Announcement Date: 5 April, 2014Improvements:
[cluster] Cluster service avoid hardcoded vhid and deadratio parameters
[installation] Postinstallation script global.conf checking
Bug fixes:
[farmguardian] Farmguardian startup bug fixes
[farmguardian] Fix FarmGuardian execution for L4 farms
[farms] Datalink infinite loop getDevData() showing farm info
[farms] Manage HTTP/S farms connection bug due to wrong lb core child pid
[farms] L4 persistency multiprotocol command fix
[farms] Set priority value for L4 by default
[farmguardian] Farmguardian startup status fixed
[farms] Farms name hyphens filtering fixed
[farmguardian] FarmGuardian up status for the backends before stopping the FG service
[farms] L4 netstat connections
[farms] HTTP/S farm redirect fix when a redirect name is used
[logs] Regression: delete debug messages
[farms] Backend states on HTTP/HTTPS farms become inconsistent when the service doesn’t exist in the configuration file (the farm restart has not been performed). Service ID will be null
[farms] New backends are not included in the backend status file
[farms] Setting L4 TTL field is not applied by default. An automatic restart is needed
[farms] DateTime errors in the log files. The DateTime has been changed to system ‘date’ command
[farms] Force drain sessions automatically when a backend is detected as down
[farms] L4 farms connections accounting are not showing properly in the status panel
[config] Delete generated global.conf, it could be rewritten an already configured load balancer by the default one
[installation] Include interactive mode post-installation and error solved
[installation] Delete config files at installation time
[gui] Shell error solved getting ZLB version
[farmguardian] Farmguardian defunct processes and tcp/udp farm malfunction
[farms] TCP and UDP with pen LB, the backend never comes alive from FGdown. These kind of farms doesn’t need to detect the state change for a backend, as they need to refresh always the blacklisted timeout