diff --git a/ChangeLog b/ChangeLog index 1c939e8fc4a9f649d65a842953282366997fae97..a7114a1ec00ee3fa7f37d26fff24db6bff3afdd3 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +2007-10-18 Per Cederqvist <ceder@shafan.lysator.liu.se> + + Make backup-one-task more robust against hung servers. + * backup-one-task: Use BatchMode and ServerAliveInterval when + fetching the exclude file, to avoid hangs. + 2007-08-09 Per Cederqvist <ceder@ryttargardskyrkan.se> The "start immediately if the previous backup took more than 24 diff --git a/backup-one-task b/backup-one-task index d865571df21f5014f1eb839fa1aec42e84995d94..2233065336e5f07ddfd7849a71b79c1e63a7188f 100755 --- a/backup-one-task +++ b/backup-one-task @@ -158,6 +158,7 @@ then # Fetch an up-to-date exclude file. rm -f "$exclude" scp -B -q -i /root/.ssh/backupkey \ + -o BatchMode=yes -o ServerAliveInterval=120 \ "$server":"$serverpath"/.lysrdiff-exclude "$exclude" >/dev/null 2>&1 # Create an emtpy exclude file if non existed on the server. touch "$exclude"