2018 Mercedes-Benz GLC 300 SUV
4 stars based on
However, there are some important differences between these various mechanisms that are not well understood. This document attempts to shed some light on this matter and provides guidelines to use to troubleshoot as well. Cisco Media Gateway Controller Software. However, the mechanisms discussed commonly exist in other software releases. The information in this document was created from the devices in a specific lab environment.
All of the devices used in this document started with a cleared default configuration. If your network is live, make sure that you understand the potential impact of any command.
For more information on document conventions, refer to the Cisco Technical Tips Conventions. Provisioning data are the data files that result from MGC configuration commands that are issued during an MML provisioning session. These data files carry a. The variable provsession refers to the session name chosen in MML when the provisioning session was started.
In a redundant MGC configuration with an 57 reference manual 16164 binary logging options and variables and a standby MGC, it is important to keep provisioning data synchronized between both hosts to prevent loss of service during a failover condition. There are several ways of doing this, and each method 57 reference manual 16164 binary logging options and variables applicable to a particular situation.
They share these similarities:. File attributes name, size, and date of last modification of data files on both hosts are compared. Those files with nonmatching attributes are transferred to the standby MGC. Dynamic reconfiguration is triggered on the standby host to reflect the newly synchronized active configuration.
It pushes those directories and files that are different or do not already exist to the standby MGC. To do this, set the parameter diskmonitor. The default value is Otherwise, functionality and system performance may be adversely affected. If you delete a configuration, you should delete it on both systems, if possible.
This command operates on only one configuration: This leaves unaltered the actual flow of provisioning data from active to standby. If you have a standalone MGC, set this value to false. If you have two MGC hosts 57 reference manual 16164 binary logging options and variables a failover configuration, set this value to true on both hosts to accommodate failover conditions in which the current active host can become the standby host.
When the initial MGC configuration on the active host is deployed, you must change the pom. After setting this parameter to trueyou can start the MGC on 57 reference manual 16164 binary logging options and variables standby host. As the MGC comes up, the data on the standby host is synchronized with the data on the active host.
The standby host then goes into the standby state. During the upgrade of a redundant MGC configuration, verify that the pom. Also, do not forget to change the setting back to true once the peer host has been upgraded successfully as well.
If you set pom. When provisioning synchronization fails, the failure reported by prov-sync or prov-dply in MML is usually not very informative. Provisioning synchronization compares file attributes name, size, and time of last modification and transfers the files for which the attributes do not match. However, the files must be accessible first.
If POM has trouble processing certain files or directories, it should be apparent from relevant error messages in the platform. File access is determined by file ownership and file permission modes. Furthermore, permission modes for data files should allow read and write access to mgcusr and mgcgrp. Directories containing these data files should include e X ecute permission for mgcusr and mgcgrp as well.
The most significant file permissions are those for the user and group mcgusr and mgcgrp. When you verify file ownership and permissions for data files of the active configuration, remember to verify both locations where these data files are kept: In a VSC configuration, remember to verify dial plan data files ending with.
If file permissions do not appear to be correct, they may have been manually changed by accident, or the umask value for mgcusr may no longer be correct. The umask value determines the file permissions creation mask for the user shell execution environment. Usually, you should not change file permissions and file ownership manually. If necessary, however, this shows how to restore values back to normal:.
This directory contains the minimum data files required to configure 57 reference manual 16164 binary logging options and variables MGC. These are the data files that an initial installation of the MGC application uses. Users that wish to provision the MGC from scratch can use this directory as the source.
They are discussed below. For large configurations or during periods of high network usage, the prov-dply and prov-sync commands may time out and should be reexecuted.
Preferably, this is done when network usage has decreased, or, in the case of a prov-sync timeout, after unwanted configurations have been removed with the config-lib utility.
If you need to make repeated attempts to complete the synchronization successfully, you may also be running into the Network Problems described below. Layer 1 network problems can prevent file transfers from completing successfully. The UNIX command netstat -i allows you to verify if any input or output errors or any collisions have occurred on the Ethernet connections. Typically, you should compare its output taken before the provisioning synchronization, and again after the provisioning synchronization.
A common problem is mismatched Ethernet duplex settings between network 57 reference manual 16164 binary logging options and variables and switches. You should not have any collisions when working in full duplex; that is typically the way you want to work. If your equipment is connected to a Cisco Catalyst switch, you can use the show port and set port commands to view and change port settings.
The variable is the interface number. To examine parameter values, use the ndd -get command. The setting corresponding to the desired mode should be set to 1 and all other parameters 57 reference manual 16164 binary logging options and variables be set to 0.
This can be useful if one of the interfaces is connected to an older switch that is not autonegotiating the line speed or mode correctly. To list all parameters supported by the network card driver, use:. These ports are default values, controlled by the pom. If it fails, you should verify whether prov-dply works instead because prov-dply only operates on the active configuration. Verify that the diskmonitor. Be sure you delete them on both systems, if they are available on both.
There are different reasons why the standby host remains OOS. Complete these steps to determine if the problem is with provisioning synchronization:.
If it does, provisioning synchronization is likely to be the problem. You should proceed to check the logs and verify file ownership and permissions for the active configuration, as explained in the section File Ownership and Permissions.
If the issue happened after you upgraded the MGC software or patch level, it is possible that after the upgrade of the standby, but before the upgrade of the active host, the standby was booted with pom. This causes POM to retrieve the old data files from the active host. It means that the upgraded software on the standby host tries to run with old data files.
The format of the old data files may or may not be compatible with the new data files, which depends on the compatibility level between both versions of software. This would not present a problem if you performed prov-sync before the upgrade and set pom.
The safest way to recover from this situation is to revert to the previous release or patch level, and follow the correct procedure in a second attempt. For issues with provisioning synchronization, the process of interest is POM. You can debug POM 57 reference manual 16164 binary logging options and variables the active host, the standby host, or on both. However, it is typically best to debug such issues on the host that will be standby.
Follow this procedure to debug issues with prov-sync and prov-dply on the standby host:. In MML, issue set-log: After the 57 reference manual 16164 binary logging options and variables has completed, issue the command set-log: Add the parameter pom. To verify this, issue rtrv-ne in MML. If the problem is that the standby remains OOS, allow the debug to run long enough to capture the problem.
Remove the line pom. The procedure to debug issues with pom. The only differences are that you execute all steps on the host that will fail over the current active host and that, instead of using the prov-sync or prov-dply commands, you should issue the command or action that will trigger the failover. Be careful when you debug such an issue. It is possible that the configuration of both MGCs is no longer in synchronization, and the setup may fail over to an undesired configuration.
The POM debug can be quite verbose. Enable it just before you issue the command that triggers the synchronization, and disable it immediately after command completion. It is best not to enable the debug during high system load.
There are many other messages collected, but these two are the most interesting and informative. To see the dynamic reconfiguration as a result of synchronization, you can also search for XETable and cfgEventHandler.
To identify problems with file transfers, search for lines containing tftp and operationStatus.