Clone
Michael Blow
committed
on 14 Jun 17
Re-register NC with CC on reestablished IPCHandle
In case of failed NC -> CC IPCHandle due to CC crash/restart, the NC
needs to re-register … Show more
Re-register NC with CC on reestablished IPCHandle

In case of failed NC -> CC IPCHandle due to CC crash/restart, the NC

needs to re-register with the CC in order to rejoin the cluster, as the

CC ignore heartbeats from unregistered nodes.

- Improve toString on IPCHandle

- Add tests for killing & restarting CC / NCs to NCServiceExecutionIT

- Retrigger NCService on detected dead node

- Ensure jobIds are not reused on CC restart

- NCService shouldn't truncate NC log

Change-Id: I6f93ca9ab37e56e02bafcdecd1e2d0cf664faef6

Reviewed-on: https://asterix-gerrit.ics.uci.edu/1830

Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>

BAD: Jenkins <jenkins@fulliautomatix.ics.uci.edu>

Integration-Tests: Jenkins <jenkins@fulliautomatix.ics.uci.edu>

Reviewed-by: Till Westmann <tillw@apache.org>

Show less

master + 4 more