connection 88.198.41.74:39240
2013-11-03 05:11:27 accepted connection 5.9.203.20:43255
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=144.76.102.176:30757, peer=144.76.102.176:30757
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=144.76.102.176:30757
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=5.9.110.78:55205, peer=5.9.110.78:55205
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=5.9.110.78:55205
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=5.9.245.121:60001, peer=5.9.245.121:60001
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=5.9.245.121:60001
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=88.198.41.74:39240, peer=88.198.41.74:39240
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=88.198.41.74:39240
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=5.9.203.20:43255, peer=5.9.203.20:43255
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=5.9.203.20:43255
2013-11-03 05:11:27 accepted connection 91.121.58.230:55388
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=91.121.58.230:55388, peer=91.121.58.230:55388
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=91.121.58.230:55388
2013-11-03 05:11:27 accepted connection 199.193.117.231:20562
2013-11-03 05:11:27 send version message: version 70001, blocks=267626, us=5.9.24.81:8333, them=199.193.117.231:20562, peer=199.193.117.231:20562
2013-11-03 05:11:27 receive version message: version 70001, blocks=1, us=127.0.0.1:8333, them=0.0.0.0:0, peer=199.193.117.231:20562
2013-11-03 05:11:32 ProcessMessages(ping, 0 bytes) : Exception 'CDataStream::read() : end of data' caught, normally caused by a message being shorter than its stated length
2013-11-03 05:11:32 ProcessMessage(ping, 0 bytes) FAILED

The thing is, I don't want to firewall these nodes as I receive blocks from some of them.  Anyone have any clue as to wtf the deal is?   Is there a way to stop this w/o firewalling the nodes themselves?  Does the message even matter?  Can I just ed -s debug.log <<< $'g/ProcessMessage/d\nw' it out of there or what?

It's the 'BS Satoshi 0.8.99' nodes.
wtogami
Sr. Member
****
Offline Offline

Activity: 263
Merit: 250



View Profile
November 04, 2013, 07:17:29 AM
 #2
https://github.com/forrestv/p2pool/issues/137
Apparently we know one way the zero byte pings can happen.  It happened by accident with a recent p2pool change.  In this 0.8.99 instance however it may not be an accident...
If you appreciate my work please consider making a small donation.
BTC:  1LkYiL3RaouKXTUhGcE84XLece31JjnLc3      LTC:  LYtrtYZsVSn5ymhPepcJMo4HnBeeXXVKW9
GPG: AEC1884398647C47413C1C3FB1179EB7347DC10D
Krellan
Member
**
Offline Offline

Activity: 106
Merit: 10


View Profile
November 04, 2013, 08:17:01 AM
 #3
Strange!

I worked on a recent patch for Bitcoin that added ping time measurement (using ping/pong).  During development, I ran into some strange nodes out there that were doing something similar.

https://github.com/bitcoin/bitcoin/pull/2937#issuecomment-23457819

Here's the referenced IRC chat:

http://bitcoinstats.com/irc/bitcoin-dev/logs/2013/08/28#l1377673880

Many of these IP addresses match!!

Only thing is, this was for the *pong* message, not ping.  They were sending me pong messages of zero size.  I now check the size before doing any processing, and just throw it out if it's too small, to avoid the spam messages you get about the exception.  I don't remember ever getting any undersized *ping* messages from them.

I remember the version also, it was 0.8.99.  I think this just means you're running the latest Bitcoin from GitHub, compiled yourself, and thus not an official release?  Not sure about that, though.

Didn't think this had anything to do with p2pool, though.  Interesting.

Josh
1JUZr4TZ5zuB4WdEv4mrhZMaM7yttpJvLG Smiley
wtogami
Sr. Member
****
Offline Offline

Activity: 263
Merit: 250



View Profile
November 04, 2013, 09:55:57 AM
 #4
Quote from: Krellan on November 04, 2013, 08:17:01 AM
Strange!

I worked on a recent patch for Bitcoin that added ping time measurement (using ping/pong).  During development, I ran into some strange nodes out there that were doing something similar.

https://github.com/bitcoin/bitcoin/pull/2937#issuecomment-23457819

Here's the referenced IRC chat:

http://bitcoinstats.com/irc/bitcoin-dev/logs/2013/08/28#l1377673880

Many of these IP addresses match!!

Only thing is, this was for the *pong* message, not ping.  They were sending me pong messages of zero size.  I now check the size before doing any processing, and just throw it out if it's too small, to avoid the spam messages you get about the exception.  I don't remember ever getting any undersized *ping* messages from them.

I remember the version also, it was 0.8.99.  I think this just means you're running the latest Bitcoin from GitHub, compiled yourself, and thus not an official release?  Not sure about that, though.

Didn't think this had anything to do with p2pool, though.  Interesting.

Josh


p2pool only connects to a local node and pretends to be another client to getblocks.  None of these bad peers are p2pool itself.

Hmm, with your commit 0.9 users will have no visible way of seeing these protocol breaking bad nodes.  Perhaps a -debug mode only warning would be helpful?
If you appreciate my work please consider making a small donation.
BTC:  1LkYiL3RaouKXTUhGcE84XLece31JjnLc3      LTC:  LYtrtYZsVSn5ymhPepcJMo4HnBeeXXVKW9
GPG: AEC1884398647C47413C1C3FB1179EB7347DC10D
zvs (OP)
Legendary
*
Offline Offline

Activity: 1680
Merit: 1000


https://web.archive.org/web/*/nogleg.com


View Profile WWW
November 04, 2013, 12:27:22 PM
 #5
I don't see this behavior with a second node that has 50 connections (& same client version as other server), runs p2pool, & has listen=0 ...  also doesn't appear on my local node that has a single bitcoind connection & runs p2pool

I am thinking you won't see this message unless these odd peers are connected to you *and* you're running p2pool?

I know it'd be caused by (ed: whatever interaction is occuring with):

p2p.py:        self.pinger.start(30)
danneu
Newbie
*
Offline Offline

Activity: 32
Merit: 0



View Profile
November 05, 2013, 08:27:06 PM
 #6
Quote from: zvs on November 03, 2013, 02:01:35 PM
The thing is, I don't want to firewall these nodes as I receive blocks from some of them.

Inv-sending nodes aren't exactly a rarity, so I don't understand your concern.

Frankly, nodes are far too lenient with the messages they respond to or accept.

That includes 0-byte pings and spam.
Krellan
Member
**
Offline Offline

Activity: 106
Merit: 10


View Profile
November 06, 2013, 07:28:36 AM
 #7
Quote from: wtogami on November 04, 2013, 09:55:57 AM
Hmm, with your commit 0.9 users will have no visible way of seeing these protocol breaking bad nodes.  Perhaps a -debug mode only warning would be helpful?

Aye, you're right.  I have that in already, but only for pong.  Don't have it for ping.  Should be easy enough to add.

Josh
1JUZr4TZ5zuB4WdEv4mrhZMaM7yttpJvLG Smiley
Pages: [1]
  Print  
Bitcoin Forum > Bitcoin > Development & Technical Discussion > the bs "Satoshi:0.8.99" and 0 byte pings
 « previous topic next topic »
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!