This is sorted out.  I am in the process of migrating from another IPsec 
solution to strongSwan and therefore I have both products installed 
side-by-side for the purpose of comparing behaviors.  The old product is 
installed at /usr while strongSwan is installed at /usr/local, for now.  Both 
products have an 'ipsec' command.  It happens that the PATH environment 
variable is being altered when pluto is started such that /usr/sbin precedes 
/usr/local/sbin, and pluto runs the _updown script using 'ipsec _updown' 
(reliance on PATH).  Hence the 'wrong' ipsec command was being called to start 
the _updown script.


Bill


-----Original Message-----
From: [email protected] on behalf of 
William Bloom
Sent: Tue 7/6/2010 12:53 PM
To: [email protected]
Subject: [strongSwan-dev] Netlink Error Prevents _updown Execution
 

Using strongSwan 4.4.0 on RHEL 5.1, I've noticed that the _updown script isn't 
being executed after phase 1 and phase 2 complete successfully in pluto.  With 
plutodebug=all, I see that, in fact, 'ipsec _updown' command lines are indeed 
being executed.  However, an strace log reveals that the 'ipsec _updown' 
subprocess is exiting before starting the _updown script due to a NETLINK error 
(a 'sendto()' using NETLINK address family reports connection failure).  The 
only use of NETLINK with which I am acquainted is the Linux audit subsystem - 
but auditing appears to be configured properly and working fine on my system.

Is there a configuration flaw or build procedure misstep that might explain 
this?


Bill

_______________________________________________
Dev mailing list
[email protected]
https://lists.strongswan.org/mailman/listinfo/dev


_______________________________________________
Dev mailing list
[email protected]
https://lists.strongswan.org/mailman/listinfo/dev

Reply via email to