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

Up2date installation on HA killed slave

Hi

After investigating on how to do the Up2Date-installation in an active/passive HA environment I finally felt safe doing it.

15 minutes after clicking "Upgrade to latest version now" I ended up with a DEAD slave. Now it has passed 30 minutes and still the SLAVE is dead.

The log says:

2010:02:17-22:14:35 (none) ha_daemon[3539]: id="38A2" severity="error" sys="System" sub="ha" name="Node 2 died during up2date process!"
2010:02:17-22:14:35 (none) ha_daemon[3539]: id="38C1" severity="info" sys="System" sub="ha" name="Node 2 is dead, received no heart beats!"
2010:02:17-22:14:37 (none) slon_control[3668]: Killing slon reporting [21816]
2010:02:17-22:14:37 (none) slon_control[3668]: Killing slon pop3 [21817]
2010:02:17-22:14:45 (none) ha_daemon[3539]: id="38A3" severity="debug" sys="System" sub="ha" name="Netlink: Lost link beat on eth5!"
2010:02:17-22:14:47 (none) ha_daemon[3539]: id="38A3" severity="debug" sys="System" sub="ha" name="Netlink: Found link beat on eth5 again!"
2010:02:17-22:14:56 (none) ha_daemon[3539]: id="38A3" severity="debug" sys="System" sub="ha" name="Netlink: Lost link beat on eth5!"
2010:02:17-22:14:57 (none) slon_control[3668]: Slon reporting exited with value 0!
2010:02:17-22:14:57 (none) slon_control[3668]: Slon pop3 exited with value 0!
2010:02:17-22:14:59 (none) ha_daemon[3539]: id="38A3" severity="debug" sys="System" sub="ha" name="Netlink: Found link beat on eth5 again!"

What am I doing wrong?

This thread was automatically locked due to age.

Parents

0 BAlfson over 13 years ago

While there are a few cases like this, they don't seem to be widespread. If you're haveing a consistent problem, I suspect a hardware problem. Since you have to re-install for V8, you might want to change the "sync" ethernet adapters, choosing ones in more-common use.

This problem happened with my client when we moved them from "Cluster" to "Hot Standby" mode. To fix it, I disabled HA from the Master, waited for the Slave to do a factory reset and powered it down. I then disconnected all Ethernet cables to the slave, powered it up, up2dated manually, confirmed the MTUs, powered down, reconnected and powered the Slave on. Since then (about 18 months ago), there have been no issues.

CHeers - Bob

Sophos UTM Community Moderator
Sophos Certified Architect - UTM
Sophos Certified Engineer - XG
Gold Solution Partner since 2005

MediaSoft, Inc. USA
Cancel
Vote Up 0 Vote Down

Cancel

Reply

0 BAlfson over 13 years ago

While there are a few cases like this, they don't seem to be widespread. If you're haveing a consistent problem, I suspect a hardware problem. Since you have to re-install for V8, you might want to change the "sync" ethernet adapters, choosing ones in more-common use.

This problem happened with my client when we moved them from "Cluster" to "Hot Standby" mode. To fix it, I disabled HA from the Master, waited for the Slave to do a factory reset and powered it down. I then disconnected all Ethernet cables to the slave, powered it up, up2dated manually, confirmed the MTUs, powered down, reconnected and powered the Slave on. Since then (about 18 months ago), there have been no issues.

CHeers - Bob

Sophos UTM Community Moderator
Sophos Certified Architect - UTM
Sophos Certified Engineer - XG
Gold Solution Partner since 2005

MediaSoft, Inc. USA
Cancel
Vote Up 0 Vote Down

Cancel

Children

No Data