a huge issue exists in linux kernels such that ipsec traffic doesn't obey 
the non-blocking socket option when the ipsec connection is not 
established yet.  so trying to test services over an ipsec tunnel that 
goes down and can't be brought back up (remote system dead) will cause 
multi-minute delays during the zenstatus loop which will cause lots of 
other services to randomly fail/clear.  this happened to me today and took 
me a while to track down what was going on.

but, i found what seems to be a workaround.

echo 1 > /proc/sys/net/core/xfrm_larval_drop

and put:

net.core.xfrm_larval_drop = 1

into /etc/sysctl.conf to persist over reboots.

now i just need to see if this really fixes things once the remote site 
comes back up.  there is quite a lengthy discussion online about bad side 
effects of using this, but i think it will be ok for me and most other 
users.
_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to