[Openswan Users] netkey to klips error: EXPECTATION FAILED

John A. Sullivan III jsullivan at opensourcedevel.com
Mon Jun 16 11:00:20 EDT 2008


On Mon, 2008-06-09 at 23:19 -0400, Paul Wouters wrote:
> 
> > pluto[2610]: "Manchester2RAS" #1737: EXPECTATION FAILED
> > at /src/openswan-2.5.17/programs/pluto/kernel.c:2581:
> > kernel_ops->eroute_idle != NULL
> > 
> > As it says, we are running openswan 2.5.17 on RedHat Enterprise Linux 5.
> 
> Please switch to 2.6.14. We are no longer going to maintain the 2.5.17
> line. 2.6.14 is also what is shipped with RHEL 5.2
> 
> > I found very little information googling for this error other than a
> > recently closed bug report in openswan 2.6.  What's causing it and how
> > do I fix it in openswan 2.5.17? Thanks - John
> 
> kernel_netlink.c has a bunch of stubs for the kernel_ops functions in
> netkey, where before they used the same codepath as klips (sometimes with
> wrong assumtions such as klips doing group delete, but netkey not doing
> those).
> 
> Paul

Thanks, Paul.  That did the trick.  However, we are now picking up a
number of disturbing errors such as:

Jun 16 09:54:00 niagras1 pluto[2825]: "Buffalo2RAS" #1802: next payload
type of ISAKMP Hash Payload has an unknown value: 252

Jun 16 09:54:00 niagras1 pluto[2825]: "Buffalo2RAS" #1802: malformed
payload in packet

Jun 16 09:54:00 niagras1 pluto[2825]: "Buffalo2RAS" #1802: too many
(909522487) malformed payloads. Deleting state

Jun 16 09:31:39 niagras1 pluto[2825]: "Buffalo2RAS" #1637:
alloc_bytes1() was mistakenly asked to malloc 0 bytes for st_skey_ar in
duplicate_state, please report to dev at openswan.org

We suspect this may be due to enabling compression between the 2.6 and
the 2.4 devices.  We'll see what happens when our client disables
compression - John
-- 
John A. Sullivan III
Open Source Development Corporation
+1 207-985-7880
jsullivan at opensourcedevel.com

http://www.spiritualoutreach.com
Making Christianity intelligible to secular society



More information about the Users mailing list