Filter
Top Products
Node recovery
Page 30 | How to Configure and Use AMF on Allied Telesis Switches
Node recovery
Automatic node recovery
AMF has been designed so that when a node fails it can be replaced with an unconfigured
device of the same type, and AMF will automatically upgrade and configure the new device
from the most recent backup. Often the replacement device will be a factory default, brand
new “out of the box” device, but it may be that you want to replace the failed unit with one
that has been previously used elsewhere. In this instance it is necessary to return the
replacement device to a “clean” state so that AMF can recognize it as a suitable replacement,
and begin automatic recovery. (See section "A “Clean” node" on page 31)
When a failed node is replaced with an unconfigured device, AMF immediately disables
forwarding on the device, shuts down all non-AMF ports, and applies the AMF safe
configuration. (See section "AMF safe configuration" on page 34.) AMF then checks
whether any of the AMF master nodes has a valid backup for the replacement node, and if it
finds one it begins to attempt automatic node recovery. Once automatic node recovery has
completed, it will then reboot the replacement node which will then rejoin the AMF network
with identical files and configuration, to the failed node it replaced.
Here is some example console output showing automatic node recovery:
Warning: Nochangesshouldbemadetothedevice'sconfigurationwhileanoderecoveryis
underway.Alogmessagewillappearontheconsoleorotherloggedin session indicating
when recovery has finished (whether successfully or with errors). This messagecanalsobefound
byviewingthe logwiththeshow logcommand
.
23:03:15 awplus ATMF[863]: ATMF network detected
23:03:15 awplus ATMF[863]: ATMF safe config applied (forwarding
disabled)
23:03:25 awplus ATMF[863]: Shutting down all non ATMF ports
23:03:26 x510_1 ATMF[863]: Automatic node recovery started
23:03:26 x510_1 ATMF[863]: Attempting to recover as x510_1
23:03:26 x510_1 ATMF[863]: Checking master node availability
23:03:32 x510_1 ATMF[863]: Master has joined. 2 members in total.
23:03:32 x510_1 ATMF[863]: x908_VCS_2 has joined. 3 members in total.
23:03:32 x510_1 ATMF[863]: x908_VCS_1 has joined. 4 members in total.
23:03:37 x510_1 ATMFFSR[2950]: Retrieving recovery data from master
node Master
23:05:18 x510_1 ATMFFSR[2950]: File recovery from master node
succeeded. Node will now reboot
Flushing file system buffers...
Unmounting any remaining filesystems...
Restarting system.