[j-nsp] Trouble with just one IPSec tunnel among many
sfouant at shortestpathfirst.net
Tue Nov 17 22:08:58 EST 2015
Have you tried debugging under [edit security flow traceoptions] yet? There is a wealth of information in there using the flag 'basic-datapath' that may help you.
JNCIE-SEC, JNCIE-SP, JNCIE-ENT, JNCI
m (703) 625-6243
> On Nov 17, 2015, at 9:42 PM, Jonathan Call <lordsith49 at hotmail.com> wrote:
> I have an SRX250 (SRX A) and an SRX240h2 (SRX B) connected via a PSK IPSec tunnel. They both have multiple IPSec tunnels configured to other SRX devices on our network. Recently the tunnel between the two stopped passing traffic. Both IKE and IPSec security association were UP on both sides. (show security ike security-association and show security ipsec security-association) If I pinged from SRX A to something on SRX B I would see the echo request come into SRX B and the echo reply go out. On SRX A I would only see the outbound packet flow with no response. When I would ping from SRX B to SRX A, SRX A acted as if nothing was happening at all. I ran a traceoptions to capture the ICMP traffic from SRX A to SRX B. SRX A looked like it was sending the traffic out the tunnel interface but had no reply content. The tracefile on SRX B was completely empty.
> After confirming there was nothing abnormal in the routing tables on both of them I tried clearing the IKE and IPSec sessions to rebuild the VPN. Both Phase 1 IKE and Phase 2 IKE came back up but now ping fails in both directions. When I look at the 'show security ipsec statistics index' information on each side SRX B shows some ESP packets being encrypted and decrypted but not very many. The SRX A ESP statistics are all zeros.
> Restarting ipsec key management seems a bit extreme since it will take out all of the other IPSec sessions. Is there some other troubleshooting I can try before I resort to that?
> Thank you,
> juniper-nsp mailing list juniper-nsp at puck.nether.net
More information about the juniper-nsp