jetNEXUS ALB-X Accelerating Load Balancer Software Version 4.2.0 (Build 1722) 1st September 2016 ================================================================================================= New =================================================== - Integrated App Store to purchase internal and external applications such as Web Application Firewall, flightPATHS and many more... - Pre-Authentication/Single Sign On - Microsoft RDP based session persistence - Fully functional RESTful API control - Multiple health-check monitors per service - Native SSL support - increased efficiency (lower CPU) and lower number of overall connections - Add response time based load balancing - fastest responding server gets new connections - Manual Weighted Load Balancing for Least-Connections, Round-Robin, Fastest - Cloud based software update - Support Button - This will allow a customer to select a timeframe and click one button and it will zip up all the relevant files that support will need to troubleshoot - Dynamically detect and convert slash notation to dotted decimal for all relevant places - e.g. /24 to 255.255.255.0 - Added a friendly prompt to "restart services" after a Sheebang that needs it - the restart can then be completed when ready - Graphing of real server response time - Real server DNS health-check monitor - Real Server SNMP OID health-check monitor - Real Server HTTPS health-check monitor - Added cookie parameter "httponly" on a per service basis - Added cookie parameter "secure" on a per service basis - Added the ability to record the update script to assist support - Now able to add and delete a custom monitor from the GUI - Ability to tune the failover settings from the GUI - broadcasts per second and packet loss toleration - automatically calculate failover time - New licensable features - Pre-Authentication, GSLB, Web Application Firewall, etc. - Software update - new Salesforce based latest download, - Software update stores the 10 most-recent - Software restore - software version now stored on appliance - Test button added for remote log storage - flightPATH - ability to see which rule is applied and where - Ability to use domain names for real servers up to 120 characters - Service type HTTP now supports WebSockets - Added IPv6 functionality - Support for SNI certificate selection and in the service's health-check monitors - Security log shown in users page Improvement ===================================================== - Saving and restoring config will now restore bonding too - Failover improved for Cisco Switches by adding ARP Replies upon failover - GUI automatically redirects from http:// to https:// so typing in the IP gets connected - Conserve CPU usage for non-keep-alive requests - Greatly improved GUI responsiveness and scalability - VMware OVA and upgrade disk size increased to 25 gigabytes - Capture filters data for IP addresses - Gateway connectivity substitutes for transparency connectivity for ease of use - WebSocket protocol compatibility added - SSL Session resumption control - SSL renegotiation control - Troubleshooting page provides diagnostics and bundled-up support information - Default monitor timeout is now 10 seconds - Virtual service foldback into another service adjusts health-checking to allow the VIP to always start - Default capture size raised from 2M byte to 10M byte - Remopte desktop health-check - Library drop-down lists are alphabetical - Auto-open Servers tab when adding a service - E-Mail alerts are improved to include server and service names and multiple e-mail addresses are supported - Health-check monitors use | to allow multiple values to be checked for - History page adds total data in/out - W3C logging for Layer-4 data - Statistics are recorded for UDP data - Column titles maintained in scrolled status page - Search filter in flightPATH page - Delete confirmations added - SNMP up-time does not reset at midnight UTC - Warnings added for changes to adapter details that will cause loss of service - Date and time changes are shared across the cluster - DICOM VS takes better default timeout and health-check monitor period - Custom WMI CPU health-check monitor - Real Server Monitoring now support SNI with ability to add HTTPS GET Resolved ===================================================== - Improved diagnostics management after sheebang updates - NIC naming in ISO install was sometimes wrong - Watchdog restart timers removed; watchdog more vigilant; corrected fail-over reliability immediately after reboot - Caching provided reliably with older pre 4.0 configurations when restored - Check the interface is up before moving web console - Remove bad file number problems - Stop high CPU problems with large numbers of connections - Addressed GUI path security vulnerability - Stopped problems with fast use of the GUI - High numbers of flightPATH rules on a service allow correct editing - Character \ replicated across the cluster in flightPATH - Control timing of consecutive GUI updates for stability - flightPATH rewrite stability when mixed with other flightPATH rules - UDP Monitor renamed as ICMP unreachable monitor for clarity - Stop high CPU after a restart with some large SSL configurations - Derelease TCP Half Monitoring - HTTP Cache is more efficient with pre-expired content - Certificate replication in clusters - Removed arcane hard timer and soft timer - Health-check monitor uses User-Agent v4 for HTTP requests - Dashboard is a normal page, not a special default Known Issues ========================================================== - Changing adapter can result in Cluster listener on 27377 needing system power restart for both cluster boxes to update the change of network - Bad calculated weight in Layer 4 TCP/UDP