linux-kernel - Re: [bug] stuck localhost TCP connections, v2.6.26-rc3+

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20080526141252.GA31352@elte.hu>
Date:	Mon, 26 May 2008 16:12:52 +0200
From:	Ingo Molnar <mingo@...e.hu>
To:	Ilpo J?rvinen <ilpo.jarvinen@...sinki.fi>
Cc:	LKML <linux-kernel@...r.kernel.org>,
	Netdev <netdev@...r.kernel.org>,
	"David S. Miller" <davem@...emloft.net>,
	"Rafael J. Wysocki" <rjw@...k.pl>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [bug] stuck localhost TCP connections, v2.6.26-rc3+


* Ingo Molnar <mingo@...e.hu> wrote:

> > ...but you can still tcpdump that particular flow once the situation 
> > is discovered to see if TCP still tries to do something, no? One 
> > needs to tcpdump couple of minutes at minimum. Also please get 
> > /proc/net/tcp for that flow around the same time.
> 
> ok, will try those.

i got lucky ... the bug just reproduced again - find the netstat and 
/proc/net/tcp output below. New config attached.

i also ran this for 15 minutes:

  [root@...ope ~]# tcpdump src and dst 10.0.1.15 -n
  tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
  listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes

but there was no output at all - i.e. TCP has not attempted to do 
anything to these connections.

there's a hung distcc task on the system, waiting for socket action 
forever:

  [root@...ope ~]# strace -fp 19578
  Process 19578 attached - interrupt to quit
  select(5, NULL, [4], [4], {82, 90000} <unfinished ...>

disturbing that task via strace did not change the state of the socket - 
and that's not unexpected as it's a select(). [TCP state might be 
affected if strace impacted a recvmsg or a sendmsg wait directly.]

i'll keep this system in its hung state for an hour, just in case you 
can think of anything else to extract out of it.

	Ingo

------------->
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address               Foreign Address             State      
tcp        0      0 10.0.1.15:22                10.0.1.16:42843             ESTABLISHED 
tcp        0 207232 10.0.1.15:37198             10.0.1.15:3632              ESTABLISHED 
tcp        0      0 10.0.1.15:22                10.0.1.16:36284             ESTABLISHED 
tcp    72283      0 10.0.1.15:3632              10.0.1.15:37198             ESTABLISHED 
Tcp:
    1931 active connections openings
    149 passive connection openings
    0 failed connection attempts
    0 connection resets received
    4 connections established
    436004 segments received
    98819 segments send out
    4 segments retransmited
    0 bad segments received.
    0 resets sent
UdpLite:
  sl  local_address rem_address   st tx_queue rx_queue tr tm->when retrnsmt   uid  timeout inode                                                     
   0: 00000000:0E30 00000000:0000 0A 00000000:00000001 00:00000000 00000000   100        0 7526 1 f6110640 300 0 0 2 -1                              
   1: 00000000:0016 00000000:0000 0A 00000000:00000000 00:00000000 00000000     0        0 7132 1 f61c8000 300 0 0 2 -1                              
   2: 0F01000A:0016 1001000A:A75B 01 00000000:00000000 02:000AF5CA 00000000     0        0 39283 4 e6826400 21 6 15 4 -1                             
   3: 0F01000A:914E 0F01000A:0E30 01 00032980:00000000 04:00001AAA 00000000   500        0 31436 2 f6113200 22 0 0 4 -1                              
   4: 0F01000A:0016 1001000A:8DBC 01 00000000:00000000 02:0009141B 00000000     0        0 10784 2 f6110000 21 4 0 4 -1                              
   5: 0F01000A:0E30 0F01000A:914E 01 00000000:00011A5B 00:00000000 00000000     0        0 0 1 f6115dc0 22 8 0 2 -1                                  

View attachment "config" of type "text/plain" (47962 bytes)