| 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 <2018> 2019 2020 2021 2022 2023 2024 | Index | 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 <2018> 2019 2020 2021 2022 2023 2024 |
| <== Date ==> | <== Thread ==> |
|---|
| Subject: | Re: Port scan with nmap causes infinite loop in casDGClient::processDG() [Re: CA gatway runs away when zero length PV name in UDP search request] |
| From: | Michael Davidsaver <[email protected]> |
| To: | "Hartman, Steven M." <[email protected]>, Benjamin Franksen <[email protected]> |
| Cc: | EPICS Tech Talk <[email protected]> |
| Date: | Tue, 23 Jan 2018 13:44:38 -0800 |
On 01/23/2018 10:54 AM, Hartman, Steven M. wrote: > Nonetheless, a malformed packet crashing a server would be considered in bug in the server implementation and should be fixed. I don't think anyone is going to argue that these sort of issues shouldn't be fixed. The problem is as usual a question of time and/or money. Actively finding and _fixing_ packet validation issues has never been a priority for anyone. FYI, if someone could spend time on this, a place to start might be: https://github.com/mdavidsaver/catvs which is a framework I started for verifying consistency between CA implementations. This works be constructing packets with a python script. It is straightforward to create invalid/corrupt messages. A test case for zero length PVs could be added here https://github.com/mdavidsaver/catvs/blob/master/catvs/server/test_search.py#L16