https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4051
Gerfl <gkrames@xxxxxxx> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|OutOfMemoryError isn't being|Memory leak in TCP dissector
|caught at the top level |(OutOfMemoryError isn't
| |being caught at the top
| |level)
--- Comment #2 from Gerfl <gkrames@xxxxxxx> 2009-09-28 04:35:43 PDT ---
Thank you for this hint.
I have 4GB virtual memory, so the out of memory error at approx. 3.5GB is
explained.
BUT: Why does tshark run out of memory so quickly?
Refined test case:
- Script performing FTP uploads in an endless loop;
FTP server (vsftpd) is throtteled to 100 KByte/sec
(roughly 1 MBit/s on the wire).
- Capture filter is set to -f 'tcp and (host <ipOfServer)'
- Watching thsark's memory consumption (using "top") shows ever
memory consumption increasing.
- Extract from ~/.wireshark/preferences:
tcp.summary_in_tree: FALSE
tcp.check_checksum: TRUE
tcp.desegment_tcp_streams: FALSE
tcp.analyze_sequence_numbers: TRUE
tcp.relative_sequence_numbers: TRUE
tcp.track_bytes_in_flight: FALSE
tcp.calculate_timestamps: FALSE
tcp.try_heuristic_first: FALSE
ip.decode_tos_as_diffserv: TRUE # also tried FALSE, no difference
ip.defragment: TRUE
ip.summary_in_tree: FALSE
ip.check_checksum: TRUE
ip.tso_support: TRUE
- Disabled dissectors FTP and FTP-DATA.
- Out of memory crash after capturing 400MB.
The above taken together points to a memory leak somewhere in the
TCP dissector area.
Maybe a "TCP version" of Bug#3913 (where test case was UDP)?
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.