I forgot to include the some details/configuration ...here they are..
Here's some of the configuration/infos: Remote Office server: OS: CentOS release 5.4 (Final) Tikanga IPSec: openswan-2.6.21-5.el5_4.2 Quagga/BGP: quagga-0.98.6-5.el5 HQ Server: OS: Fedora Core release 6 (Zod) Tikanga IPSec: openswan-2.4.9-1 Quagga/BGP: quagga-0.99.7-1.fc6 config setup protostack=netkey nat_traversal=yes virtual_private= oe=off nhelpers=0 conn nyctomtl authby=secret pfs=yes left=a.a.a.175 leftnexthop=a.a.a.161 right=b.b.b.2 rightnexthop=b.b.b.1 auto=start esp=aes128-sha1 config setup include /etc/ipsec.d/examples/no_oe.conf conn mtltonyc authby=secret pfs=yes left=a.a.a.175 leftnexthop=a.a.a.161 right=b.b.b.2 rightnexthop=b.b.b.1 auto=start esp=aes128-sha1 keylife=4h ikelifetime=8h On Thu, 12 Aug 2010 11:52:13 -0400, itnews wrote: For the past few week we're experiencing issue with 2 of our offices. Each office run linux as firewall with a gre tunnel over IPSec to our MTL office. We also have bgp/quagga running over the gre. At least once per week for a period of about 10 minute the remote office is unable to access ressource over tunnel. What's weird, is that during the downtime period we can ping successfully ressource as well as telnet to the ressource's port. However if we try to initiate a full communication to the remote host, then the session hang forever, until the the we're back online. Looking at the logs show nothing during the time of the issue and the server's load/memory and bandwidth consumption is fine. Has anyone else experience this type of problem.
_______________________________________________ mlug mailing list [email protected] https://listes.koumbit.net/cgi-bin/mailman/listinfo/mlug-listserv.mlug.ca
