The configuration at the moment of a failure is
very often useful to better understand the erros
in the logs and debug the underlying issue.

Signed-off-by: Klaus Aehlig <[email protected]>
---
 roles/buildmaster/files/srv/buildbot/lib/scripts/collect-logs | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/roles/buildmaster/files/srv/buildbot/lib/scripts/collect-logs 
b/roles/buildmaster/files/srv/buildbot/lib/scripts/collect-logs
index 591fd0b..331cded 100755
--- a/roles/buildmaster/files/srv/buildbot/lib/scripts/collect-logs
+++ b/roles/buildmaster/files/srv/buildbot/lib/scripts/collect-logs
@@ -17,6 +17,11 @@ for host; do
   rsync -av \
     root@$host:/var/log/ganeti/ $hostdir/var/log/ganeti/
 
+  # Collect the configuration information
+  mkdir -p $hostdir/var/lib/ganeti
+  rsync -av \
+    root@$host:/var/lib/ganeti/ $hostdir/var/lib/ganeti/
+
   # Also collect vcluster logs, if present
   if ssh root@$host test -d /srv/ganeti/vcluster; then
       mkdir -p $hostdir/srv/ganeti/vcluster
-- 
2.6.0.rc2.230.g3dd15c0

Reply via email to