[j-nsp] Weird SRX flow timeout issue
Andrew Yager
andrew at rwts.com.au
Mon Nov 12 06:50:49 EST 2012
Hi,
We're working with a client on a strange issue with an SRX.
The client has a Postgres application that regularly runs long queries across a route-based IPSEC VPN connection (taking several hours to return a result).
By default the SRX closes the flow after 30 minutes (1800 seconds) as there is no activity on the wire during this time.
We created a custom application object and applied it to the VPN ruleset, and after one hour it closes all sessions that match the application object, and any new flows are created with a 20 second flow-timeout, completely ignoring the timeout rule on the application definition until a reboot.
e.g. default behaviour:
>show security flow session destination-port 5432
Session ID: 1047, Policy name: vpn-usa2-out-postgres/7, Timeout: 1800, Valid
In: 10.2.2.5/49354 --> 192.168.2.10/5432;tcp, If: vlan.3, Pkts: 1370, Bytes: 91835
Out: 192.168.2.10/5432 --> 10.2.2.5/49354;tcp, If: ge-0/0/15.0, Pkts: 2518, Bytes: 3441234
Total sessions: 1
working for the first hour:
Session ID: 27259, Policy name: vpn-usa2-out-postgres/7, Timeout: 40000, Valid
In: 10.2.2.5/49227 --> 192.168.2.10/5432;tcp, If: vlan.3, Pkts: 2, Bytes: 120
Out: 192.168.2.10/5432 --> 10.2.2.5/49227;tcp, If: ge-0/0/15.0, Pkts: 0, Bytes: 0
Total sessions: 1
after the first hour (on a brand new session)
Session ID: 29151, Policy name: vpn-usa2-out-postgres/7, Timeout: 20, Valid
In: 10.2.2.5/49214 --> 192.168.2.10/5432;tcp, If: vlan.3, Pkts: 3, Bytes: 180
Out: 192.168.2.10/5432 --> 10.2.2.5/49214;tcp, If: ge-0/0/15.0, Pkts: 0, Bytes: 0
Total sessions: 1
All subsequent sessions are crated with a 20 second timeout.
Config snippets:
[applications]
application postgres {
protocol tcp;
destination-port 5432;
}
[security policies from-zone untrust to-zone trust]
policy vpn-usa-in-postgres {
match {
source-address US1;
destination-address Local;
application postgres;
}
then {
permit {
tunnel {
ipsec-vpn USA;
pair-policy vpn-usa-out-postgres;
}
}
}
}
(and corresponding pair-policy).
Any thoughts on whether this is a JunOS bug or config error?
Thanks,
Andrew
--
Andrew Yager, Managing Director (MACS Snr CP BCompSc MCP MCE JNCIA-Junos)
Real World Technology Solutions Pty Ltd - IT people you can trust
ph: 1300 798 718 or (02) 9037 0500
fax: (02) 9037 0591 mob: 0405 152 568
http://www.rwts.com.au/
More information about the juniper-nsp
mailing list