SRX Services Gateway
Highlighted
SRX Services Gateway

SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-29-2016 03:04 PM

Hi All, 

 

I have an issue on my 210 cluster when failing redundancy group 0. 

Everything fails over fine (IPsec and GRE tunnels) but the routing sub system seems to have a serious problem

BGP establishment takes a good 30 seconds +. I have tried to adjust hold timers, implement bfd etc with no luck,

 

Any ideas on this or help would be massively appreciated

 

5 REPLIES 5
Highlighted
SRX Services Gateway

Re: SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-29-2016 06:28 PM

Do you see this behavior with RG0 failover or RG1 failover. With RG1 failover we dont expect this behavior but with RG0 failover we may see this issue.

Can you confirm if BGP is running over VPNS ?

 

 

Thanks,
Suraj
Please Mark My Solution Accepted if it Helped, Kudos are Appreciated too
Highlighted
SRX Services Gateway

Re: SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-29-2016 11:54 PM

Hi, 

 

Thanks for the response.

I dont see it with any other redundancy group except R0. 

 

No MPLS VPN's, I simulated this in my lab without the IPsec and Gre as well. 

So just the BGP loopback peering across 2 links to another neighbor. 

 

The time it takes is extremely long, on average around 30 seconds for this to come up. If i static route traffic there are no issues whatsoever. 

I believe this is an issue with the actual routing sub-system coming up rather than BGP itself, as I have configured BFD, hold timers of 10 seconds etc with no success

 

Highlighted
SRX Services Gateway

Re: SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-30-2016 12:15 AM

Hi whiteac22,

 

RG0 failover is intrusive and we will have this kind of issues. There are customers using Graceful Restart option for better results, but its not supported with Branch SRX series.

I feel this behavior is accepted with SRX branch series RG0 failovers. You may open a ticket with JTAC to confirm the same.

Thanks,
Suraj
Please Mark My Solution Accepted if it Helped, Kudos are Appreciated too
Highlighted
SRX Services Gateway

Re: SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-30-2016 04:54 AM

Hi, 

 

Fair enough. Thanks!

 

I have got a slight work around, i want to use graceful restart to keep the routes on the peer active and im going to use a static route with a higher preference as a fail safe on one of my links (not ideal i know, but it will keep things up for that 30 seconds)

 

The problem is the srx actually closes the session before failing over, is there any way to stop this?

 

Highlighted
SRX Services Gateway

Re: SRX 210 Cluster delay on BGP when failing redundancy-group 0

‎06-30-2016 05:44 AM
SRX closes which session?
Also AFAIK, SRX210 cannot act as a graceful helper also.
Thanks,
Suraj
Please Mark My Solution Accepted if it Helped, Kudos are Appreciated too