ExOS 6.3.13 Release Notes

Firmware Name: ExOS 6.3.13.2248
Release Date: August 8, 2013
Platforms: 2060, 2061, 4060, 4061, 5000, 6010, 6060, 6062, 7000, 8060, 8062, 10060, 10062, Virtual
Required Versions: 5.4, 5.5, 6.0, 6.1, 6.3.0 (or previous 6.3 BETA releases)
WARNINGS:

• After installing ExOS 6.3, it is not possible to roll back to any ExOS version prior to ExOS 6.0.2. It is strongly recommended to have ExOS 6.0.2 (or newer) installed on your system prior to upgrading to ExOS 6.3 .
• When rebooting into ExOS 6.3 following the installation, it is strongly recommended to use the "Slow Reboot" mode.

If you are upgrading to ExOS 6.3 from ExOS 6.0 or earlier:

• A database upgrade is performed and monitoring reports may be unavailable for up to 10 minutes.

When upgrading to ExOS 6.3 from ExOS 5.5 or earlier:

• WAN Memory persistence will be lost due to changes in the underlying storage architecture.
• ARM objects are automatically upgraded to APS objects, however, all historical ARM data is lost. If you would like to keep records of your historical ARM data then it is recommended to generate PDF reports with this data prior to upgrading.
• Virtual Machine disk images will be lost due to changes in the underlying storage architecture. This only affects customers running virtual machines on the Exinda appliance such as Exinda Mobile Client or Windows Server. Please contact Exinda TAC for a manual workaround.

If you are upgrading to ExOS 6.3 from ExOS 5.3 or earlier:

• This upgrade is not supported. Please upgrade to ExOS 5.4 or newer first.
• Network Management Systems that use the Exinda SNMP MIB will need to be updated to use the new Exinda SNMP MIB after upgrading to ExOS 6.3.
• After upgrading, refresh the Web UI browser window using Ctrl+F5 as CSS (style-sheets) and JavaScript functions have changed

Upgrading from V5 or V6 (appliances already on V5 or V6):

• On V5 or V6 systems with valid Software Subscription (SS), Hardware Maintenance (HM) or Premium Support (PS), navigate to System | Maintenance | Firmware Update and use one of the following URLs:

If you are running a release prior to 6.1 and your existing RAM usage, as seen on the "Monitor / System / Ram Usage" graph, is over 80%, then upgrading may cause your appliance to be pushed beyond its capabilities. Contract Exinda to inquire if a RAM upgrade kit is available for your platform before upgrading to 6.3. It is highly recommended that 6.3 only be installed on appliances with 4GB of RAM or higher.
Images:
Exinda 2061, 4060, 4061, 5000, 6010, 6060, 6062, 7000, 8060, 8062, 10060, 10062 (64-bit):

64 bit version Image Size: 319.2 MB MD5: bb0a152f3a74a6c84f8bac1e86102a97

Exinda 2060 (32-bit):

32 bit version Image Size: 279.7 MB MD5: bddc379d7bf820657fe3161441cf25a8

The image can also be downloaded to your desktop and uploaded to the appliance. Downloading and installation may take some time. Ensure that the partition other than your currently booted partition displays the latest ExOS version: "6.3.13.2248 (i386) 2013/08/06" or "6.3.13.2248 (x86_64) 2013/08/06".
You must reboot the appliance for the image to install. Once the system reboots ensure that the status bar displays the correct ExOS version (6.3.13.2248). If the old ExOS version is displayed it indicates a problem with either the installation of the new version (system will automatically fall back to existing version) or that the new version was not set to "to boot next" prior to rebooting.
Virtual appliances:
Refer to the virtual appliance documentation specific to the hypervisor being used for instructions on installing and upgrading the virtual appliance images. Please update to the latest version after downloading the initial image. This can be done by clicking on the version string in the WUI header to go to the proper screen.
VMWare ESXi (64-bit):

OVF Image Size: 10,752 bytes (10 KB)
VMDK (optional, for offline deployments) Image Size: (605.7MB)

XenServer (64-bit):

Zip File Image size: 624,108,947 bytes (595 MB)

Exinda AD Windows Client:
Customers using the Exinda AD client are advised to upgrade to the latest version of the client, 1.0.11, however, version 1.0.6, 1.0.7, 1.0.9, and 1.0.10 are still fully compatible with ExOS 6.3.

Fixed and Added in 6.3.13 (from 6.3.12):

• [B-01508] If downloading of a new firmware release fails, attempting to download the firmware again will restart from where the first failure occurred. This will help customers on a slow connection such as a satellite link.
• [B-01955] QoS: when using multi-queue mode, the number of supported policies has been risen from 200 to 1900.
• [B-02006] The system watchdog is now enabled for all systems after the upgrade. It is highly recommended that the system watchdog be left enabled.
• [D-01469] VLAN tag rewriting through policy rules does not rewrite the tag.
• [D-01481] AD: active directory groups that begin with a '#' character were not handled correctly. This has been fixed.
• [D-01514] monitoring: If detailed records is enabled for monitoring, the data that was being displayed on the monitoring Subnet page was not reflective of the detailed records that are being tracked. The information shown was a summary of the detailed records. This has been corrected so that if detailed records are enabled, these records are used to generate the report on this page.
• [D-01603] SDP: as of 6.3.12 and 6.4.0, the data sent to SDP is not consistent with previous versions. This change ensures that this version and subsequent versions send the same data to SDP as was sent in version 6.3.9 and prior.
• [D-01694] When the following conditions are met, file share users were denied access to the file share:
  • pre-6.3 compatibility mode for SMB acceleration is enabled
  • SMB acceleration is disabled
  • there are policies for the client and server combination that instruct the Exinda to accelerate the traffic
  • there is only a single Exinda appliance in the path between the client or Server
• [D-01728] DVCs: When the amount of traffic in a DVC far exceeds the configured bandwidth cap for the DVC, the resulting shaped traffic would be less than the configured bandwidth cap. This has been addressed and DVC bandwidth is now preserved when throttling a heavy volume of traffic.
• [D-01833] Under some circumstances the URL report in a subnet will show "no data available". This has been corrected.
• [B-02090] Updated Layer 7 signatures. For full information, please follow the following: List of Application Signatures in Exinda Releases

Fixed and Added in 6.3.12 (from 6.3.11):

• [B-01434] monitoring, cms: This is the first version that supports the CMS Technology Preview release
• [D-01251] diagnostics: Added additional debugging information to system processes
• [D-01420] monitoring: fixed performance issues that customers were having 6.3.10 and 6.3.11 where logging into the web UI and generating reports were taking a long time
  • Please Note that even after the upgrade, login and report generation may be slow if your system was slow before upgrading. The performance will improve on every 5 and 35 minutes after the hour as a clean up process runs. The system could take up to 2 days to completely return to top performance.
  • If you need the system to return to top performance, please call support for assistance after waiting for the first clean up process to run ( up to 30 minutes after the initial upgrade). Support have a mechanism to run the complete cleanup process by hand.

Fixed and Added in 6.3.11 (from 6.3.10):

• [D-01284] system: Clean up current day's files if directory size is > 128MB
• [D-01241] diagnostics: added support for capturing more information during a core dump (kdump)
• [APP-8188] classification: fixed an issue where the wrong URL domain would be displayed in the real time monitor
• [APP-8186] Adaptive Response: AR rules with a Monthly duration that also had an External Exception did not work.
• [APP-8093] VoIP: The MOS scoring has been improved to be more accurate
• [APP-8088] NCP: Fixed a communication issue with NCP traffic
• [D-01232] monitoring: User-drill-down by-host report is broken when using legacy detailed reporting enabled
• [D-01200] SMBAD: fixed an issue with meta data cache population differences with NetApp filers

Fixed and Added in 6.3.10 (from 6.3.9):
The monitoring database summary data has been extended in this release to contain all the data from the detailed records. This enables full drill-down even when detailed records are disabled. (Via the 'Setup -> Monitoring -> Detailed Records' option.)

The detailed records are only required to allow earlier firmware versions to be display correct data and for some historical data. New appliances will have detailed records disabled by default as they are no longer required.

• [APP-7766] monitoring: Increase the available information when detailed records are disabled
• [APP-7766] monitoring: Disable detailed records for new systems since they are only required for access to historical data
• [APP-8148] monitoring: Account for Ethernet header (source and destination MAC and VLAN tags) and FCS in byte counters
• [APP-8163] acceleration smb: Correctly handle negotiation with a server with less capabilities than the client
• [APP-7531] wui: 'Monitor -> Control' graphs sometimes show the wrong value for 'Maximum BW' in the table.
• [APP-8177] system: Remember when default application groups are modified so they are not reset back to defaults on upgrade
• [APP-8160] system: Don't log excessively when one of the temperature sensors is busy and cannot be accessed
• [APP-8151] system: Don't crash when accelerating and receiving some specific forms of malicious TCP fragments
• [APP-8097, APP-8117, APP-8118, APP-8119] Reduce lock contention and optimise policy matching for increased performance and reduced CPU usage.

Fixed and Added in 6.3.9 (from 6.3.8):

• [APP-7638] wui: voip - when multiple class had the same R value only the last call was stored
• [APP-7638] wui: voip - the packet count did not packets from up to 10s at the end of the call
• [APP-8100] wui: 'Dashboard -> Benefits -> Link Utilization' peak information did not always align with the actual peaks
• [APP-8115] wui: Revert the subnet report changes added for 6.3.8 to prevent internal error in some situations
• [APP-8096] system: Start irq balance earlier to ensure high performance on boot on multi core system with heavy traffic already running at boot time
• [APP-8099] system: Optimise application matching performance when assigning policies to packets

Fixed and Added in 6.3.8 (from 6.3.7):

• [APP-7878] wui: The flash version of the voip graphs was not showing all the data
• [APP-8083] wui: Subnet report does not show correct data when sorted by name
• [APP-1787] wmd: Handle hostid changes better to allow easier integration of replaced appliances
• [APP-7582] system: Don't allow SDP to create an application with id 0
• [APP-7938] wui: Real-time reduction shows no information when there were are more entries than the configured table size (also in 6.1.10)
• [APP-7922] wui: Modifying an SNMP community would add a new community - modified to handle multiple SNMP communities (also in 6.1.10)
• [APP-7746] wui: Fix wui to work when optimizer circuit names contain a hash (#)
• [APP-7875] wui: Add paging for large user and group lists
• [APP-7841] acceleration smb: Fix issues with printing due to incorrect large files handling
• [APP-8076] acceleration sslad: Fix rare issue with sslad on multicore cpus sometimes blocking a subset of source ips
• [APP-7439] system: Add support for 115200 baud serial console speed
• [APP-8072] system: Include the hostid in body of emails
• [APP-8052] system: Enable the system watchdog by default for new installs
• [APP-7852] system: Fix applications with an invalid id of 0 to have a proper id during upgrades
• [APP-8074] optimizer: Fix for occasional 30s stall during bootup
• [APP-7877] monitoring: Updated L7 signatures. For full information, please follow the following: List of Application Signatures in Exinda Releases?

All fixes for 6.3.0 - 6.3.7 and 6.1.6 and 6.1.7 are also included

Known Issues:

• [APP-7426] pre-population: NTLM authentication for HTTP is not supported - NTLMv1 added in 6.3.1. NTLMv2 not supported.
• [APP-7066] system: management access (ssh, telnet, http) connections reset when the management IP is on a VLAN on a bridge - fixed in 6.4.0
• [APP-3410] system: May display Unsaved Changes after Factory Default or Firmware Upgrade. - fixed in 6.3.1 and later
• [APP-3275] monitoring: Graphs/tables show data for "last 60 minutes" show "this hour" in the drill-down reports.
• [APP-1999] optimizer: Cannot add a VC in a circuit if another VC exists with the same name in another circuit. - this will not be changed.
• [APP-668] cli: command completion does not work for names with multiple words (that contain a space). e.g. show policy my policy.