Thread Info
  • State Not Answered
  • Locked Locked
  • Replies 2 replies
  • Subscribers 3 subscribers
  • Views 4315 views
  • Users 0 members are here
Options
Suggested
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Puzzling High-Availibility log

SteveHart

We have two UTMs configured in High Availability mode. Thinks seem to be working OK for now.

We're getting this pretty much continuously in our High-Availability log.

2016:03:10-00:00:05 perimeter1-2 ha_daemon[4301]: id="38A3" severity="debug" sys="System" sub="ha" seq="S:  394 05.320" name="Netlink: Found link beat on eth2 again!"
2016:03:10-00:00:05 perimeter1-2 ha_daemon[4301]: id="38A1" severity="warn" sys="System" sub="ha" seq="S:  395 05.320" name="All monitored interfaces with link again!"
2016:03:10-00:00:05 perimeter1-2 ha_daemon[4301]: id="38A0" severity="info" sys="System" sub="ha" seq="S:  396 05.320" name="state change UNLINKED(1) -> ACTIVE(0)"
2016:03:10-00:00:05 perimeter1-1 ha_daemon[4275]: id="38A0" severity="info" sys="System" sub="ha" seq="M:  968 05.880" name="Node 2 changed state: UNLINKED(1) -> ACTIVE(0)"
2016:03:10-00:00:06 perimeter1-2 ha_daemon[4301]: id="38A3" severity="debug" sys="System" sub="ha" seq="S:  397 06.442" name="Netlink: Lost link beat on eth2!"
2016:03:10-00:00:09 perimeter1-2 ha_daemon[4301]: id="38A1" severity="warn" sys="System" sub="ha" seq="S:  398 09.443" name="No link on interface eth2"
2016:03:10-00:00:09 perimeter1-2 ha_daemon[4301]: id="38A0" severity="info" sys="System" sub="ha" seq="S:  399 09.443" name="state change ACTIVE(0) -> UNLINKED(1)"
2016:03:10-00:00:09 perimeter1-1 ha_daemon[4275]: id="38A0" severity="info" sys="System" sub="ha" seq="M:  969 09.884" name="Node 2 changed state: ACTIVE(0) -> UNLINKED(1)"

Does this imply a bad network cable, bad switch or something else?



This thread was automatically locked due to age.
  • 0

    Yes, UNLINKED means an interface link is down. So, if you were to do ethtool eth2 you will see the "Link detected" value will be "No" This normally indicates a cabling issue, or check your switch.

  • 0

    Hi,

    the unlinked status is a status to determine the device with the highest connectivity.

    Lets say

    Node 1  => Currently master, uses eth0, 1 and 4

    Node 2 => Currently slave, uses eth0, 1 and 4

    so both are equally connected.

    Now the master looses Link on eth4, which means its connectivity is lower than that of Node 2.

    This determines the Node 1 as unlinked.

    HA reacts to this by switching the nodes to ensure connectivity.

    The reasons could be seen in the kernel log. Search for interface or link down status.

    If there are, see if you can find issues on other equipment that could influence this status.

    The things you have mentioned, like defective cables, switches and similar issues could influence this.

    If you can't find issues on the network devices / hardware that connects to the interface that is problematic,

    open a case with support so we can investigate if there are hardware issues. 

    The LEDs on the interface can also be an indicator if the whole interface has issues. If both LEDs are dead,

    its a strong indicator for device issues. If only the red one is blinking it means the interface is up but the link is down.

     

    Hope this helps. Cheers!

Unfiltered HTML