kernel: llt info v-14-1-10205 link 1 bond0 node 1 trouble
kernel: LLT INFO V-14-1-10024 link 1 (bond0) node 1 active



Need to increase the peertrouble.


# lltconfig -T query


Current LLT timer values (.01 sec units):
heartbeat = 50
heartbeatlo = 100
peertrouble = 200 <------- before
peerinact = 1600
oos = 10
retrans = 10
service = 100
arp = 30000




# more */etc/llttab

::::::::::::::
set-node ba08
set-cluster 10
link bond1 eth-00:ba:ba:ba:ba:02 - ether - -
link-lowpri bond0 eth-00:ba:ba:ba:ba:01 - ether - -



Link1 (bond0) is the lo-pri link. LLT heartbeats once every second on lo-pri
links and twice every second for normal links. Typically, since the lo-pri
links are public links, LLT does not send any other traffic on it.

The public network on which the lo-pri link is connected has larger
network delays.

When heartbeats are missed for more than "peertrouble" number of seconds, LLT
throws the message "link in trouble". Since we are seeing some network delays
on the lo-pri link, Increse the peertrouble time to say 4 seconds (on all
nodes) by the following command:

lltconfig -T peertrouble:400 # to be fired on all nodes



# lltconfig -T query

Current LLT timer values (.01 sec units):
heartbeat = 50
heartbeatlo = 100
peertrouble = 400 <------- After
peerinact = 1600
oos = 10
retrans = 10
service = 100
arp = 30000




Solution:


To make it the parameter persistent during the reboot,
add the following line in llttab on all nodes



vi /etc/llttab and add the line below.

set-timer peertrouble:400


 

 

kernel: llt info v-14-1-10205 link 1 bond0 node 1 trouble