[lvs-users] Persistent connections not persisting after failover

Simon Horman horms at verge.net.au
Thu Aug 21 08:43:14 BST 2008


On Wed, Aug 20, 2008 at 10:35:21AM -0400, Nicholas Guarracino wrote:
> On Tue, Aug 19, 2008 at 9:28 PM, Simon Horman <horms at verge.net.au> wrote:

<snip>

> > If so, the key to persistence working is the persistence template.
> > It is an entry that goes into the connection table with 0 as the from-port
> > and acts as a parent entry for other connections from the same
> > host/netmask - I won't attempt to fail to explain that relationship
> > yet another time.
> >
> > You should see these templates being synchronised to the backup
> > director like any other connection. And you should be able to verify
> > this using either ipvsadm -Lcn or cat /proc/net/ip_vs_conn
> >
> > Can you verify that this is the case?
> > Could you look at the timeouts for the templates?
> 
> Yes, the templates are being synchronized.
> On the director:
> TCP 00:51  NONE        159.63.77.32:0     10.204.54.170:443  10.204.54.165:443
> TCP 00:54  NONE        159.63.77.16:0     10.204.54.170:443  10.204.54.166:443
> 
> On the backup director:
> TCP 02:51  NONE        159.63.77.32:0     10.204.54.170:443  10.204.54.165:443
> TCP 02:54  NONE        159.63.77.16:0     10.204.54.170:443  10.204.54.166:443
> 
> > I suspect that the problem is that the problem is that when
> > templates are synchronised their timeout is not synchronised.
> > Instead, at the other end it is set to 3 minutes. So perhaps they
> > are timing out and disapearing premeturely?
> >
> > I'm looking at the bottom of ip_vs_process_message() in
> > net/ipv4/ipvs/ip_vs_sync.c in 2.6.27-rc2
> >
> > Actually, I think that this works a little differently in 2.6.18,
> > but the affect seems to be much the same according to a quick
> > test that I ran.
> >
> > Clearly this is not ideal, and it would be nice to fix.
> > But does it explain your problem?
> 
> I'm not sure. My timeout is only 1 minute but it does look like when
> the template gets synchronized to the backup director the timeout
> becomes 3 minutes. That would be the opposite situation from what you
> described though, wouldn't it?

Yes, I would imagine so.

I wonder if for some reason the templates on the backup aren't being used.




More information about the lvs-users mailing list