Wireshark-bugs: [Wireshark-bugs] [Bug 10064] New: dumpcap regression: 1.10.6 says Capturing on '
Date: Wed, 30 Apr 2014 11:57:33 +0000
Bug ID 10064
Summary dumpcap regression: 1.10.6 says Capturing on 'Loopback' but doesn't for a while (race condition)
Classification Unclassified
Product Wireshark
Version unspecified
Hardware x86
OS Mac OS X 10.4
Status UNCONFIRMED
Severity Major
Priority Low
Component Capture file support (libwiretap)
Assignee bugzilla-admin@wireshark.org
Reporter jthackray@gmail.com

Build Information:
TShark 1.10.6 (v1.10.6 from master-1.10)

Copyright 1998-2014 Gerald Combs <gerald@wireshark.org> and contributors.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Compiled (64-bit) with GLib 2.39.91, with libpcap, with libz 1.2.8, with POSIX
capabilities (Linux), without libnl, with SMI 0.4.8, with c-ares 1.10.0, with
Lua 5.2, without Python, with GnuTLS 2.12.23, with Gcrypt 1.5.3, with MIT
Kerberos, with GeoIP.

Running on Linux 3.13.0-24-generic, with locale en_US.UTF-8, with libpcap
version 1.5.3, with libz 1.2.8.
      Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz

Built using gcc 4.8.2.
--
We use "dumpcap" as a testing tool, in preference to "tcpdump", because the
latter has a race condition, when it says "Capturing" and it isn't (for a
while). This means it can miss the start of packets being received.

dumpcap historically didn't have this issue. In Ubuntu 13.10, (dumpcap 1.10.2
SVNRev 51934 from /trunk-1.10) it prints "Capturing on 'Loopback'" and
_actually starts capturing_ from that moment on.

This avoids adding a "sleep 2" or some other random number of seconds into test
framework code that can be called thousands of times (all these sleep
statements add up, and they're non-deterministic)

However, since dumpcap 1.10.6, dumpcap now prints "Capturing on 'Loopback'" but
there's a few milli/microseconds when it misses packets.

Surely it can start capturing and then print "Capturing on 'Loopback'" or
whatever message. This output can then be used to start sending an HTTP (or
whatever) request for testing purposes.

Is there any other way to know reliably when dumpcap is _actually_ capturing?
This seems like a regression from earlier releases.

Many thanks,
Jon.


You are receiving this mail because:
  • You are watching all bug changes.