[Swan-dev] what git commit or patch fixes the maximum number of retransmit reached problem after libreswan 3.15

Paul Wouters paul at nohats.ca
Fri Jun 3 21:25:52 UTC 2016


On Fri, 3 Jun 2016, CHEN, JIANFU wrote:

>  
> I am current using libreswan 3.15. And found that after libreswan restarts, if the peer side not reachable after about 1 minutes, it gives a message
> Max number of retransmission reached, and will not try to connect to the peer side anymore even if later the peer side is reachable. The message is shown as below:
>  
> Jun  3 16:28:28 R9 pluto[23315]: "net22" #1: initiating Main Mode
> Jun  3 16:29:32 R9 pluto[23315]: "net22" #1: max number of retransmissions (8) reached STATE_MAIN_I1.  No response (or no acceptable response) to our first IKEv1 message
> Jun  3 16:29:32 R9 pluto[23315]: "net22" #1: deleting state #1 (STATE_MAIN_I1)
>  
> I tried the libreswan 3.17, and found the problem has been fixed. But for some reason, we plan to still use libreswan 3.15. But plan to fix this problem to add a patch based on libreswan
> 3.15.
>  
> Does anyone know which patch or git commit fixes this problem.
>  
> Thanks,
>  
> Jeff

commit 3c8dc46d53e3e5004b88f30b5ec3d06d5337951c
Author: Wolfgang Nothdurft <wolfgang at linogate.de>
Date:   Wed Nov 11 14:40:03 2015 +0100

     pluto: fix keyingtries=0 doesn't try forever

     Since the change in 11fbe4b1182a06b6794575405639b9727ffdd23b pluto
doesn't try forever with keyingtries=%forever (keyingtries=0), because
try <= try_limit can never match.

     Signed-off-by: Tuomo Soini <tis at foobar.fi>

Paul


More information about the Swan-dev mailing list