https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6293
--- Comment #5 from Guy Harris <guy@xxxxxxxxxxxx> 2011-09-01 23:34:14 PDT ---
That's not at all correctly handling TCP segmentation. The length of the
tvbuff handed to the routine calling tcp_dissect_pdus() does not, in any way
whatsoever, reflect the lengths of any packets contained, in whole or in part,
in the tvbuff; tcp_dissect_pdus() can *ONLY* be used in protocols where all
packets are at least N bytes long, for some value of N, and the contents of the
first N bytes can be used, in *ALL* cases, to determine the length of the
packet.
Unfortunately, not having a protocol specification for the HDFS wire protocol,
I don't know whether that's the case. This HDFS issue:
https://issues.apache.org/jira/browse/HDFS-1123
is still unresolved, so there doesn't seem to be a complete protocol
specification, unless this is some other protocol, as per the over-a-year-ago
comment
Todd Lipcon added a comment - 30/Apr/10 05:08
I agree that we should do a better job of documenting the current protocol,
but we shouldn't spend too much time on it, since everyone is in agreement that
we'd like to move to Avro this year. A quick pass to update the comments is
probably worth doing, but a formal spec may be overkill for a protocol we plan
to deprecate imminently.
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.
