Redcell Management Center - Release Notes - Alcatel

Redcell Management Center - Release Notes - Alcatel
Redcell Management Center – v.5.3
Release Notes
`
1.
Release Numbering
Dorado Software uses the following format for release/version numbers:
<ProductID>.<Major Version>.<Minor Release>.<Maintenance Release>.<Build #>
The ProductID for <this product> is "RMC" This release of the Redcell Management Center is formally
RMC.5.3 When contacting Dorado Customer Support, please refer to the complete release number, which
can be found in the Redcell Management Center “About” Box.
2.
Significant New Features
This is a significant new release of Redcell Management Center. It contains new interfaces and even more
revised functionality than appears in those interfaces. For a complete look at this functionaly, consult the
User Guide. Highlights of the new functionality in this release include the following:
•
Notifications and actions let you more completely configure Redcell responses to events
generated by internal or external processes.
•
Support for the optional Redcell Change Management add-on. Change Management
includes ProScan — a feature that proactively scans device configurations for
compliance with business policies.
•
Support for the optional Redcell Advanced Inventory Management & Reporting add-on.
•
To ensure that any application server node does not miss server-to-server heartbeats that
may erroneously initiate fail-over processes, you must connect clustered application
server nodes via a LAN with maximum latency of 1000ms. That said, you can put
clustered server nodes several miles/kilometers apart, as long as the connection does not
exceed the maximum latency (using Fibre ethernet, for example). WAN connectivity is
not recommended.
•
Usability Improvements in user interface.
•
Group Operations support for EAPS, EDP and VLAN.
•
Active config commands can now included variables (“tags”).
•
CPE WAN Management Protocol (CWMP) support (alpha) (CWMP Web Services and the
generic System Heartbeat service are now in DNC.)
3.
Feature Changes
•
Some additional Topology changes:
• You can invoke the alarm window on all of the devices except connection lines and
special indicators.
• Sometimes reordering or zooming makes the view window shift so that too many
objects are off the screen.
• Centering the object may not work as the user would expect. The user would
normally expect "center" to cause the object to be placed in the center of the view
screen but that does not really happen if the object is next to the viewable edge.
• Chassis view of a port shows just a single port. The color appears on this
representation of the port when it has an open alarm.
• The "cut-through" option erroneously appears for a port in the topology view. This is
similarly erroneous for group operations.
• "Shift-left-mouse" selects multiple objects instead of "ctrl-left-mouse."
•
The Domain Field in the client logon screen is not used by the some versions of the
application. The information provided during installation overrides DOMAIN. In such
installations, the file <installdir>owareapps\installprops\lib\installed.properties contains the follwing statement that overrides the domain field.
Even if users enter data in the domain field it is ignored.
OWARE.CONTEXT.SERVER.URL=jnp://[hostname or ip address]:1099
•
For systems running NTFS, you can extend the maximum size of your database file
beyond the default limit of 2GB. If you believe you may have a very large number of
database entries in a short period of time and/or do not want to adjust your DAP
policies or schedule. For example, this could be the case if you are running the
application on a network of 100 or more switches and are expecting many traps
from these switches in a short period of time, and do not wish to remove them from
the database through DAP. You can modify the %SystemRoot%\my.ini file in
Windows to allow for a larger database by editing the
data_file_path=d:/work/oware3rd/mysql/4_0_13/ibdata/ibdata1:600M:
autoextend:max:2000M line. Change the last parameter from 2000M (2GB) to a
larger number such as 4000M for 4GB. You must reboot after adjusting the ini file.
•
When RADIUS, TACACS or TACACS+ are enabled and the device is in multi-user mode,
the RADIUS/TACACS/TACACS+ user ID and password must match the multi-user user ID
and password on the device before Redcell can manage the device
•
Inventory Report Manager has been moved under the File menu item and removed from
the left tree view.
•
The first two Discovery screens have been simplified. See the User Guide for more
information.
•
The Report Manager is now available only from the menu (select Report Manager >
Inventory Reports from the File > Open > Network Services menu), not from the Navigation
Pane.
•
After Event Monitor receives a clear alarm (linkUp) trap, the applicatin updates the port
icon in the device's chassis view to reflect its cleared state (white background on port). To
reset the affected port back to its pre-alarm state you must click on the Refresh button in
chassis view. This removes the white trapUp/cleared state to return to the pre-trap/alarm
state (black). SCR 25912
4.
Known Issues
•
Equipment Configuration: Required Security Levels: Access to some system command
modes on devices may be defined by specifying an access privilege level for a user
account. For example, Riverstone devices require access privilege level 15 to access
Enable (privileged) mode. This application requires the user account for authenticating
with and managing a device has a privilege level of 15. For the Juniper Netscreen driver
this application requires root privileges for some configuration operations. Ensure that
user accounts associated with Authentication objects are configured on the device with
root privileges and has both read and write access. Similarly, Juniper M/T/J users require
users assigned to the superuser class for some operations. Consult your equipment’s
manuals for instructions about how to set up user access.
•
Performance: Redcell cannot respond faster than the network permits, but the
complexity of devices is also important. For example, it can take time to delete a device
that has 4000 interfaces. Such a device can take up to several minutes to delete,
depending on your installation’s hardware.
•
Manuals: The panel documented in Discovery Wizard, labelled Manage via, is not
enabled. This is slated for a future release.
•
Installation: Redcell cannot install on Solaris where only a Core System Support level of
installation exists. Solaris must have at least an End User System level installation before
Redcell can install.
•
Installation: If you cancel installation in progress, registry entries and files may remain
on the host where you are installing. Work around: Restart the installation/upgrade
process again then when it's done you can uninstall. Note: If you install Oracle, because
the installation is manual, this application does not create db_setup.log.
•
You cannot update IP addresses in the discovery parameters form. For example: select the
position between the 8 and the 0 in the address 192.168.0.1. You can delete the 0 but
cannot change the 0 to a 1. To change it, go the end of the string and backspace across
the 1 and 0 to type in 1.1 to get to the desired string 192.168.1.1. (SCR 9056)
•
Redcell allows the creation of duplicate links where link type and end points are the
same. (SCR 19536).
•
Discovered subcomponents appear even if the subcomponents check box is unchecked.
(SCR 20520)
•
Multiple NICs Support: Redcell does support the use of multiple NICs, but you must
have the actual IP address of the NIC in owareapps/installprops/lib/installed.props (not
localhost). If you have multiple NICs on different networks, multicast traffic sent out on
a network interface card (NIC) is bound to the network it is on. So if the appserver is
bound to 192.168.20.100 with the "-l" option, a client on 192.168.2.100 will never
connect unless the router that connects those networks routes multicast. (SCR 20053)
•
Infrequently, the Redcell login screen does not allow a cursor to appear. Resize the screen
to overcome this difficulty. (SCR 21018)
•
Redcell Topology: If you click Remove on an item in a view, it does not delete the item
from your system, it simply hides it in a Topology view. Some hidden items may still
appear. (SCR 20692)
•
When discovering new devices (managed objects or MOs), the device discovery engine
does not attempt to match IP addresses in LinkEndpoint connector objects with IP
addresses found. In the current implementation of the Link Discovery engine, the
"discovered entities" that do not support SNMP fail the the link data collection step
(observe warning message in the link discovery job containing a discovered entity) and
are removed from the link discovery process since they have no collected data. Here, the
IP address of the discovered entity is not compared to the LinkEndpoint address.
Workaround: Delete the link with the single LinkEndpoint referencing a managed
object and rediscover the link after both managed objects have already been discovered.
(SCR 20926)
Note: This is the same behavior in device resync.
•
You cannot delete a script command if any template uses it. The command script field
length is reduced from from 512 to 50 so that you can query this field on event template
table. (SCR 21014)
•
Cut-thru with SSH requires 1.a user name be configured on the device (this issue is
across all drivers) 2. An SSH (and not sshv2 ) management interface is configured for
the device. (SCR 21198)
•
In the tree view of the equipment manager (and other managers) if you right click on a
node without selection and execute a menu option it excutes against the wrong node
(the previous selection). Otherwise if you select first (left click) and then right click it
operates correctly. (SCR 23360)
•
Inventory Config (Settings > Configuration > Inventory Config) and the Add New
Authentication screen (in Discovery Wizard) do not have help screens. (SCR 22804, SCR
23281)
•
Intermittent Application Server. After the system runs properly, on a subsequent
restart, Application Server fails to properly initialize providing the following error in the
log file:
2005-03-01 10:52:32,102 INFO
[Di Mrz 1 10:52:32 C
[org.javagroups.FFIPartition]
ET 2005] [ERROR] JChannel.connect(): exception:
java.net.SocketException: IP_ADD
_MEMBERSHIP failed (out of hardware filters?)
2005-03-01 10:52:32,112 ERROR
[org.jboss.ha.framework.server.ClusterPartition] S
tarting failed
ChannelException: java.net.SocketException:
IP_ADD_MEMBERSHIP failed (out of har
dware filters?)
at org.jgroups.JChannel.connect(JChannel.java:224)
at
org.jboss.ha.framework.server.ClusterPartition.startService
(ClusterPa
rtition.java:291)
….. truncated
This problem only occurs when the Application Server setup autostarts on system
reboot. The issue stems from contention in gaining access to the sockets layer for IP
communications. If the NIC drivers have not fully initialized when application
server calls the JChannel.connect statement, there are no underlying resources to
allow communication. JChannel.connect starts communication between the
application and sockets programming layer for multicast applications. Dorado uses
multicast for communication between client and server.
Resolution: You must manually edit the configuration file for the JavaGroups
service. The configuration file is \oware\jboss-2.3.2\owareconf\clusterservice.xml. The XML configuration contains the following entry:
<UDP mcast_addr=”@MULTICAST_IP@” mcast_port=”45566”
ip_ttl=”64” ip_mcast=”true”
mcast_send_buf_size=”150000” mcast_recv_buf_size=”80000”
ucast_send_buf_size=”150000” ucast_recv_buf_size=”80000”
loopback=”false” />
Change the loopback setting to true. This change forces the JavaGroups API to bind
to the software loopback interface instead of the hardware interfaces and will resolve
the problem.
Alternative Workaround: If the problem still occurs, one workaround is to prevent
the Application Server from autostarting on system boot. On Windows, this means
removing the Application Server icon from the Windows startup group. After the
system is up and running, you can manually start the Application Server. It should
initialize with no errors.
•
Redcell supports only syslog messaging over UDP, not syslog messaging over TCP at this
time (SCR 23592)
•
When you change the SSH key, subsequent logins can fail if you select the Regenerate key
option and create a new key. Redcell caches the key value in a file:
<installdir>\oware3rd\cygwin\.ssh\known_host. You must delete this file
since the application looks there for the key. The application then generates a new file
with the new key. This applies to all device drivers using SSH. (SCR23802)
•
During upgrades loaddb could fail because of the following user account problems:
o
Database user Oware does not have the role OWRole assigned to it.
o
OWRole didn't have the required priviledges like create table, create index etc.
You can fix these problems by making the required changes to the Database User and
OWRole using oracle enterprise manager console or using SQL*Plus.
Another workaround is to run loaddb with the -d option. This forces the loaddb script to
recreate the database user.
•
Redcell date dialogs may appear with something other than the current date. (SCR
24336)
•
Default authentication timeouts for discovered equipment may be too short. SNMP and
Telnet may need as much as 15 seconds timeout instead of the default 5 seconds,
especially on busy networks. (SCR 24104)
•
In discovery, when looking at the Additional Authentication panel, the combo box at the
top of the screen specifies the management interface type to add to the equipment when
you select New. Based upon the selected type querying this screen (click Go) retrieves a
list of authentication objects appropriate for that management interface type.
Authentication credentails are classified based upon communication types (SNMP vs.
HTTP) along with credential requirements (SNMP v1/v2c is dififerent then SNMPv3
because the attribution is different. The protocol that communicates using those
credentials (the management interface type) is not associated to the credentials. For
example for CLI three protocols can use CLI-based credentials—Telnet, SSH, SSHv2.
When you select one of those protocols with the combo box the application retrieves the
same set of authentication objects from those available for all protocols. (SCR 23426)
•
V3 traps have the same format as V2 traps. To accurately reflect the trap’s version within
the protocol field, Redcell would need to duplicate every trap template to get the
template (it does not). Redcell does capture the SNMP trap version in the Version field,
which the user can add as a column to the view. (SCR 24032)
•
SNMP v3 Configuration
•
To receive and acknowledge SNMP v3 inform messages from a device, you must do the
following:
1.
Open the Redcell EMS in the equipment manager.
2.
Create and save a SNMP v3 Management Interface with the local IP address and the
engine ID (calculated below).
3.
Create an SNMP V3 User similar to the one on the device and save it.
The engine id for a Redcell System concatenates 00000063000100A1 with the local IP
address (in hex). For example, if the IP Addess is 192.168.0.154 then the engine ID is
00000063000100A1C0A8009A
These steps let the system receive inform messages from the device and acknowledge
them. (SCR 24338)
•
Redcell relies on the ARP table to provide infomation about ethernet link during
discovery. The ARP table is routinely cleaned up based upon activity. As a result some
link are visible only inconsistently. (SCR 23331)
•
Device discovery does not work for SNMP ports other than 161. (SCR 24354)
•
Trap Forwarding fails on windows XP service pack 2 fails with the error The following
error occurred in sendto:10004 This error occurs because of security changes in
winsock libraries preventing transmission of packets with source address not matching
any of the local interfaces. (SCR 24364)
•
The stopmedserver command does not work. Workaround: Press Ctrl+C while focus
is in the shell where the mediation server is running. (SCR 24411)
•
Editing deleted managed objects produces an obscure error message (“"Error occured
%1”). You cannot edit such deleted objects. (SCR 24584)
•
Opening a “physical” port can display an exception the appserver console: that begins
with “ERROR [redcell.common.util.RedCellRuntimeUtil]” (and continues). This error is
benign and can be ignored. (SCR 24457)
•
When you export an MO with Manage by Hostname checked, then re-import it, that
checkbox is unchecked after the import finishes. Workaround: Manually re-check the
Manage by Hostname checkbox. (SCR 24736)
•
Up and down buttons in discovery scheduler disappear when you clicks on Remove
button. (SCR 24835)
•
Development backup / restoration with dbbackup / dbrestore scripts may fail for larger
(850MB+) MySQL databases. Workaround: Use the MySQL native utilities for both
owbusdb and owmetadb databases. (SCR 24006)
•
A device login banner can not contain the same character(s) that are used during the
device login sequence. For example hahsmarks (#) greater than symbols (>) etc.. (SCR
24992)
•
After a successful delete, selecting an entry from a table does not enable the edit/delete
buttons. Ctrl+click the last row to unselecte and the reselect and the buttons enable.
(SCR 25127)
•
No audit record is created for add or delete of new scheduled item. (SCR 25775)
•
If you close a web client without logging out (by clicking the “x” in its upper-right
corner), then an immediate re-login fails. The following error message appears:: Error
Message: Web Licensing Error - (License needs to be updated allow more than 2 concurrent users
for -> PRODUCT: Oware EDITION: WEB)
•
When you create two different sessions on the server, and are licensed only a single
session, the application will not work. The session does not end for 60 minutes, the
default session timeout. Workaround: Close the web client with File > Exit. (SCR 23955).
•
When in edit mode, you can select multiple table entries when it is inappropriate. (SCR
23944)
•
Clicking on a port or turner refreshes the screen. To display the port edit screen, you
must select a port, and then click Open. (SCR 24222)
•
The cursor position onscreen may temporarily conceal field contents in, for example, the
Equipment Editor screen (SCR 24224)
•
Web client may display time-out messages if left open for more than an hour.
Workaround: close and re-open the client. (SCR 23889)
•
Web client URL must refer to a single server, even if the application servers are clustered.
Clustering still benefits server-side processes in such situations. If a single server fails,
however, you must enter a new URL containing another server name.
•
Occasionally, some edit buttons may not work correctly on Firefox browsers.
Workaround: Press button in its non-text area. (SCR 31452)
•
Create a manual link between two devices and save that link. Then delete that device
from inventory and the link created removes that end point, as expected. However, if
you re-discover the device and attempt to add an interface from that device to the link,
the link will never save that end point. Each time you add the end point, save, and reopen the link the end point is not there. (SCR 32554)
•
Redcell does not support interface identifier numbers that are very large. If the query
“Fetching children, please wait” freezes it may be on account of this limitation.
Workaround: Rename the interface with a telnet session. (SCR 32594)
•
This application can mistakenly apply Active configs to devices with similar sysobjectIDs.
For example, if you have sysObjectIDs of 1.3.6.1.4.1.3224.1.17, and 1.3.6.4.1.3224.1.171,
the .171 device can use the tagged commands that should be only available to the .17
device. (SCR 32829)
•
If you specify "no ping" for discovery then discovery assigns no ICMP authentication
object to the device. Resync continues to do ping verification and fails because the
device does not have an ICMP authentication. Work around: Remove the ICMP
management interface and retry resync. (SCR 32892)
•
Active Config archives do not always save. You can see this if you create new tagged
config with two tagged items, then add a new configuration archive, then provide values
for both tags, and hit Apply. Hit Save, then re-open Active config and configuration
archive and the second tag value is not present. Workaround: To get the second tag to
save you must select outside of the second tags cell you just edited before applying and
saving. You must click the enter key or tab key or another table cell for the application
to recognize the last cell that still had the cursor focus. (SCR 32961)
•
Occasionally, some edit buttons may not work correctly on Firefox browsers.
Workaround: Press button in its non-text area. (SCR 31452)
•
Create a manual link between two devices and save that link. Then delete that device
from inventory and the link created removes that end point, as expected. However, if
you re-discover the device and attempt to add an interface from that device to the link,
the link will never save that end point. Each time you add the end point, save, and reopen the link the end point is not there. (SCR 32554)
•
Redcell does not support interface identifier numbers that are very large. If the query
“Fetching children, please wait” freezes it may be on account of this limitation.
Workaround: Rename the interface with a telnet session. (SCR 32594)
•
To disable Mediation server on application server machines by ensuring the property
below appears in \owareapps\installprops\lib\installed.properties:
oware.appserver.mediation.setup=false
This step is done for performance reasons, since mediation takes precedence over
application server processing. Another machine must be set up as a mediation server if
you do this.
•
If a device needs only a password to log into an SSHv2 direct access / cut-through session
on a Solaris client, Redcell cannot connect, since having a user name and a password is
still a requirement for the cut through viewer. Workaround: If only SSHv2 fails, and
only on Solaris clients, then either use SSHv1 or Telnet or Windows clients. (SCR 30716)
•
Java Daylight Savings Time Patch: Sun has put out an advisory discussing time
problems associated with the daylight savings change and java based applications.
http://java.sun.com/developer/technicalArticles/Intl/USDST/ and
http://java.sun.com/javase/tzupdater_README.html
The United States has planned a change to its DST observance beginning in 2007. The
Energy Policy Act of 2005 mandates that DST will start on the second Sunday in March
and end on the first Sunday in November. In 2007, the start and stop dates will be March
11 and November 4, respectively. These dates are different from previous DST start and
stop dates. In 2006, the dates were the first Sunday in April (April 2, 2006) and the last
Sunday in October (October 29, 2006). Some countries are still evaluating whether they
will adopt the new rules for themselves. You should anticipate more changes in DST and
time zone rules for countries that typically align with U.S. DST rules.
Test Procedure: after applying patches on OS and our application, when the daytime
savings change happens, it is also changed on server side. To reproduce
1. Advance the system clock to 2007 daylight savings time (March.11, 2007)
2. Apply patch for your operating system (XP, Windows 2000)
3. apply DST patch for Java application
3.1 copy the folder of tzupdater2007a which is extracted from tzupdater1_0_1.zip
3.2 open cmd
3.3 change directory to Dorado\tzupdater2007a
3.4 run command of oware
3.5 then run command of java -jar tzupdater.jar -u
4. Run startappserver
5. Observe the time change in the logs. When the time changes on the OS, the
time on the server changes too. (SCR 33308)
Migration:
•
When upgrading the Redcell system to a new version, you must install all the same
components that were in the previous version.
•
If you cancel at a certain point during the installation, you must redo the installation
fully before you can delete it. (SCR 24299)
•
Shut down the application when you install an upgrade. The installer should migrate
any existing data automatically.
•
To upgrade some Redcell application and device driver features, you must install licenses.
Now, Redcell maintains a central depository (database) for license files. This means every
client and server machine used in installation no longer must register a license file
individually. Refer to the Installation Guide for details of how to install license files.
•
Upgrade Requirement for Oracle Users: The Oracle version must be 9.2.0.1 or later.
•
Oracle migration does not update several tables correclty. Workaround: Manually
update the affected tables. The workaround should be executed after the dbevolve step of
migration. To do it, execute the following SQL statments on the Oracle database:
ALTER TABLE OWARE_BASE_ARCHIVED ADD ( class_name
Varchar2(128), type Char(1), archived_count Number(12),
archivable Char(1) Default 'N' )
ALTER TABLE OWARE_BASE_ARCHIVED_LOG ADD archivable Char (1)
Default 'N'
SCR (27168)
•
Discovery and General Screen: Some fields in the Equipment Editor General screen must
be populated for resync to function correctly. These include Firmware Version, Software
Version, and SysObjectID. Also: If you are creating managed objects before you discover
them, by filling out the Equipment Editor screens, you must completely fill out the
General screen. (SCR 24672)
•
Discovery’s additional authentication step allows you to set the port, retries and timeout
to invalid values. On all three the range allowed is an integer (-2147483648 to
2147483647). The valid range is Port: 1 to 65535, Retries: 0 to 100, timeout: 0 to 100.
(SCR 24970)
•
When Windows (2000) restarts explorer.exe because of operating system errors,
application server continues to run but the tray displays no Process monitor (although it
too still runs). Workaround: If you kill pmtray.exe in the task manager, you can
restart server monitor by invoking pmtray.exe and the icon reappears on the task bar.
(SCR 25907)
•
You cannot edit a management VLANs for Extreme equipment in Equipment Editor.
Workaround: Use a cut-through to edit this on the command line. (SCR 26805)
•
By default, the Redcell cut through client uses the property oware.local.ip.address
to get the local interface for binding and communicating with application/mediation
server. When the client connects with the application/mediation server using VPN, the
cut through fails because the client is listening on the local interface while the
application/mediation server communicates on the VPN address. Workaround: Add the
property oware.ezmediation.cut-through.use.interface.name=<Interface
Description> to owinstalled.properties file on the Redcell client. This property
specifies the interface description for uniquely identifying the interface used for
communicating with the application/mediation server for cut through. The customer
can get the interface description name by running the command ipconfig on
windows and ifconfig -a on solaris/linux.
For example, some customers use the Nortel VPN application to connect to the
application server so for them the property will be
oware.ezmediation.cut-through.use.interface.name=Nortel. Only the first
word from in the interface display name is required if it uniquely identifies the interface.
(SCR 27099)
•
You cannot select a login name that contains spaces for the installing user. Installation
will fail. (SCR 29716)
•
When reseeding components that reseed reports (reports.ocp, among others), the current
5.2.3 to 5.3.0 migration path requires reseeding of all components and therefore requires
a workaround for this issue. Workaround: Prevent reports from seeding by renaming
rptcompmgr.properties before you reseed. Here are the steps:
1.
Rename:
changemgnt/lib/cmcompmgr.properties
reports/lib/rptcpmpmgr.properties
2.
Execute ocpistall -s (this seeds all components) as required for migration.
3.
Set the renamed files back to :
changemgnt/lib/cmcompmgr.properties
reports/lib/rptcpmpmgr.properties
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising