ebase

Maintenance Guide

Version: V3.0

ZTE CORPORATION
NO. 55, Hi-tech Road South, ShenZhen, [Link]
Postcode: 518057
Tel: +86-755-26771900
Fax: +86-755-26770801
URL: [Link]
E-mail: support@[Link]
LEGAL INFORMATION
Copyright © 2011 ZTE CORPORATION.
The contents of this document are protected by copyright laws and international treaties. Any reproduction or
distribution of this document or any portion of this document, in any form by any means, without the prior written
consent of ZTE CORPORATION is prohibited. Additionally, the contents of this document are protected by
contractual confidentiality obligations.
All company, brand and product names are trade or service marks, or registered trade or service marks, of ZTE
CORPORATION or of their respective owners.
This document is provided “as is”, and all express, implied, or statutory warranties, representations or conditions
are disclaimed, including without limitation any implied warranty of merchantability, fitness for a particular purpose,
title or non-infringement. ZTE CORPORATION and its licensors shall not be liable for damages resulting from the
use of or reliance on the information contained herein.
ZTE CORPORATION or its licensors may have current or pending intellectual property rights or applications
covering the subject matter of this document. Except as expressly provided in any written license between ZTE
CORPORATION and its licensee, the user of this document shall not acquire any license to the subject matter
herein.
ZTE CORPORATION reserves the right to upgrade or make technical change to this product without further notice.
Users may visit ZTE technical support website [Link] to inquire related information.
The ultimate right to interpret this product resides in ZTE CORPORATION.

Revision History

Revision No. Revision Date Revision Reason

R1.0 2012-11-28 First edition

Serial Number: SJ-20120313112131-024

Publishing Date: 2012-11-28 (R1.0)

Contents
About This Manual ......................................................................................... I
Chapter 1 ebase Introduction.................................................................... 1-1
1.1 Functions........................................................................................................... 1-1
1.2 Application Scenario........................................................................................... 1-3
1.3 System Architecture ........................................................................................... 1-4
1.4 Operation Environment ....................................................................................... 1-7

Chapter 2 ebase Routine Maintenance..................................................... 2-1

2.1 Maintenance Tools ............................................................................................. 2-1
2.1.1 CLI Client ................................................................................................ 2-1
2.1.2 OMM ....................................................................................................... 2-2
2.1.3 Preventive Routine Maintenance Tool ........................................................ 2-3
2.1.4 Full Backup Tools ..................................................................................... 2-3
2.1.5 load out/in Tool......................................................................................... 2-5
2.2 Maintenance Contents........................................................................................ 2-6
2.2.1 Checking the Operation Status.................................................................. 2-7
2.2.2 Checking for a Core File ........................................................................... 2-9
2.2.3 Checking the Number of Links .................................................................. 2-9
2.2.4 Checking the Scheduled Backup Task ..................................................... 2-10
2.2.5 Checking the Synchronization Module ......................................................2-11
2.2.6 Checking the Related Interface Program.................................................. 2-13
2.2.7 Checking Table Definitions ...................................................................... 2-14
2.2.8 Checking Configuration Files................................................................... 2-15
2.2.9 Checking the Dual-Server Cluster ........................................................... 2-17
2.2.10 Checking the Version Number and Bit Number ....................................... 2-18

Chapter 3 ebase Backup and Recovery ................................................... 3-1

3.1 Backup Period ................................................................................................... 3-1
3.2 Physical Backup and Recovery ........................................................................... 3-2
3.2.1 Physical Backup....................................................................................... 3-2
3.2.2 Physical Recovery.................................................................................... 3-3
3.2.3 Verifying a Physical Recovery ................................................................... 3-4
3.3 Logical Backup and Recovery ............................................................................. 3-5
3.3.1 Making a Table-Level Backup.................................................................... 3-5
3.3.2 Making a Table-Level Recovery................................................................. 3-6

I
 3.3.3 Making a Database-Level Backup ............................................................. 3-7
3.3.4 Making a Database-Level Recovery .......................................................... 3-7
3.3.5 Making a System-Level Backup ................................................................ 3-8
3.3.6 Making a System-Level Recovery ............................................................. 3-9
3.3.7 Verifying a Logical Recovery ..................................................................... 3-9

Chapter 4 Common Problems ................................................................... 4-1

4.1 Common ebase Operation Problems ................................................................... 4-1
4.1.1 Determining the Cause of the Stopped ebase Process ............................... 4-1
4.1.2 Troubleshooting the “Extrabase: create listen socket err, exit” Error............. 4-1
4.1.3 Troubleshooting the “can't find xbase_trace.log” Error................................. 4-2
4.1.4 Copying Data Between Two ebase Servers................................................ 4-3
4.1.5 Modifying the Shared Memory................................................................... 4-3
4.1.6 Troubleshooting the ebase's Failed Startup due to a Configuration
Modification ............................................................................................ 4-4
4.1.7 Troubleshooting the ebase's Failed Startup due to File Permission
Issues .................................................................................................... 4-5
4.1.8 Locking the ebase's Memory to the Physical Memory ................................. 4-5
4.2 Common Client Problems ................................................................................... 4-6
4.2.1 Running the Tracing Command ................................................................. 4-6
4.2.2 Troubleshooting an Error Reported at a Login through the Client ................. 4-7
4.2.3 Troubleshooting Failed Data Insertion in the Database ............................... 4-7
4.2.4 gen_clust_index Description ..................................................................... 4-8
4.3 Synchronization Module Problems....................................................................... 4-8
4.3.1 Troubleshooting a Synchronization Error.................................................... 4-8
4.3.2 Troubleshooting a Synchronization Failure due to the Disabled Archive
Mode ..................................................................................................... 4-9
4.3.3 Troubleshooting a Synchronization Failure................................................4-11
4.3.4 Troubleshooting Message Accumulation for a Synchronization Module ...... 4-13
4.4 Common xbasemoni Faults............................................................................... 4-14
4.4.1 Fault One .............................................................................................. 4-14
4.4.2 Fault Two............................................................................................... 4-15
4.5 OMM Fault....................................................................................................... 4-15
4.6 iTool Faults ...................................................................................................... 4-16
4.6.1 Maintenance Timeout ............................................................................. 4-16
4.6.2 User Description .................................................................................... 4-17
4.6.3 Troubleshooting the Failed Selection of an ebase Server When Adding a
New Device .......................................................................................... 4-17
4.6.4 Troubleshooting a Failed Import of a Maintenance Package...................... 4-18

II
Appendix A Frequently Used Commands............................................... A-1
Appendix B iTool Description .................................................................. B-1
B.1 Automatic Preventive Maintenance Operations .................................................... B-1
B.2 Maintenance Items............................................................................................. B-8

Glossary .......................................................................................................... I

III
IV
About This Manual
Purpose
This manual describes the functions, system architecture, application scenarios,
maintenance tools, maintenance operations, backup and recovery, and common
troubleshooting procedures of the ebase.

Intended Audience
This manual is intended for:

l Network management engineers

l Maintenance engineers

Prerequisite Skills and Knowledge

Familiarity with the following is helpful:
l UNIX/Linux commands and procedures for handling simple UNIX/Linux system faults
l Oracle database commands and procedures for handling simple Oracle database
faults
l Sybase database commands and procedures for handling simple Sybase database
faults

What Is in This Manual

This manual contains the following chapters:

Chapter Summary

1, ebase Introduction Describes the functions, system architecture, and application scenarios
of the ebase.

2, ebase Routine Describes the common maintenance tools for the ebase and the
Maintenance maintenance contents.

3, ebase Backup and Describes the backup and recovery operations for the ebase and recovery
Recovery verification operations.

4, Common Problems Describes the common problems with ebase components and the
recommended troubleshooting procedures.

A, Frequently Used Describes the commands frequently used during the ebase routine
Commands maintenance.

B, iTool Description Describes the maintenance operations using the iTool and the maintenance
items.

I
Conventions
This manual uses the following typographical conventions:

Typeface Meaning

Italics Variables in commands. It may also refer to other related manuals and documents.

Bold Menus, menu options, function names, input fields, option button names, check boxes,
drop-down lists, dialog box names, window names, parameters, and commands.

Constant Text that you type, program codes, filenames, directory names, and function names.
width

| Separates individual parameter in series of parameters.

Caution: indicates a potentially hazardous situation. Failure to comply can result in

moderate injury, equipment damage, or interruption of minor services.

Note: provides additional information about a certain topic.

xsql> Command prompt on the xsql client.

Command prompt in an operating system. The # is for the root user, and the $ is for
#/$
other users.

II
 Chapter 1
ebase Introduction
The ebase, short for EXTRABASE, is a high-performance in-memory database
management system. The ebase is a relational database management system (RDBMS)
that supports the standard SQL syntax and has a similar syntax style to Sybase.
The ebase operates and stores data based completely on a physical memory. Consisting
of a main server program and a number of peripheral programs, the ebase functions as a
repository for caching data and cannot be replaced with a commercial database.
Table of Contents
Functions ...................................................................................................................1-1
Application Scenario...................................................................................................1-3
System Architecture ...................................................................................................1-4
Operation Environment...............................................................................................1-7

1.1 Functions
The ebase provides a mechanism to accurately control memory usage. The ebase
also supports automatic installation, upgrade, preventive maintenance and GUI-based
configuration in accordance with iTool standards, which facilitates the use and
maintenance.
The functions of the ebase are as follows:
l Support the standard SQL syntax.

Note:
There is no system table in the ebase. Users can run the show tables and show proc
edures commands to view the definitions of tables and stored procedures in the data
dictionary. For the SQL commands and syntaxes that the ebase supports, refer to
Appendix A.

The ebase supports the creation and management of database objects (tables,
indexes, and stored procedures) using DDL commands. The ebase supports
DQL and DML commands including SELECT, INSERT, UPDATE, DELETE, and
TRUNCATE TABLE that are compliant with SQL-92. The ebase also supports data
import and export using SQL commands. In addition, the ebase provides built-in

1-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

functions, for example, aggregate functions, which can be used in SQL commands
and stored procedures.
l Support the data index function.

Note:
Up to four indexes are allowed for each table. The performance of the inserted data
is degraded because each index occupies extra memory space.

The ebase supports Red Black Tree (RBT) indexes and range search. Indexes can
be created, deleted, and rebuilt.
l Support the data synchronization function.
Database synchronization refers to transferring the updated data in the in-memory
database to a third-party database in a quasi-realtime way to ensure data consistency.

The ebase provides a module that synchronizes data from a database to another
homogeneous or heterogeneous database, for example, from an Oracle or Sybase
database to the ebase, or from the ebase to an Oracle or Sybase database.
The OMM's operation and maintenance user interface (UI) provides a tool for users
to accomplish quick data synchronizations.
l Support iTool-based automatic installation and upgrade.
The ebase completely follows iTool standards to implement automatic installation
and upgrade. Engineering personnel first collect engineering information, use the
commissioning and upgrade planning UI to plan the nodes to be deployed, pack
the collected information into an equipment package or a product package, and
then send the package to the site. On-site engineers only need to perform simple
operations to accomplish automatic installation and upgrade.

l Support configuration management through the OMM GUI.

The ebase provides a system configuration management GUI on the OMM for users
to view current or historical data, modify configurations, and apply modifications.
l Support automatic preventive maintenance using iTool.

The ebase provides an iTool-based maintenance package. Users can use the
package to check the system operation status of each ebase node and configure
alarms for exceptions. Users only need to configure and save node information to
complete preventive routine maintenance.

l Support whole-process tracing.

1-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 1 ebase Introduction

Caution!
A whole-process tracing affects service operation and ebase performance, so contact
ZTE technical support for confirmation before you perform a whole-process tracing.

The ebase provides a simplified tool for tracing the execution process of a single SQL
command or a single stored procedure.

1.2 Application Scenario

The ebase is a memory database system and all its libraries, tables and indexes are stored
in the memory. This ensures a high performance, but the memory size restricts the scale
of the data that the ebase can manage. Therefore, the ebase is intended for scenarios
where the amount of data to be stored does not exceed the size of the physical memory
and that requires high read and write performance, for example, as a temporary database
cache for a service processor.

Note:
Because the ebase's synchronization module is designed for data management instead
of for synchronization or disaster recovery for a large amount of data, the synchronization
module is only suitable for synchronizing tables with little record changes.

The ebase and the service processor are co-located on the same server as the front-end
server. The service processor has a direct access to the ebase through the JDBC/ODBC
interface. The ebase synchronizes the updated service data to the commercial database
(for example, Oracle database) on the back end in real time.

Figure 1-1 shows the typical application scenario of the ebase.

1-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Figure 1-1 Typical ebase Application

1.3 System Architecture

Software Architecture
The ebase is a complete in-memory database system that supports the standard SQL
interface and does not support the transaction function. External applications can be
connected to the system through the ODBC or JDBC.
Figure 1-2 shows the ebase's software architecture.

Figure 1-2 Software Architecture

The software module of the ebase is described as follows:

l The connection pool receives SQL requests from clients, executes the received
requests, and then returns results.

1-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 1 ebase Introduction

l The analyzer analyzes SQL statements and generates graphs. The optimizer chooses
the optimal paths, for example, the optimal index according to the query criteria.
l The memory pool manages its own memory by frequently releasing small blocks of
memory during the process of analysis.
The memory storage engine accomplishes the data management function, including
the bitmap module and the page storage management module.
à The bitmap module is divided into level 1 bitmap management and level 2 bitmap
management. Level 1 bitmap management determines occupied or idle pages.
Level 2 bitmap management determines occupied or idle records.

à The page storage management module assigns table pages on which

management functions are recorded.
l Each index is an RBT tree. Indexes are sorted by key, and each tree node stores
a pointer pointing to a data record, which allows users to query ranges by index
efficiently. If synchronization is required for a table, any insertion or modification
operation to the table is written to a synchronization log, and the synchronization log
module writes the cached logs in the system to a synchronization log file periodically.

Components
The ebase includes the following components:
l extrabase: the ebase's core program, which is responsible for data storage
management, index management, data persistence, and processing of application
query requests.
l xsql: the ebase's built-in command line client tool. It is similar to the Sybase's isql and
allows users to connect to a database using xsql commands. Users can enter SQL
statements in the command line interface.
l ODBC library: The ebase provides a C programming interface. Applications can
invoke the corresponding Application Programming Interface (API) to execute SQL
statements for execution results.
Open Database Connectivity (ODBC) is a unified interface standard intended for
database access. ODBC provides client programs with a set of standard program
design interfaces for database applications, and the set of program design interfaces
can be used to compile application programs used to perform add, delete, modify,
query, and search operations to a database.

l JDBC library: Java programming interface provided by the system. Applications can
invoke the corresponding API to execute SQL statements for execution results.
Java DataBase Connectivity (JDBC) is a Java API used to execute SQL statements.
JDBC comprises a group of classes and interfaces compiled in the Java language.
JDBC provides a reference for compiling database application programs.

l Synchronization module: a tool provided by the system to accomplish incremental

synchronization from an ebase to a commercial database or from a commercial
database to an ebase.

1-5

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

à xsync_x2sx: accomplishes incremental synchronization from an ebase to a

Sybase database or to another ebase.
à xsync_x2o: accomplishes incremental synchronization from an ebase to an
Oracle database.

Note:
The log archive mode must be enabled for the Oracle database.

à xsync_s2x: accomplishes incremental synchronization from a Sybase database

to an ebase.
à xsync_o2x: accomplishes incremental synchronization from an Oracle database
to an ebase.
The ebase also supports the full synchronization function, including:

à Full synchronization from an Oracle database to an ebase.

Full synchronization includes synchronization by a fixed list and synchronization
by a selected list. A fixed list is a list registered in the xbase_sync_tabs
table. A selected list is a list registered by a user manually in the
LOAD_OUT_ORACLE_LIST table.
Before you start a full synchronization in an environment containing the
synchronization module, you must stop the synchronization module first or
ensure that no dml operation is performed to the source table during the process
of the full synchronization. If you start a full synchronization without stopping the
synchronization module first, an operation may be lost. For example, if a delete
operation that is performed before a full synchronization is completed fails to
delete the target record, this delete operation is lost.
à Full synchronization from an ebase to another ebase.
The source and destination ebases must use the same username, for example,
xbase.

à Full synchronization on the OMMP GUI.

The OMMP GUI provides the one-button full synchronization function that
supports full synchronizations from an Oracle database to an ebase and from an
ebase to another ebase (both must use xbase as the username).

Processes
The ebase database system provides two processes:

l xbasemoni

1-6

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 1 ebase Introduction

The ebase's monitor process, which monitors the extrabase process' status. The
xbasemoni process restarts the extrabase process if the extrabase process is in
abnormal status.
l extrabase
The ebase's main process, which implements the ebase in-memory database's
functions, including storage management, index management, data persistence, and
processing of application query requests.

Threads
The ebase database system is designed with a multi-thread architecture, which means
that multiple threads operate together, including:
l srv_lock_timeout_and_monitor_thread: monitor thread, which monitors the locks in
the database and provides clients with monitoring information.
l com_accept_or_recv: main communication thread, which establishes connections to
the database for clients and assigns a worker thread for each connection.
l srv_worker_thread: main thread for database processing. This thread analyzes SQL
statements, generates a syntax tree based on the analysis, executes the syntax tree,
and returns the final results to the client. Every established connection is assigned
such a thread.
l checkpoint: periodically flushes dirty pages in the shared memory to database files.

1.4 Operation Environment

Hardware Requirements
The hardware configuration principle for the ebase is multiple CPUs, large memory, and
high CPU core frequency.
The minimum hardware configurations:
l CPU: 4 Core
l Mem: 8 G
l Disk space: The idle disk space is at least twice of the memory size and must exceed
20G.
l The settings for the shmmax and shmall parameters of the operating system must
be greater than the size of the maximum shared memory required by the ebase.

Run the following commands if you are the root user:

echo 16202652842 (adjust in accordance with the actual requirements) >
/proc/sys/kernel/shmmax

echo 26843545600 (adjust in accordance with the actual requirements) >

/proc/sys/kernel/shmall

1-7

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Software Requirements
The ebase supports the Linux, AIX, Solaris, HP-UX, HP-IA and UNIX operating systems.
It is recommended that ebase operates in a 64–bit operating system.
The lowest operating system versions:
l HP-IA
The latest patch(es) must be installed, including PHCO_37940, PHKL_37121, and
PHKL_37122.
l SUSE Linux
SUSE Linux Enterprise Server 9 (Linux version 2.6.5-7.315-smp) and later.
l RedHat Linux
RedHat Enterprise Linux Server release 5.5 (Linux version 2.6.18-194.el5) and later.

1-8

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2
ebase Routine Maintenance
Maintenance tools for the ebase include the CLI client and the OMM. This chapter
describes the common maintenance tools and routine maintenance items.

Note:
This manual only describes the common maintenance tools. For information about other
tools, contact ZTE technical support.

Table of Contents
Maintenance Tools .....................................................................................................2-1
Maintenance Contents ...............................................................................................2-6

2.1 Maintenance Tools

2.1.1 CLI Client
Users can use the ebase-provided xsql command line interface (CLI) client to connect to
a database and enter SQL statements to execute the corresponding commands. For a
description of the common commands, refer to Table 2-1.

Note:
For SQL statements supported by the ebase, refer to “Appendix A”.

Table 2-1 Common Maintenance Commands

Command Description

xsql -v Prints the version number of the [Link] library that

the xsql is using.

xsql -s ebase IP -p ebase Logs in to an ebase.

port

xsql -s ebase IP -p ebase port Logs in to an ebase and executes a script file.
–i script file

2-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Command Description

source filename Performs an SQL operation in the xsql to execute the contents of a
file or execute file contents in batch mode.
If the target file is not in the current directory, the absolute path to
the file must be specified.

Enter \c at the xsql prompt Cancels the input contents. This command is similar to the reset
command of the isql.

go or “;” Submits a finished statement for execution.

2.1.2 OMM
The ebase operation and maintenance system (OMM) is based on the OMMP platform
and accomplishes the following functions: equipment management, performance
statistics, alarm management, configuration management, whole-process tracing and fast
synchronization. The OMM can be used to monitor the ebase server's operation status
and perform routine maintenance operations.

l Equipment management
Equipment management refers to maintaining the data of all the managed NEs
and configuring their office IDs, module numbers, IP addresses, and device types
(including ebase device information).
l Performance statistics
Performance statistics refers to collecting system data in order to provide accurate,
timely and necessary data for the carrier, network administration department, and
device manufacturers. The collected data serves as experience and bases for scheme
planning, configuration and adjustment.

l Alarm management

Note:

Alarm management is mainly applied to real-time fault observations and historical fault
queries.

Alarm management refers to detecting exceptions and locating, isolating and

eliminating faults in network operation based on the observation and analysis of
device status information.

l Configuration management
Ebase configuration management includes five steps: configuration data
warehousing, data configuration on the user interface, configuration file generation,

2-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

configuration delivery and notification, and application of the configuration. During

version upgrade and routine maintenance, users can maintain configuration data for
different NEs through the OMM without logging in to the hosts one by one. This also
simplifies the version upgrade process.
l Whole-process tracing
Whole-process tracing refers to using a graphic to show and monitor the detailed
information about the executed SQL statements and the execution results.
l Fast synchronization
The ebase OMM allows users to perform a full synchronization by pressing a button
on the page as shown in Figure 2-1.

Figure 2-1 ebase Fast Synchronization Tool

2.1.3 Preventive Routine Maintenance Tool

The iTool preventive routine maintenance tool achieves the automation of maintenance
items, facilitates maintenance operations, and reduces maintenance difficulty.

Note:

For operations to the iTool, refer to B.1 Automatic Preventive Maintenance Operations.
For maintenance items provided by the iTool, refer to B.2 Maintenance Items.

2.1.4 Full Backup Tools

The ebase provides two tools to make full backups, one is for the logical full backup function
and the other for the physical full backup function.

2-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Note:
For backup and recovery operations, refer to 3 ebase Backup and Recovery.

l Logical full backup

Use a tool to export data of all the tables in a database and save the exported data
as a local file.
l Physical full backup
Use the export command to save datafiles and data dictionaries in all the directories
on a database server to the corresponding files.

Logical Full Backup

The ebase provides two binary programs: loadoutall and loadinall, which can export data
of all the tables in a database as text files. A table name is a filename. By default, filenames
are automatically separated with a vertical bar (“|”), and lines are separated with a linefeed.
l The loadoutall program exports data of all or specified tables in a database to a
specified directory. Users can modify the field delimiter (vertical bar) and the line
delimiter (linefeed) are required.
l The loadinall program imports data from the backup directories to the corresponding
tables in a database. The field delimiter and line delimiter must be consistent with
those for the backup directories.
For common operations to the loadoutall program, refer to Table 2-2.

Caution!
The following operations also apply to the loadinall program. The sequence of the
parameters is the same. The delimiters must be consistent with the delimiters used for
the loadoutall program, otherwise, an error may occur.

Table 2-2 Common Operations to the loadoutall Program

Operation Command

Make backup and recovery without loadoutall /home/xbase/databak

modifying the delimiter.
loadoutall zxdbp_133 /home/xbase/databak

Specify the “#@” as the field loadoutall -f "#@" /home/xbase/databak

delimiter, and the carriage return as
loadoutall -f "#@" zxdbp_133 /home/xbase/databak
the default line delimiter. A username
can be specified.

2-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

Operation Command

Specify the “%#” as the line delimiter, loadoutall -l "%#" /home/xbase/databak

and the “|” as the field delimiter. A
loadoutall -l "%#" zxdbp_133 /home/xbase/databak
username can be specified.

Specify the “#@” as the field loadoutall -f "#@" -l "%#" /home/xbase/databak

delimiter, and the “%#” as the line
loadoutall -f "#@" -l "%#"zxdbp_133 /home/xbase/databak
delimiter. A username can be
specified.

1. Escape characters including “$” and “\” should not be used a delimiter.
2. To use the carriage return as the line delimiter, you only need to keep the line delimiter unspecified.

Physical Full Backup

The ebase provides ebase_export.sh as a backup script and ebase_import.sh as
a recovery script. Users can add a scheduled task in crontab to accomplish scheduled
backup. When a fault occurs, users can use the ebase_import.sh script to recover
data.

Note:
The ebase_export.sh script keeps backup data for three days, and the ebase_impor
[Link] script recovers data of the previous day.

The ebase_import.sh and ebase_export.sh scripts can be used in the SUSE Linux,
HP-UX, HP-IA, AIX, and Solaris operating systems, and any user with an installed ebase
can use these two scripts.
Before using these scripts, users must create the backup folder in the $HOME directory
first. After a backup is completed, a file named in the date and a corresponding tar file are
generated in the $HOME/backup directory, for example, 20120503 and [Link].
The 20120503 file is a logical backup, and the [Link] file is the corresponding
tar file.

2.1.5 load out/in Tool

The load out/in tool is used to back up and recover table data for a database.

l load out: exports data from a table to a file.

Common command: “load data infile '/home/xbase/tb_zxinaccode' into table

tb_zxinaccode field_term '|';”

à The load out command can export data of a specified field. Command: “ load data
outfile '/home/xbase/tb_zxinaccode' ziduan1,ziduan2 from table tb_zxinaccode
field_term '|';”.

2-5

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

à The load out command can load data of all the tables in a database to a specified
directory. Command: “load data all outdir 'path/directory' field_term '|';”.
l load in: imports data from a file to a table.
Command command: “load data outfile '/home/xbase/tb_zxinaccode' from table
tb_zxinaccode field_term '|';”
For a description of the command parameters, refer to Table 2-3.

Table 2-3 load out/in Command Parameter Descriptions

Parameter Description

tb_zxinaccode Name of the table to be backed up or recovered.

/home/xbase/tb_zxinaccode Path to the folder that stores the exported file.

field_term '|' Delimiter used to separate fields. Any other special character
or character string can be used as a delimiter, for example,
#, and @.
The data must not include the “|” delimiter.

2.2 Maintenance Contents

Maintenance Description
The ebase requires regular maintenance to ensure its normal operation. The maintenance
for the ebase includes manual maintenance and maintenance performed by the iTool. This
section describes the items, operations, and reference criteria for manual maintenance.
For the recommended maintenance items and periods for the ebase, refer to Table 2-4.

Note:
The maintenance periods and items can be adjusted according to site requirements. If you
are not sure, contact ZTE technical support for confirmation.

Table 2-4 Maintenance Periods and Corresponding Maintenance Items

Maintenance Period Maintenance Item

Daily maintenance 2.2.1 Checking the Operation Status

2.2.3 Checking the Number of Links

2.2.2 Checking for a Core File

2.2.4 Checking the Scheduled Backup Task

2.2.5 Checking the Synchronization Module

2-6

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

Maintenance Period Maintenance Item

Monthly maintenance 2.2.6 Checking the Related Interface Program

2.2.7 Checking Table Definitions

2.2.8 Checking Configuration Files

2.2.9 Checking the Dual-Server Cluster

Half-yearly maintenance 2.2.10 Checking the Version Number and Bit Number

Hazardous Operations
The following operations are hazardous and can lead to data corruption.

Caution!
The following situations should be avoided when the ebase is still operating.

l Power failure
A power failure causes the loss of the data generated since the latest checkpoint. It
can also cause data corruption.
A power failure causes the loss of the shared memory.
l Forced termination of a process

Caution!
The kill -9 command must not be used. If you must use this command, contact ZTE
technical support for confirmation first.

If you run the kill -9 command while a process is still running, it terminates the process
directly, which causes incomplete data writing and thus damages the datafile.
l Server switchover

If two servers are switched over at an improper time when some dirty pages on the
primary server are not written to the disk, file damages may occur.

2.2.1 Checking the Operation Status

Maintenance Purpose
This procedure describes how to check the ebase's operation status.

2-7

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to verify that the ebase processes are normal:
$ xstat
3. Run the following commands to enter the $HOME/log directory and check the extr
[Link] file:
$ cd $HOME/log
$ more [Link]
4. Run the following commands to enter the $HOME/log directory and check the
xbasemoni monitor process's log file [Link]:
$ more [Link]
5. Run the following commands as the root user to check the CPU and memory
resources occupied by the extrabase process:
$ su - root
# top

Reference Criteria
l The xstat command shows that the extrabase and xbasemoni processes are
operating.
l The logs in the [Link] file do not contain any keywords indicating an error
or exception, for example, error, failed.
l The logs in the [Link] file do not contain any keywords indicating an error
or exception, for example, error, failed.
l The top command shows that the CPU utilization of the extrabase process remains
below 50% most of the time, and the memory utilization (in the %MEM column) does
not have a significant increase during a certain period.

Note:
The memory utilization criterion varies, depending on the site situation. If you are not
sure, send the check results to ZTE technical support for confirmation.

Exception Handling
1. If the extrabase and xbasemoni processes are not operating:
Check whether they are terminated manually, and check whether the extrabase.l
og and [Link] files indicate any exception.

2-8

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

2. If the [Link] or [Link] file indicates an exception:

Analyze the logs and troubleshoot accordingly. If you cannot troubleshoot, save the
logs and contact ZTE technical support.
3. If the CPU utilization remains high:
Run the # free -m command to check whether swap has been used, and whether it is
the peak hour. Check the [Link] file for logs indicating an exception. Save
the logs and check results, and contact ZTE technical support.

2.2.2 Checking for a Core File

Maintenance Purpose
This procedure describes how to check for a core file and determine whether the ebase
processes have crashed.

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to verify that the core file exists:
$ find ./ -name core -print

Reference Criteria
No result is output.

Exception Handling
If any core file exists, send it to ZTE technical support.

Related Information
In the directory containing a core file, run the following command as the user who installed
the ebase to determine the program that generated the core file:
$ file core file name

2.2.3 Checking the Number of Links

Maintenance Purpose
This procedure describes how to verify that the number of links is normal. The ebase
restarts when the number of links exceeds the set number.

Steps
Follow these steps:

2-9

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to check the max_n_session item in the [Link]
configuration file in $HOME/etc:
$ more $HOME/etc/[Link]
3. Run the following command to log in to the ebase database:
$ xsql -s IP -p port
4. On the xsql client, run the following command to check the number of links:
xsql> show session;

Reference Criteria
The number of links shown by the show session command is lower than the 80% of the
value set for the max_n_session parameter.

Exception Handling
If the number of current links exceeds 80% of the value set for the max_n_session
parameter, contact ZTE technical support for confirmation. After the confirmation,
increase the value for the max_n_session parameter in accordance with the actual
requirements and then restart the ebase at a proper time.

Note:
The value set for the max_n_session parameter should be appropriate. Each link
occupies a certain amount of memory, and a too great value wastes the memory resource.

2.2.4 Checking the Scheduled Backup Task

Maintenance Purpose
This procedure describes how to check the scheduled backup task.

Note:
It is not allowed to edit a running scheduled task by using the crontab -e command. If
necessary, contact ZTE technical support for confirmation.

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.

2-10

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

2. Run the following command to check the scheduled task:

$ crontab -l

Reference Criteria
The scheduled task is set to back up data for the ebase every day.

0 3 * * * /home/xbase/bin/ebase_export.sh >/dev/null 2>&1

Exception Handling
If the scheduled task is not set to back up data for the ebase every day, deploy the ebase_
[Link] script again. For the detailed procedure, contact ZTE technical support.

2.2.5 Checking the Synchronization Module

Maintenance Purpose
This procedure describes how to verify that the synchronization module is operating
properly.

Steps
Follow these steps:
1. Determine the synchronization module that the service is using.

Note:
If you cannot determine the synchronization module, contact ZTE technical support.

2. Run the following command as the ZXINOS user to verify that the synchronization
process is operating properly:
$ zxstat
3. Check the corresponding log file of the synchronization module. For the
correspondence relationships between synchronization modules and log files, refer
to Table 2-5.
Table 2-5 Synchronization Modules and Corresponding Log Files

Synchronization Module Corresponding Log File

xsync_o2x xsync_o2x.log

xsync_s2x xsync_s2x.log

xsync_x2o xsync_x2o.log

xsync_x2sx xsync_x2sx.log

2-11

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Reference Criteria
l The synchronization module exists.
l The corresponding log file does not contain any keyword indicating an error or
exception, for example, error or failed.

Exception Handling
If the process does not exist, check the synchronization logs for exceptions, and then
handle the exceptions accordingly.
If you cannot handle the exception, contact ZTE technical support.

Related Information
The following takes the o2x synchronization module as an example to describe the
procedure for handling an exception. The procedures for other synchronization modules
are similar.
1. Run the following command to log in to the Oracle database as the sysdba user:
$ sqlplus “/ as sysdba”
2. Execute the following scripts and record the number of results:

col STREAMS_NAME format a15

col rule_owner format a15
col rule_name format a12
col RULE_CONDITION FORMAT A35
col ORIGINAL_RULE_CONDITION FORMAT A35
select streams_name,streams_type,RULE_OWNER, RULE_NAME,RULE_TYPE FROM
DBA_STREAMS_RULES WHERE STREAMS_NAME LIKE 'O2X_CAPTURE%' AND
STREAMS_TYPE = 'CAPTURE' ORDER BY streams_name;
select streams_name,streams_type,RULE_OWNER, RULE_NAME,RULE_TYPE FROM
DBA_STREAMS_RULES WHERE STREAMS_NAME LIKE 'O2X_APPLY%' AND
STREAMS_TYPE = 'APPLY' ORDER BY streams_name;

3. Run the following command, check the number of the registered tables, and then check
whether the number obtained in step 2 is twice the number of the registered tables:
SQL> select count(*) from xbase_sync_tabs;
l Yes: proceed to step 4.
l No: proceed to step 6.
4. Execute the following scripts and check whether each captured and applied process
is enabled:

select capture_name,status from dba_capture where capture_name like 'O2X_CAPTURE%';

select apply_name,status from dba_apply where apply_name like 'O2X_APPLY%';

l Yes: proceed to step 5.

2-12

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

l No: proceed to step 6.

5. Run the following command to log out of the database:
SQL> exit
6. Check whether the Oracle database log file alert_zxin.log contains erro, fail, or
ora-number, and check whether the errors are related to the stream.

Note:
By default, the database log file is in the $HOME/zxindbf/admin/zxin/bdump/
directory. If the directory contains multiple log files, refer to the latest one.

l Yes: proceed to step 7.

l No: proceed to step 8.
7. Troubleshoot according to the log information, and then verify that the problem is
solved.
l Yes: end.
l No: proceed to step 8.
8. Contact ZTE technical support.

2.2.6 Checking the Related Interface Program

Maintenance Purpose
This procedure describes how to check the [Link] used by the SDF, IMP,
synchronization module, or any other program that uses ODBC to connect to the ebase.
The purpose is to ensure that the [Link] being used is the same as the li
[Link] in the ebase version.

Steps
Follow these steps:
1. Determine the processes that are using [Link]. Contact ZTE technical
support for more information.
2. Log in to the ebase server as the user who installed the ebase.
3. Run the ldd target program command, for example, ldd sdfpro.

Reference Criteria
The [Link] that the target program is using (command result) is the same
as the [Link] in $HOME/lib, including the filename and size.

2-13

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Exception Handling
If the [Link] that the target program is using is different than the libxbase
[Link] in $HOME/lib, stop the target program and replace its [Link].
Contact ZTE technical support if necessary.

2.2.7 Checking Table Definitions

Maintenance Purpose
This procedure describes how to check an index and ensure that the index is consistent
with the index in the script.

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to log in to the ebase database:
$ xsql -s IP -p port
3. On the xsql client, run the following command to check the index of a major table:
xsql> desc table table_name;

Reference Criteria
The checked index is consistent with the index in the script.

Note:
The gen_clust_index output from the desc command is a built-in object of the system, not
an index.

Exception Handling
If the checked index is different from the index in the script, create the index manually in
accordance with the script.

Note:
The ebase supports dynamic creation of an index.

Contact ZTE technical support if necessary.

2-14

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

2.2.8 Checking Configuration Files

Maintenance Purpose
This procedure describes how to check configuration files for the ebase and ensure that
they meet requirements.

Caution!
Ebase datafile and directory must not be deleted or moved, otherwise, there will be serious
consequences.

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to check the $HOME/etc/[Link] configuration file:
$ more $HOME/etc/[Link]
3. Run the following command to check the $HOME/etc/[Link] configuration
file:
$ more $HOME/etc/[Link]
4. Check whether any device is running multiple ebase instances:

Reference Criteria
l If the ebase must be synchronized with any other database, the
redolog_used_by_xsync parameter must be set to 1.
l If the setting for the page_count parameter is greater than 8 G, ensure that the
settings for shm_max and shm_all are sufficient for the operating system. For the
Linux operating system, the default setting for shm_all is 8 G, and the ebase cannot
start up if the setting for shm_all is greater than the default.

Note:

The modification to these parameters must be fixed to the startup of the OS, otherwise,
the modification is null after a restart.

l In the case of a cluster (including a 1+1 cluster and an n+m cluster), ensure that the
data, dict, and redolog directories reside on the shared storage device.
l The server_share_mode parameter is set to 0.

2-15

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

l The default settings for sql_timeout and sql_timeout2 are 50 and 60 respectively.
The settings for these two parameters should not be too great, otherwise, the service
will be interrupted for a too long time.

Note:
The settings can be modified in accordance with service requirements. Contact ZTE
technical support for confirmation if necessary.

l The IP address and port number configured for the destination ebase in the xbasem
[Link] must be consistent with that configured in [Link].

Note:
The setting for the XbaseSrvIP parameter in [Link] must be consistent
with the setting for the bind_ip parameter in [Link]. And the setting for the
XbaseSrvPort parameter in [Link] must be consistent with the setting for
the bind_port parameter in [Link].

l The work_mode parameter is set to 0 (which means work_mode=0).

l If it is required to run multiple ebase instances on one device, ensure that the
key_of_shm of each ebase's configuration file ($HOME/etc/[Link]) is unique,
and their paths of the bind_port, data, dict, redolog directories are not repeated.

Exception Handling
Modify the settings to meet actual requirements. Contact ZTE technical support if
necessary.

Note:
The dynamic in a description for a configuration item indicates that the configuration item
is dynamically effective, which means a modification to the configuration item takes effect
in one minute after the modification is completed. If a description for a configuration item
does not contain dynamic, a modification to the configuration item takes effect only when
the ebase is restarted.

2-16

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 2 ebase Routine Maintenance

2.2.9 Checking the Dual-Server Cluster

Maintenance Purpose
This procedure describes how to check the dual-server cluster and ensure that the two
servers operate properly.

Steps
Follow these steps:

1. Log in to the active and standby servers as the user who installed in the ebase, for
example, zxin10, and then perform these steps:
l In the $HOME/etc directory on the active and standby servers, check whether
the [Link] and [Link] in the configuration files are consistent.
l Check whether the dual-server cluster scripts are correct:
l Run the following command to check the processes:
$ xstat

l Run the following command to check the shared memory:

$ ipcs -m
2. Log in to the active server as the user who installed the ebase, for example, zxin10, and
then enter the shared storage directory to check whether the shared storage directory
is correctly mounted.

Reference Criteria
l The [Link] and [Link] in the configuration files of the active and
standby servers are consistent.
l Use the xbasemoni monitor process to start ebase processes, and use the xbasemoni
command to stop ebase processes.
l The xbasemoni and extrabase processes are not running on the standby server, and
the shared memory specified by the key_of_shm configuration item in the xbase
.cfg configuration file does not exist on the standby server. The xbasemoni and
extrabase processes are running on the active server.
l The paths configured in the [Link] configuration file are on the disk array. The
paths for the datafiles on the disk array include map_file_path, data_dict_path,
redo_log_path.

Exception Handling
1. If the configuration files are not consistent:

Modify the settings and check whether the configuration items take effect dynamically.
If not, restart the ebase to apply the modification.
2. If the dual-server cluster scripts are incorrect:

2-17

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Stop the dual-server cluster at a proper time and then modify the scripts. Contact ZTE
technical support if necessary.
3. If the paths of the datafiles are not on the disk array:
Stop the dual-server cluster, copy the datafiles to the disk array, modify the xbase.c
fg item, and then start the dual-server cluster.
4. If the xbasemoni or extrabase process or the shared memory is found on the standby
server, delete them manually.
a. Run the kill -9 xbasemoni command.
b. Run the xstop command.

c. Delete the shared memory that the ebase is using on the standby server.
Contact ZTE technical support if necessary.

2.2.10 Checking the Version Number and Bit Number

Maintenance Purpose
This procedure describes how to check whether the ebase version number and bit number
used on site meet the requirements.

Steps
Follow these steps:
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to check the ebase version number:
$ extrabase -v
3. In the $HOME/bin directory, run the following commands to view the bit number:

$ cd $HOME/bin
$ file extrabase

Reference Criteria
l The version number is the same as the required version number.
l The extrabase process is 64–bit.

Exception Handling
If the version number or bit number used on site is different from the required, contact ZTE
technical support to confirm whether to replace the version.

2-18

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 3
ebase Backup and Recovery
Backup should be performed regularly to prevent data corruption and security issues
caused by improper operations. When the ebase fails, data or datafiles can be recovered
immediately using the corresponding backup files.
Table of Contents
Backup Period............................................................................................................3-1
Physical Backup and Recovery ..................................................................................3-2
Logical Backup and Recovery ....................................................................................3-5

3.1 Backup Period

Backup Classification
Based on the backup object, ebase backup is classified as physical backup and logical
backup.
l Physical backup refers to the process of making a checkpoint to keep the memory
consistent with datafiles and copying datafiles and the data dictionary.
Physical backup files can be used to recover data in different ebase environments.
l Logical backup refers to executing the load out SQL command to export table data
from a database to a file as a backup.
Logical backup is classified as table-level backup, database-level backup, and
system-level backup. Users can recover data using the corresponding backup files
in accordance with the specific fault.
Based on the backup method, ebase backup is classified as automatic backup and manual
backup.
l Automatic backup refers to setting a scheduled task in the crontab and using the
scheduled task to automatically execute the ebase_export.sh script. The ebase
_import.sh script can be used to recover data directly.
The ebase_import.sh script keeps backup data for three days, and the ebase_i
[Link] script recovers data of the previous day.

Note:
Automatic backup is system-level logical backup.

3-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

l Manual backup refers to backing up ebase data manually.

Note:
Manual backup includes physical backup and logical backup and can be performed
according to site requirements.

Backup Period
l Automatic backup: The scheduled task performs backup every day at the backup time
when the service is idle.
l Manual backup: A backup should be performed manually before a service upgrade,
ebase upgrade or a major operation to the system.

3.2 Physical Backup and Recovery

3.2.1 Physical Backup
This procedure describes how to back up the ebase datafile. When the datafile is damaged
or corrupted, it can be recovered immediately using the backup file.

Prerequisite
l The ebase is operating properly. To determine the ebase's status, refer to 2.2.1
Checking the Operation Status.
l There is sufficient free disk space.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to log in to the ebase on the xsql client:
$ xsql -s IP -p port

3. Run the following command to back up the datafile:

xsql> export '/home/zxin10/exportdata';

4. Enter the backup directory to verify that the backup files are generated. The generation
time of the backup files should be the current time.

3-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 3 ebase Backup and Recovery

Note:
The datafile and data dictionary are used together and must be backed up
simultaneously.

Two files are generated in the /home/zxin10 directory: one is the exportdata
datafile, and the other is the [Link] data dictionary.
– End of Steps –

3.2.2 Physical Recovery

This procedure describes how to use a backup file to recover the ebase datafile when the
datafile is damaged or corrupted.

Prerequisite
l The ebase is operating properly.
l The backup file is stored in the /home/zxin10 directory, and you have the right to
read and write it.
l The free disk space is sufficient.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.

2. Run the following command to log in to the ebase on the xsql client:
$ xsql -s IP -p port
3. Run the following command to recover the datafile:

Note:
The /home/zxin10 directory contains the exportdata datafile and the exportda
[Link] data dictionary. You only need to run the import ‘/home/zxin10/exportdata’
the command to complete the recovery.

xsql> import '/home/zxin10/exportdata';

The following shows an example.

3-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Note:
After the import replaces the backup datafile with the original datafile, the ebase
must be restarted to import data to the memory again. Because the ebase exits
automatically when the import finishes the execution and then it is started by the
xbasemoni process, the connection is broken message that appears is normal.

xsql> import '/home/zxin10/exportdata';

connection is broken
PSM:initialization OK!!

4. After the recovery is completed, verify that the recovery is successful. For the detailed
procedure, refer to 3.2.3 Verifying a Physical Recovery.
– End of Steps –

3.2.3 Verifying a Physical Recovery

This procedure describes how to verify that a physical recovery is successful and the ebase
operates properly after the recovery.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to check the [Link] log file:
$ more [Link]
The following shows an example of logs for the import command that is executed
successfully:

11-22 21:51:32 F[psm0psm.c] L[5408] PSM:psm_import_copyfile():open database backup

file[/home/hanyi/exportdata] ok!!
11-22 21:51:32 F[psm0psm.c] L[5419] PSM:psm_import_copyfile():open dest database
file[/home/hanyi/data/[Link]] ok!!
11-22 21:51:32 F[psm0psm.c] L[5431] PSM:psm_import_copyfile():convert back up file into
system map file ok!!
11-22 21:51:32 F[load0load.c] L[1053] mem_backup_step: Import finished successfully and
ebase will [Link] xbasemoni to restart ebase!

3. Run the following command to log in to the ebase through the xsql client:
$ xsql -s IP -p port
4. Verify that the correct tables and data are recovered, and the recovered data is valid
and complete.
– End of Steps –

3-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 3 ebase Backup and Recovery

3.3 Logical Backup and Recovery

Logical backup and recovery refers to exporting and importing table data and is divided
into three levels: table level, database level and system level. The application scenarios
are as follows:

Note:
By default the system-level backup and recovery is recommended.

l Table level
Export the data of one or multiple tables to specified files. Table-level backup and
recovery is applied when the service requires the export of data for a table.
l Database level
Export the data of all the tables in a database to specified files, and the data of each
table is saved as an independent file. Database-level backup and recovery is applied
when the service requires the export of all the table data in a database.
l System level

Export the data of all the tables in the ebase to specified files, and the data of
each table is saved as an independent file. System-level backup and recovery is
applied when a scheduled task performs a daily backup or the site requires a backup
operation.

3.3.1 Making a Table-Level Backup

This procedure describes how to export the data of one or multiple tables to specified files.
When the data in a table is damaged or corrupted, the data can be recovered using the
corresponding backup file.

Prerequisite
l The ebase is operating properly.
l The free disk space is sufficient.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to log in to the ebase through the xsql client:

$ xsql -s IP -p port
3. Run the following command on the xsql client to back up data for the target table:

3-5

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Note:
The field_term '|' parameter indicates that the field delimiter is “|”. The default line
delimiter is the carriage return.

xsql> load data outfile '/home/zxin10/table1' from table table1 field_term '|';
4. (Optional) Enter the /home/zxin10 backup directory to verify that the table1
backup file is generated.
The generation time of the table1 file must be the current time.
– End of Steps –

3.3.2 Making a Table-Level Recovery

This procedure describes how to recover one or multiple tables using the corresponding
backup file(s).

Prerequisite
l The ebase is operating properly.
l The backup files are in the backup directory, and you have the right to read and write
these backup files.
l There is sufficient free disk space.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to log in to the ebase through the xsql client:
$ xsql -s IP -p port
3. Run the following command on the xsql client to recover data for the target table:

Note:
The field_term '|' parameter indicates that the field delimiter is “|”. The default line
delimiter is the carriage return. The field and line delimiters used in this command
must be consistent with the delimiters used in the backup procedure.

xsql> load data infile '/home/zxin10/table1' into table table1 field_term '|';
4. After the recovery is completed, verify that the recovery is successful. For the detailed
procedure, refer to 3.3.7 Verifying a Logical Recovery.
– End of Steps –

3-6

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 3 ebase Backup and Recovery

3.3.3 Making a Database-Level Backup

This procedure describes how to export the data of all the tables in a database to specified
files. The data of each table is stored in an independent file. When the data in the database
is damaged or corrupted, the data can be recovered using the backup files.

Prerequisite
l The ebase is operating properly.
l The free disk space is sufficient.
l The ebase installation directory (bin) contains the loadoutall tool.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to back up the data of all the tables in the target database.
The following takes the zxdbp_133 as an example.

Note:
The default field delimiter is “|”, and the line delimiter is the carriage return. This
command exports the data of all the tables in zxdbp_133 to the /home/xbase/da
tabak directory. A file is generated for each table, and the filename is the table name.

$ loadoutall zxdbp_133 /home/xbase/databak

3. (Optional) Enter the backup directory to verify that the backup files are generated.
– End of Steps –

3.3.4 Making a Database-Level Recovery

This procedure describes how to load the data of backup files to the tables of a database
to recover the database.

Prerequisite
l The ebase is operating properly.
l The free disk space is sufficient.
l The ebase installation directory (bin) contains the loadoutall tool.
l The backup files are in the recovery directory, and you have the right to read and write
these backup files.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.

3-7

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

2. Run the following command to recover the target database. The following takes
zxdbp_133 as an example.

Note:
The field and line delimiters used in the loadinall command must be consistent with
the delimiters used in the loadoutall command. Otherwise, the recovery will fail.

$ loadinall zxdbp_133 /home/xbase/databak

3. After the recovery is completed, verify that the recovery is successful and the ebase
is operating properly. For the detailed procedure, refer to 3.3.7 Verifying a Logical
Recovery.
– End of Steps –

3.3.5 Making a System-Level Backup

This procedure describes how to export data of all the tables in the ebase to specified files.
Each table is assigned a file to store data for the table.

Prerequisite
l The ebase is operating properly.
l The free disk space is sufficient.
l The ebase installation directory (bin) contains the loadoutall tool.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to complete a system-level backup:

Note:
The default field delimiter is “|”, and the line delimiter is a carriage return. This
command loads the data of all the tables in the system to the /home/xbase/databak
directory. A file is generated for each table, and the filename is the table name.

$ loadoutall /home/xbase/databak
3. (Optional) Enter the backup directory to verify that the backup files are generated and
the backup is successful.
The generation time of the backup files must be the current time.
– End of Steps –

3-8

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 3 ebase Backup and Recovery

3.3.6 Making a System-Level Recovery

This procedure describes how to load the data from the system database files to the tables
of the system database.

Prerequisite
l The ebase is operating properly.
l The free disk space is sufficient.
l The ebase installation directory (bin) contains the loadoutall tool.
l The backup file is in the backup directory, and you have the right to read and write the
backup file.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to recover the system database:

Note:
The field and line delimiters used in the loadinall command must be consistent with
the delimiters used in the loadoutall command, otherwise, the recovery will fail.

$ loadinall /home/xbase/databak
3. After the recovery is completed, verify that the recovery is successful. For the detailed
procedure, refer to 3.3.7 Verifying a Logical Recovery.
– End of Steps –

3.3.7 Verifying a Logical Recovery

This procedure describes how to verify that a table-level, database-level, or system-level
recovery is successful.

Steps
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Check whether the print by the load/loadinall command reports any error:
If the following message is displayed, it indicates that the recovery is successful.

In all, loadin [nn] tables successfully

3. Run the following command to check the [Link] log file:

$ more [Link]

3-9

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

The following log indicates a successful recovery:

F[load0load.c] L[315] 1 rows loaded in [table=table1] successfully

4. Run the following command to log in to the ebase through the xsql client:
$ xsql -s IP -p port
5. Verify that the correct tables and data are recovered, and the recovered data is valid
and complete.
– End of Steps –

3-10

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4
Common Problems
Table of Contents
Common ebase Operation Problems..........................................................................4-1
Common Client Problems...........................................................................................4-6
Synchronization Module Problems .............................................................................4-8
Common xbasemoni Faults ......................................................................................4-14
OMM Fault ...............................................................................................................4-15
iTool Faults...............................................................................................................4-16

4.1 Common ebase Operation Problems

4.1.1 Determining the Cause of the Stopped ebase Process
Fault Symptom
The ebase process stops running during operation. This procedure describes how to
determine the cause: a manual shutdown or a fault.

Fault Cause
The cause can be located by analyzing the $HOME/log/[Link] file.
l The following log indicates that the process is stopped manually:

shutdown_step: received shutdown normal, exit!

l If the above log is not found, the process may have been terminated by the kill -9
command. If a core file is found, it indicates that the ebase exited due to a fault.

4.1.2 Troubleshooting the “Extrabase: create listen socket err,

exit” Error
Fault Symptom
The $HOME/log/[Link] file contains an “Extrabase: create listen socket err,
exit” error.

Fault Cause
Failed to set up listening.

4-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Recommended Procedure
1. Open the $HOME/etc/[Link] configuration file as the user who installed the
ebase, for example, zxin10, and then record the values of the bind_ip and bind_port
parameters.
2. Run the following commands as the root user to check whether the server IP address
is incorrect and whether the server port number has been occupied:
# netstat -an | grep port
# ifconfig -a
l Yes: proceed to step 3.
l No: proceed to step 4.
3. Reconfigure the port number or IP address, restart the ebase to apply the
configuration, and then check whether the log is still generated:
l Yes: proceed to step 4.
l No: end.
4. Restart the server to check whether the log is still generated:
l Yes: proceed to step 5.
l No: end.
5. Contact ZTE technical support.

4.1.3 Troubleshooting the “can't find xbase_trace.log” Error

Fault Symptom
The “can't find xbase_trace.log” error is reported when the service software is operating.

Fault Cause
The log file does not exist.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase, for example, zxin10.
2. Run the following command to generate the xbase_trace.log file:

Caution!
Running the tracing command decreases the execution efficiency, so tracing should
be performed during idle hours and terminated immediately after it is completed.

$ set trace on all

3. Run the following command to stop the tracing:
$ set trace off all

4-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

4.1.4 Copying Data Between Two ebase Servers

Fault Symptom
This procedure describes how to copy data from one ebase server to another ebase server.

Recommended Procedure
1. Log in to the source and destination servers as the user who installed the ebase, for
example, zxin10.
2. Run the following command on both servers to stop the ebase service:
$ xsuperstop
3. Run the following command to pack the data, dict, and redolog directories on the
source server:
$ tar -cvf [Link] data dict redolog
4. Upload the [Link] file to the corresponding directory on the destination server
through FTP in binary mode.
5. Run the following command to extract the [Link] file:
$ tar -xvf [Link]
6. Run the following command on both servers to start the ebase service:

$ xbasemoni

4.1.5 Modifying the Shared Memory

Fault Symptom
When users modify settings for the shmmax and shmall parameters, they must also
modify the /etc/rc.d/[Link] file, otherwise the modification disappears when the
server is restarted.

Note:

For different operating systems, the [Link] filename may be different. It should be
modified in accordance with the actual requirements.

Recommended Procedure
1. Log in to the ebase server as the root user.
2. Run the following command to modify the values for the shmmax and shmall
parameters.

4-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Note:
The values for these parameters are not fixed, depending on the actual requirements.
If you are not sure, contact ZTE technical support.

# vi [Link]
echo 16202652842 > /proc/sys/kernel/shmmax
echo 4147483648 > /proc/sys/kernel/shmall

3. Run the following command to apply the modification:

# /etc/rc.d/[Link]

If the settings are not modified after the system is restarted (the /etc/rc.d/[Link] file is
not executed), run the following commands to create a link file pointing to /etc/rc.
d/[Link]:
# cd /etc/rc.d/rc5.d/
# ln -s ../[Link] S99local

4.1.6 Troubleshooting the ebase's Failed Startup due to a

Configuration Modification
Fault Symptom
The ebase cannot start up, and the $HOME/log/[Link] file contains the
following logs:

BM:page_init_pages():g_dwDataPageNum=16384, g_dwUsefulBMBytes=2048!
PSM:initializing page BITMAP OK!
PSM:configure error:Integrality Level of current configure is not same as existing database!!
extrabase_start_or_create: psm_sys_load() failed, exit.

Fault Cause
The ebase cannot start up because the interity_level parameter in the $HOME/etc/xba
[Link] configuration file is changed or replaced.

Recommended Procedure
Restore the original value of the interity_level parameter in the [Link] file and then
restart the ebase service.

4-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

Note:
The value of the interity_level parameter must not be changed during the service
operation.

4.1.7 Troubleshooting the ebase's Failed Startup due to File

Permission Issues
Fault Symptom
The ebase cannot start up, and the $HOME/log/[Link] file contains the
following logs:

PSM:psm_read_config():fail to create database state file!!

PSM:psm_sys_dict_init():fail to read system configure from file [Link]!!
extrabase_start_or_create: psm_sys_load() failed, exit!

Fault Cause
At startup, the ebase automatically reads the datafile. If the datafile does not exist, the
above logs are generated automatically.
If the logs indicate that the ebase cannot read the datafile at startup, it is because the
ebase does not have the permissions.

Recommended Procedure
1. Log in to the ebase server as the root user.
2. Run the following command to modify the permissions to the data and dict
directories:
# chmod 777 data dict
3. Restart the ebase service to verify that the startup is successful.

4.1.8 Locking the ebase's Memory to the Physical Memory

Fault Symptom
This procedure describes how to lock the memory that the ebase uses to the physical
memory so that the ebase does not use swap space. This increases the ebase's operation
speed.

Fault Cause
The lock_memory parameter in the $HOME/etc/[Link] configuration file controls
whether to lock the memory to the physical memory.

4-5

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Note:
The lock_memory parameter is effective for Linux kernel 2.6.9 or later and it is not effective
for an NT system. For the AIX, Solaris, and HP-IA systems, this parameter is effective only
when the root user is running the ebase.

Recommended Procedure
The following takes Linux as an example to describe the procedure:
1. Log in to the ebase server as the root user.
2. Modify the /etc/security/[Link] file by appending the following contents
to the file:

* soft memlock unlimited

* hard memlock unlimited

3. Modify the setting for the lock_memory parameter in the $HOME/etc/[Link]

configuration file as the user who installed the ebase.
Set the lock_memory parameter to 1.
4. Restart the ebase to apply the modified configuration.

4.2 Common Client Problems

4.2.1 Running the Tracing Command
Fault Symptom
This procedure describes how to use the ebase's internal tracing command set sqltrace on.

Caution!
Using the set sqltrace on command affects the system performance, so this command
should be used during off-peak hours.

Fault Cause
1. Log in to the ebase server as the user who installed the ebase.
2. Run the following command to start tracing:

4-6

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

Note:
The tracing will automatically stop in 30 minutes.

$ set sqltrace on
3. Check the [Link] log file in the $HOME/log directory.
The log file prints the ten statements that have been executed for the longest period
recently.

4.2.2 Troubleshooting an Error Reported at a Login through the

Client
Fault Symptom
The following error is reported during a login to an ebase database from the ebase server
through the xsql client.

$ xsql -s [Link] -p 8433

Oracle XML Developers Kit [Link].0 - Production
XML-25001: Cannot locate requested XSQL file. Check the name.

Fault Cause
The Oracle database has a tool named xsql, which has the same name as the ebase's
CLI client xsql. In the PATH environment variable, the ebase's bin directory appears at
the end of the path name, and it is required to move the bin directory to the front.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase.
2. Modify the PATH environment variable by moving the bin directory in front of oralc
e/bin so that the program in the bin directory will be found first during the program
operation.
The modified variable is as follows.

PATH=$HOME/bin:$ORACLE_HOME/bin:$PATH:/sbin:.;export PATH

4.2.3 Troubleshooting Failed Data Insertion in the Database

Fault Symptom
The insert command is correct but data cannot be inserted in the database.

4-7

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Fault Cause
The data cannot be inserted because the database is full.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase.
2. Check whether the [Link] log file contains the following log:
This log indicates that the database is full:

Memory full,no page available!

l Yes: proceed to step 3.

l No: proceed to step 6.
3. Run the following command to stop the ebase server:

$ xsuperstop
4. Increase the value of the page_count parameter in the [Link] configuration file.
5. Run the following command to start the ebase server and then test whether data can
be inserted successfully:
$ xbasemoni

l Yes: end.
l No: proceed to step 6.
6. Contact ZTE technical support.

4.2.4 gen_clust_index Description

Fault Symptom
When a user runs the desc table command on the xsql client, each table has a
gen_clust_index.

Fault Cause
The gen_clust_index is an internal object of the system. It is not a real index and can be
ignored.

4.3 Synchronization Module Problems

4.3.1 Troubleshooting a Synchronization Error
Fault Symptom
The following logs are printed on the ebase server during a synchronization operation:

4-8

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

redo_open_log_file:Fail to open file

[/home/zxin10/redolog/XLOG_10.42.102.232_20100819151222_0001357340.r].
redo_open_log_file:Fail to open file
[/home/zxin10/redolog/XLOG_10.42.102.232_20100819151222_0001357341.r].
redo_open_log_file:Fail to open file
[/home/zxin10/redolog/XLOG_10.42.102.232_20100819151222_0001357342.r].
redo_open_log_file:Fail to open file
[/home/zxin10/redolog/XLOG_10.42.102.232_20100819151222_0001357343.r].

Fault Cause
The disk space is insufficient. The ebase cannot write the redolog file properly, and the
size of the generated regolog file is 0.

Recommended Procedure
1. Log in to the ebase server as the root user.
2. Check the disk and delete the obsolete files.

Note:
Double-check and make a backup before you delete a file.

4.3.2 Troubleshooting a Synchronization Failure due to the

Disabled Archive Mode
Fault Symptom
During a synchronization from an Oracle database to the ebase, executing init_strea
[Link] to initialize the synchronization failed, and the following logs are printed:

Add supplemental log data for ZXDB_VPBX.TB_SYSTBL done!

Setup queue0 done!
Failed to add capture rules for ZXDB_VPBX.TB_SYSTBL!
Set instantiation for ZXDB_VPBX.TB_SYSTBL done!
Add apply rule for ZXDB_VPBX.TB_SYSTBL done!
Set dml handler for ZXDB_VPBX.TB_SYSTBL done!
Set ddl handler for apply0 done!

Fault Cause
The archive mode is not enabled for the Oracle database.
1. Log in to the database server as the oracle user.
2. Run the following commands to check the database mode:

4-9

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

$ sqlplus / as sysdba
SQL> archive log list;
The check results are as follows:

SQL> archive log list

Database log mode No Archive Mode
Automatic archival Disabled
Archive destination /zxinbak/arch
Oldest online log sequence 19
Next log sequence to archive 21
Current log sequence 21

Recommended Procedure
1. Log in to the database server as the oracle user.
2. Run the following commands to log in to the database:
$ sqlplus /nolog

SQL> connect / as sysdba;

3. Run the following command to shut down the database:
SQL> shutdown immediate;
4. Run the following command to start and load the database:
SQL> startup mount;
5. Run the following command to switch the database to archive mode:

SQL> alter database archivelog;

6. Run the following command to switch the database to open status:
SQL> alter database open;
7. Run the following command to start the automatic database archive function:
SQL> alter system archive log start;
8. Run the following command to check whether the database is set to archive mode
successfully:

SQL> archive log list

If the database is in archive mode, the result is as follows:

4-10

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

SQL> archive log list

Database log mode Archive Mode
Automatic archival Disabled
Archive destination /zxinbak/arch
Oldest online log sequence 19
Next log sequence to archive 21
Current log sequence 21

9. Run the following command to exit the database:

SQL> exit

4.3.3 Troubleshooting a Synchronization Failure

Fault Symptom
An Oracle 10g database is used on site. The synchronization module cannot synchronize
data properly, and the data in the ebase table is not updated when the data in the Oracle
database is updated.
l The log file prints logs properly.

06-21 22:10:25:036 F[xsync_o2x.c] L[3031] CAPTURE 0 FOUND QUEUE EMPTY

06-21 22:10:25:038 F[xsync_o2x.c] L[3031] CAPTURE 4 FOUND QUEUE EMPTY
06-21 22:10:25:038 F[xsync_o2x.c] L[3031] CAPTURE 2 FOUND QUEUE EMPTY
06-21 22:10:26:035 F[xsync_o2x.c] L[3031] CAPTURE 5 FOUND QUEUE EMPTY
06-21 22:10:26:035 F[xsync_o2x.c] L[3031] CAPTURE 3 FOUND QUEUE EMPTY
06-21 22:10:26:036 F[xsync_o2x.c] L[3031] CAPTURE 1 FOUND QUEUE EMPTY

l The configuration file is normal.

l The scripts operate properly.

Fault Cause
The Oracle database's log file alert_zxin.log contains the following logs:

4-11

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

ORA-1688: unable to extend table SYSTEM.LOGMNR_OBJ$ partition P132 by 128

in tablespace SYSAUX
Wed Jun 22 23:16:04 2011
krvxerpt: Errors detected in process 90, role builder.
Wed Jun 22 23:16:04 2011
krvxmrs: Leaving by exception: 604
Wed Jun 22 23:16:04 2011
Errors in file /home/oracle/admin/zxin/bdump/zxin_p016_18555.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01688: unable to extend table SYSTEM.LOGMNR_OBJ$ partition P132 by 128 in tablespace
SYSAUX
Wed Jun 22 23:16:04 2011
LogMiner process death detected
Wed Jun 22 23:16:04 2011
ORA-1652: unable to extend temp segment by 8 in tablespace SYSAUX
Wed Jun 22 23:16:06 2011
Streams CAPTURE C002 with pid=18, OS id=27363 stopped
Wed Jun 22 23:16:06 2011
Errors in file /home/oracle/admin/zxin/bdump/zxin_c002_27363.trc:
ORA-01280: Fatal LogMiner Error.
Wed Jun 22 23:16:06 2011
Streams CAPTURE C005 with pid=163, OS id=27369 stopped
Wed Jun 22 23:16:06 2011
Errors in file /home/oracle/admin/zxin/bdump/zxin_c005_27369.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01658: unable to create INITIAL extent for segment in tablespace SYSAUX
ORA-06512: at "SYS.LOGMNR_KRVRDLUID3", line 1799
ORA-06512: at line 1

The log messages show that the Oracle database's system tablespace SYSAUX is full,
which means that the core functions of the database are still effective, but the functions of
the SYSAUX tablespace fail or are restricted.

Recommended Procedure
Expand the SYSAUX system tablespace.

Note:

For information about Oracle commands, refer to the related Oracle documents. Contact
ZTE technical support if necessary.

4-12

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

l Use the ALTER TABLESPACE ADD DATAFILE; command to increase the system
file or original partition.
l Increase the initial size, for example, by using the alter tablespace SYSAUX default
storage(next 500M pctincrease 1); command.

4.3.4 Troubleshooting Message Accumulation for a Synchronization

Module
Fault Symptom
Messages are accumulated for the xsync_x2sx synchronization module. After the
enable_flow_control parameter in the xsync_x2sx.ini file is set to 1 to enable flow
control and the message threshold is set to the maximum, messages are still accumulated.
The redolog file contains only .r files and no .o file.
The following errors are reported by the synchronization module:

06-28 03:42:48:002 F[xsync_x2sx.c] L[2585] parse_file_entry for xbase[143]:redo_log_get_next_lcr

reported end of file[/home/xbase/data/xbase143/redolog/XLOG_172.17.245.143_20110627151146
_0000002920.r],switch to next file.
06-28 03:45:07:002 F[xsync_x2sx.c] L[2585] parse_file_entry for xbase[143]:redo_log_get_next_lcr
reported end of file[/home/xbase/data/xbase143/redolog/XLOG_172.17.245.143_20110627153607
_0000002921.r],switch to next file.
06-28 03:45:57:002 F[xsync_x2sx.c] L[2585] parse_file_entry for xbase[143]:redo_log_get_next_lcr
reported end of file[/home/xbase/data/xbase143/redolog/XLOG_172.17.245.143_20110623012541
_0000002725.r],switch to next file.
06-28 03:45:59:002 F[xsync_x2sx.c] L[2585] parse_file_entry for xbase[143]:redo_log_get_next_lcr
reported end of file[/home/xbase/data/xbase143/redolog/XLOG_172.17.245.143_20110623042554
_0000002726.r],switch to next file.

Fault Cause
According to the printed logs, messages are parsed repeatedly for 20110627 and
20110623. The repeated parsing causes the accumulated messages.

None of the .r files are changed to .o files in the redolog directory. It is because the
synchronization module and the source ebase server are installed by different users, and
the synchronization module does not have the permission to modify the files in the redo
log directory on the source ebase server, and thus the failed rename operation.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase.
2. Run the following command to modify the permissions to the redolog directory on
the source ebase server:

$ chmod -R 777 redolog

4-13

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

4.4 Common xbasemoni Faults

The xbasemoni process is used to test the ebase's operation status. This section describes
two xbasemoni faults based on log information.

4.4.1 Fault One

Fault Symptom
The following logs are printed in the [Link] file for the xbasemoni process:

F[xbasemoni.c] L[624] connect to xbase error

F[xbasemoni.c] L[685] xbase status abnormal,1
F[xbasemoni.c] L[624] connect to xbase error
F[xbasemoni.c] L[685] xbase status abnormal,2
F[xbasemoni.c] L[624] connect to xbase error
F[xbasemoni.c] L[685] xbase status abnormal,3
F[xbasemoni.c] L[568] restart xbase server!

Fault Cause
l The ebase is not started, and the xbasemoni process keeps checking the ebase's
status. If the ebase is still not started after three checks, the xbasemoni process
restarts the ebase.
l The IP address and port number configured in the [Link] file are not
consistent with the IP address and port number configured in the [Link] file.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase.
2. Run the following command to check whether the ebase process is started:
$ xstat
l Yes: proceed to step 4.
l No: proceed to step 3.
3. Check the [Link] file, troubleshoot according to the log information, and
then check whether the problem is resolved:
l Yes: end.
l No: proceed to step 6.
4. Check whether the IP addresses and port numbers are consistent in the xbasemoni
.cfg and [Link] files:
l Yes: proceed to step 6.
l No: proceed to step 5.
5. Modify the IP addresses or port numbers to keep them consistent, and then restart the
ebase to check whether the problem is resolved:
l Yes: end.
l No: proceed to step 6.

4-14

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

6. Contact ZTE technical support.

4.4.2 Fault Two

Fault Symptom
The following logs are printed in the [Link] file for the xbasemoni process:

F[xbasemoni.c] L[2297] connect to xbase[[Link]:8433] error

F[xbasemoni.c] L[2360] xbase status abnormal,1
F[xbasemoni.c] L[2297] connect to xbase[[Link]:8433] error
F[xbasemoni.c] L[2360] xbase status abnormal,2

Fault Cause
The logs indicate that the xbasemoni process is not connected to the ebase, and the IP
address or port number that the xbasemoni process monitors is not correct.

Recommended Procedure
1. Log in to the ebase server as the user who installed the ebase.
2. Check whether the IP addresses and port numbers in the [Link] and xba
[Link] files are consistent:
l Yes: proceed to step 4.
l No: proceed to step 3.
3. Modify the IP addresses or port numbers to keep them consistent, and then restart the
ebase to check whether the problem is resolved:
l Yes: end.
l No: proceed to step 4.
4. Contact ZTE technical support.

4.5 OMM Fault

Fault Symptom
The network traffic and disk traffic are displayed as 0 in performance statistics.

Fault Cause
The iostat is not installed on the device, or the environment variable is not configured for
the iostat. As a result, I/O traffic statistics cannot be collected.

Recommended Procedure
1. Log in to the ebase server as the root user.
2. Check whether the iostat is installed.

4-15

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Run the iostat command. If the iostat is not installed, the command fails, and a
message is displayed.
l Yes: proceed to step 4.
l No: proceed to step 3.
3. Install the iostat using the system installation package and then check whether the
problem disappears:
l Yes: end.
l No: proceed to step 4.
4. Check whether the corresponding environment variable is configured for the iostat:

Note:
The environment variable for the iostat varies, depending on the specific operating
system. For the detailed information, refer to the operating system related documents.

l Yes: proceed to step 6.

l No: proceed to step 5.
5. Set the environment variable for the iostat and then check whether the problem
disappears:
l Yes: end.
l No: proceed to step 6.
6. Contact ZTE technical support.

4.6 iTool Faults

4.6.1 Maintenance Timeout
Fault Symptom
A timeout is reported by the iTool during the preventive routine maintenance.

Fault Cause
The default timeout for the iTool is 60 seconds, and sometimes a maintenance item takes
longer than 60 seconds.

Recommended Procedure
1. Enter the iTool main window.
2. Click . The Setting dialog box is displayed.
3. Modify the timeout, for example, set it to 200 seconds, and then click OK.

4-16

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Chapter 4 Common Problems

4.6.2 User Description

Fault Symptom
When a maintenance item is selected, the corresponding user who performs the
maintenance must be entered.

Fault Cause
Each maintenance package requires a corresponding user who performs the maintenance.
To execute a maintenance package, the corresponding user must log in to the device first.
Table 4-1 lists maintenance contents and the corresponding users.

Table 4-1 Maintenance Contents and Corresponding Users

User Maintenance Content

ebase user ebase database

o2x user o2x module

x2sx user x2sx module

s2x user s2x module

x2o user x2o module

agent user ebase agent

root user ebBase Unix_Linux system status

Recommended Procedure
1. In the iTool main window, select Selecting Device > Edit Device.
2. Modify the users corresponding to the maintenance packages.

4.6.3 Troubleshooting the Failed Selection of an ebase Server

When Adding a New Device
Fault Symptom
During the process of adding a new device, an ebase server cannot be selected from the
Device type list.

Fault Cause
A new device can be imported to the iTool or created manually.

If the ebase's [Link] file has not been used by the iTool, the Device type list does
not include the ebase database server and ebase-omm items when you create a new
device.

4-17

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Recommended Procedure
1. Log in to the server where the iTool resides.
2. Delete the common_zh.xml and common_en.xml files in the OMM-iToolV1.01.0
7\install\itool\workspace\default\config directory.
3. Copy the common_en_ebase.xml and common_zh_ebase.xml files in the EXTRA
BASE-OMMV1.01.04\autoupdate\build\projectinfo directory to the OMM-i
ToolV1.01.07\install\itool\workspace\default\config directory.
4. Restart the iTool, and you can select ebase database server or ebase-omm from the
Device type list.

4.6.4 Troubleshooting a Failed Import of a Maintenance Package

Fault Symptom
When a user puts a maintenance package in the checkbag directory and tries to import
it, a message is displayed to indicate that the maintenance package cannot be imported.

Fault Cause
To import a maintenance package, a user must select the checkbag directory's
higher-level directory. But the iTool does not allow the user to put the maintenance
package in the checkbag directory and import it directly.

Recommended Procedure
1. Log in to the iTool server.
2. Create the checkbag directory.
3. Put the DHSS maintenance package DHSS_CHECK in the new checkbag directory
and import it.

4-18

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix A
Frequently Used Commands
Table A-1 lists the commands that are frequently used during the routine maintenance.

Note:
For information about more ebase commands, contact ZTE technical support.

Table A-1 Frequently Used Commands

Command Function Remarks

xstart Starts the extrabase process. Used in an operating

system.

xstat Shows the status of the ebase process. Used in an operating

system.

xstop Stops the extrabase process. Used in an operating

system.
This command only stops
the extrabase process.
The shared memory still
exists.

xbasemoni Starts the xbasemoni and extrabase Used in an operating

processes. system.

xsuperstop Stops the xbasemoni and extrabase Used in an operating

processes. system.

extrabase -v Shows the version number of the Used in an operating

ebase database system. system.

xsql –v Prints the version number of the Used in an operating

[Link] library used by the system.
xsql.

show procedures; Shows the name of all the stored Used on the xsql client.
procedures.

show tables; Shows the index of all the tables. Used on the xsql client.

A-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Command Function Remarks

show session; Shows the sessions of all the Used on the xsql client.
clients. The session of each client is
represented by one record.

desc table tablename; Shows the information about the fields Used on the xsql client.
and index of a specified table. The
tablename indicates the name of the
target table,

desc procedure Shows the script used to create a Used on the xsql client.
procedurename; specified stored procedure. The
procedurename indicates the name of
the target stored procedure.

select get_used_pages(); Shows the number of used memory Used on the xsql client.
pages.
1 page=64K bytes,

select get_total_pages(); Shows the total number of memory Used on the xsql client.
pages.
The number is the same as [psm]
page_count in [Link].

if (exists('XXX')) Checks whether a table or a stored Used on the xsql client.

print 'XXX exist' procedure exists.
else
print 'XXX inexist'
go

load data outfile Exports the data of a table to a file. Used on the xsql client.
'/home/tmp/[Link]' from
table example field_term '|';

load data infile Imports data from a file to a table. Used on the xsql client.
'/home/tmp/[Link]' into
table example field_term '|';

Loadoutall Exports the data of one or multiple Used in an operating

tables in a database using the system.
specified delimiters. A character included in a
For the detailed procedure, refer to field cannot be used as a
2.1.4 Full Backup Tools. delimiter.

A-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix A Frequently Used Commands

Command Function Remarks

Loadinall Imports data to one or multiple tables Used in an operating

from the corresponding backup file(s) system.
using the specified delimiters. The delimiters used in
For the detailed procedure, refer to the loadinall command
2.1.4 Full Backup Tools. must be consistent with
the delimiters used in the
loadoutall command.

export '/home/tag/tmp/[Link]'; Exports the whole database physically. Used on the xsql client.

import '/home/tag/tmp/[Link]'; Imports the whole database physically. Used on the xsql client.

$HOME/dict/sys_dict_data.dat Data dictionary of the ebase. It is an Used in an operating

XML file that saves all the DDL (scripts system.
used to create tables, indexes, and
stored procedures).

source /home/zxin10/[Link]; Executes a specified SQL script. Used on the xsql client.

alter system switch redolog; Closes the current log file and Used on the xsql client.
generates a new one.

set show plan on/off; Shows/closes indexes used to display Used on the xsql client.
statements.

set trace on all ms; Prints all the statements whose Used on the xsql client.
set trace off all; execution time is longer than the set This command affects
milliseconds. the service operation
Stops tracing. and should be used only
when the service is idle.
The command must be
stopped immediately
when the task is
completed.

set sqltrace on; Prints the top 10 statements that take Used on the xsql client.
set sqltrace off; the longest execution time during the This command affects
last 30 minutes. the service operation
and should be used only
when the service is idle.

A-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

This page intentionally left blank.

A-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B
iTool Description
Table of Contents
Ÿ Automatic Preventive Maintenance Operations....................................................... B-1
Ÿ Maintenance Items ................................................................................................. B-8

B.1 Automatic Preventive Maintenance Operations

The iTool implements automatic preventive maintenance for ebase devices. The automatic
preventive maintenance increases the efficiency and reduces the difficulty.

Prerequisite
l The ebase's preventive routine maintenance package EBASE_CHECK is copied to the
iTool's root directory checkbag.

Note:
For iTool V1.01.08 and later, the preventive routine maintenance package is imported
to the iTool's root directory.

l The common_en_ebase.xml and common_zh_ebase.xml files in the ebase

version are copied to the iTool's OMM-iToolV1.01.06U1\install\itool\work
space\default\config directory.

Steps
1. Run [Link] to log in to the iTool.

2. Click and then configure the FTP server in accordance with the site
requirements.

B-1

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Figure B-1 Ftp Server Configure Dialog Box

3. Click and then configure the office information in accordance with the site
requirements.

Figure B-2 Add Bureau Dialog Box

4. Click in the device management window. The New device dialog box is
displayed, see Figure B-3.

B-2

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Note:
You can add a device by importing it, but it is recommended to add a device by
creating it because the ebase server commissioning scheme is complicated and
involves multiple nodes.

Figure B-3 New device Dialog Box

5. In the iTool window, click Device check.

6. Right-click the target device and then select Edit information from the shortcut menu.

B-3

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Figure B-4 iTool Window

7. Click . The Add check bag dialog box is displayed, see Figure B-5.

Figure B-5 Add check bag Dialog Box

8. Select the desired maintenance items and then click OK. The selected items are added
to the target device.

9. Modify the parameters in accordance with the site requirements.

B-4

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Note:
When you click the text box for a parameter, the parameter description is displayed at
the bottom of the Edit Device dialog box, see Figure B-6.

Figure B-6 Edit Device Dialog Box

10. On the Device check tab, right-click an object at any level to start the preventive
routine maintenance.

Note:
A level can be an office, a device, or a maintenance package.

B-5

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Figure B-7 Preventive Routine Maintenance Results

11. (Optional) Click view detail to view the detailed information about the maintenance
items.

Note:
If too many messages are returned, the messages are not displayed on the Detailed
menu. You can click view detail to view the detailed information.

B-6

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Figure B-8 Check the system errors Dialog Box

12. After the preventive routine maintenance is completed, right-click the device or the
maintenance package and then select Making report from the shortcut menu to
generate a report.

B-7

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Figure B-9 Make report Menu

– End of Steps –

B.2 Maintenance Items

Table B-1 lists the maintenance items provided by the iTool and the criteria.

Table B-1 Maintenance Items for the ebase

Maintenance
User Criteria
Item

CPU status root An alarm is raised when the sum of CPU utilization in system
mode (%sy) and user mode (%us) exceeds 80%.

Memory status root Check the memory utilization.

Swap status root Check the swap utilization. An alarm is raised when the swap
utilization exceeds 10%.

Disk partition root Check the status of the disk partitions. An alarm is raised when
status the utilization of any partition exceeds 90%.

System logs root Check whether the system logs contain the ERROR keyword. If
ERROR is found, it indicates that an error has occurred, and the
latest 1000 error logs have been printed.

ebase version ebase user Check the ebase version.

ebase bit number ebase user Check the ebase program's bit number.

B-8

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Maintenance
User Criteria
Item

ebase operation ebase user Check the CPU utilization of the extrabase process. An alarm is
environment raised when the utilization exceeds 80%.

ebase user Check the memory utilization of the extrabase process. An alarm
is raised when the utilization exceeds 80%.

ebase user Partition where the data directory resides. An alarm is raised
when the partition size is less than 5 times of the file size.
Partition where the Redolog directory resides. An alarm is raised
when the partition size is less than 5 times of the file size.

ebase user Check the memory page usage. An alarm is raised when the
utilization of the total page count, used page count, or shared
memory page count exceeds 90%.

ebase ebase user 1. If the xsync_x2o or xsync_x2sx process exists, check the
configuration setting for the redo_log_used_by_xsync parameter in
[Link] file. An alarm is raised if this parameter is not
set to 1.
2. Check the value of the mem_size parameter. An alarm is
raised if the value is greater than 1024.
3. Check the value of the flush_all_pages_before_shutdown
parameter. An alarm is raised if this parameter is not set to 0.
4. Check the value of the max_n_session parameter. An
alarm is raised if the value is greater than 200.
5. Check the value of the page_count parameter. An alarm is
raised if the value exceeds the 50% of the total memory size.
6. If integrality_level=1, redo_log_buffer_size must be set to 0,
otherwise an alarm is raised.
If integrality_level=3, redo_log_buffer_size must be less
than 10 M, otherwise an alarm is raised.

xbasemoni ebase user 1. Check the settings for the XbaseSrvIP and XbaseSrvPort
configuration parameters in the [Link] file and the settings for
the bind_ip and bind_port parameters in the [Link]
file. The setting for XbaseSrvIP must be consistent with the
setting for bind_ip, and the setting for XbaseSrvPort must
be consistent with the setting for bind_port. An alarm is
raised if any of the settings is inconsistent.
2. Check whether the work_mode parameter in the
[Link] file is set to 0. An alarm is raised if this
parameter is not set to 0.

B-9

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Maintenance
User Criteria
Item

Scheduled backup ebase user Check whether the output from the crontab -l command contains
task ebase_export.sh. If ebase_export.sh is found, it indicates
that a scheduled backup task has been set for the ebase.
Otherwise, an alarm is raised.

ebase operation ebase user 1. Check whether the [Link] file contains error,
status fail, or hung. If any of these keywords is found, it indicates
that an error or a failure has occurred, and the latest 1000
error logs have been printed.
2. Check whether the [Link] file contains abnormal.
If the file contains abnormal, it indicates that an alarm
has been raised, and the latest 1000 error logs have been
printed.

core file ebase user Check whether the extrabase process has generated a core
file. If a core file has been generated, it indicates that an alarm
has been raised, and the alarm shows the path and name of
the core file.

xbase_agent ebaseagent Run the ps -ef|grep agtfunc command to check whether

user xbase_agent exists. An alarm is raised if xbase_agent does
not exist.

core file ebaseagent Check whether agtfunc has generated a core file. If a core file
user has been generated, it indicates that an alarm has been raised,
and the alarm shows the path and name of the core file.

xsync_o2x log file o2x user Check whether the xsync_o2x.log file contains error or fail.
If any of these keywords is found, it indicates that an alarm has
been raised, and the latest 1000 error logs have been printed.

xsync_o2x.ini o2x user 1. Check whether the trace_msg or trace_msg2 parameter

configuration is set to 1. During peak hours, either of the parameters must
for the o2x not be set to 1, because when the tracing function is enabled,
synchronization a large amount of logs are generated, and the service
module performance is affected. During off-peak hours, the tracing
function can be enabled to observe the service operation.
2. Run the ostool -r [Link] command to check whether
the logprintlevel parameter is set for xsync_o2x. If the
parameter is set, the set value must not be greater than or
equal to 4. It is allowed that the parameter is not set or the
set value is less than 4.
3. In the case of a dual-server cluster system, read the
configuration of the cache_dir directory, and use the df -k
command to check whether the command output contain the
configuration information about the cache_dir directory. If

B-10

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Maintenance
User Criteria
Item
the returned results contain the configuration information,
it indicates that the directory is configured on the shared
storage, which is correct.
4. The configuration file for the xsync_o2x synchronization
module contains the [xbase1]-[xbasen] segment, and
each [xbase] in the segment contains the xbase_module
parameter. Ensure that the xbase_module parameter for
each [xbase] is set to a unique value in the [xbase1]-[xbasen]
segment.
5. Check whether you can connect to each xbase through
the xsql client. In the configuration file for the xsync_o2x
synchronization module, read the settings for the xbase_ip
and xbase_port parameters in each [xbase] segment, and
then run the xsql -s xbase_ip -p xbase_port command, for
example, xsql -s [Link] -p 8433. If 0 is returned, it indicates
that a link is established successfully.

core file o2x user Check whether the xsync_o2x synchronization module has
generated a core file. If a core file has been generated, it
indicates that an alarm has been raised, and the alarm shows
the path and the name of the core file.

xsync_x2sx log file x2sx user Check whether the xsync_x2sx.log file contains error or fail.
If any of the keywords is found, it indicates that an error or a
failure has occurred, and the latest 1000 error logs have been
printed.

xsync_x2sx.ini x2sx user 1. Check whether the trace_msg or trace_msg2 parameter is

configuration set to enable the tracing function. When the tracing function
for the x2sx is enabled, a large amount of logs are generated, and the
synchronization service performance is affected. During off-peak hours,
module the tracing function can be enabled to observe the service
operation.
2. Run the ostool -r [Link] command to check whether
the logprintlevel parameter is set for xsync_x2sx. If the
parameter is set, the set value must not be greater than or
equal to 4. It is allowed that the parameter is not set or the
set value is less than 4.
3. In the case of a dual-server cluster system, read the
configuration of the cache_dir and pos_file_dir
directories, and use the df -k command to check whether
the command output contain the configuration information
about the two directories. If the returned results contain the

B-11

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Maintenance
User Criteria
Item

configuration information, it indicates that the directories are

configured on the shared storage, which is correct.
4. The configuration file for the xsync_x2sx synchronization
module contains the [xbase1]-[xbasen] segment, and
each [xbase] in the segment contains the xbase_module
parameter. Ensure that the xbase_module parameter for
each [xbase] is set to a unique value in the [xbase1]-[xbasen]
segment.
5. Check whether you can connect to each xbase through
the xsql client. In the configuration file for the xsync_x2sx
synchronization module, read the settings for the xbase_ip
and xbase_port parameters in each [xbase] segment, and
then run the xsql -s xbase_ip -p xbase_port command, for
example, xsql -s [Link] -p 8433. If 0 is returned, it indicates
that a link is established successfully.
6. The configuration file for the xsync_x2sx synchronization
module contains the [dest_db1]-[dest_dbn] segment, and
each [dest_db] in the segment contains the db_module
parameter. Ensure that the db_module parameter for each
[dest_db] is set to a unique value in the [dest_db1]-[dest_dbn]
segment.
7. Check whether you can connect to each xbase through
the xsql client. In the configuration file for the xsync_x2sx
synchronization module, read the setting for the db_type
parameter in each [dest_db] segment. If the parameter is
set to 1, it indicates that synchronization is configured for the
ebase. In this case, run the xsql -s xbase_ip -p xbase_port
command, for example, xsql -s [Link] -p 8433. If 0 is
returned, it indicates that a link is established successfully.

core file x2sx user Check whether the xsync_x2sx synchronization module has
generated a core file. If a core file has been generated, it
indicates that an alarm has been raised, and the alarm shows
the path and the name of the core file.

xsync_s2x log file s2x user Check whether the xsync_s2x.log file contains error or fail. If
any of the keywords is found, it indicates that an error or a failure
has occurred, and the latest 1000 error logs have been printed.

B-12

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

 Appendix B iTool Description

Maintenance
User Criteria
Item

xsync_s2x.ini s2x user 1. Check whether the trace_msg parameter is set to 1. During
configuration peak hours, the parameter must not be set to 1, because
for the s2x when the tracing function is enabled, a large amount of logs
synchronization are generated, and the service performance is affected.
module During off-peak hours, the tracing function can be enabled to
observe the service operation.
2. Run the ostool -r [Link] command to check whether
the logprintlevel parameter is set for xsync_s2x. If the
parameter is set, the set value must not be greater than or
equal to 4. It is allowed that the parameter is not set or the
set value is less than 4.
3. In the case of a dual-server cluster system, read the
configuration of the cache_dir directory, and use the df -k
command to check whether the command output contain the
configuration information about the cache_dir directory. If
the returned results contain the configuration information,
it indicates that the directory is configured on the shared
storage, which is correct.
4. The configuration file for the xsync_s2x synchronization
module contains the [xbase1]-[xbasen] segment, and
each [xbase] in the segment contains the xbase_module
parameter. Ensure that the xbase_module parameter for
each [xbase] is set to a unique value in the [xbase1]-[xbasen]
segment.
5. Check whether you can connect to each xbase through
the xsql client. In the configuration file for the xsync_s2x
synchronization module, read the settings for the xbase_ip
and xbase_port parameters in each [xbase] segment, and
then run the xsql -s xbase_ip -p xbase_port command, for
example, xsql -s [Link] -p 8433. If 0 is returned, it indicates
that a link is established successfully.

core file s2x user Check whether the xsync_s2x synchronization module has
generated a core file. If a core file has been generated, it
indicates that an alarm has been raised, and the alarm shows
the path and the name of the core file.

xsync_x2o log file x2o user Check whether the xsync_x2o.log file contains error or fail. If
any of the keywords is found, it indicates that an error or a failure
has occurred, and the latest 1000 error logs have been printed.

B-13

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

ebase Maintenance Guide

Maintenance
User Criteria
Item

xsync_x2o.ini x2o user 1. Check whether the trace_msg or trace_msg2 parameter is

configuration set to enable the tracing function. When the tracing function
for the x2o is enabled, a large amount of logs are generated, and the
synchronization service performance is affected. During off-peak hours,
module the tracing function can be enabled to observe the service
operation.
2. Run the ostool -r [Link] command to check whether
the logprintlevel parameter is set for xsync_x2o. If the
parameter is set, the set value must not be greater than or
equal to 4. It is allowed that the parameter is not set or the
set value is less than 4.
3. The configuration file for the xsync_x2o synchronization
module contains the [xbase1]-[xbasen] segment, and
each [xbase] in the segment contains the xbase_module
parameter. Ensure that the xbase_module parameter for
each [xbase] is set to a unique value in the [xbase1]-[xbasen]
segment.
4. Check whether you can connect to each xbase through
the xsql client. In the configuration file for the xsync_x2o
synchronization module, read the settings for the xbase_ip
and xbase_port parameters in each [xbase] segment, and
then run the xsql -s xbase_ip -p xbase_port command, for
example, xsql -s [Link] -p 8433. If 0 is returned, it indicates
that a link is established successfully.

core file x2o user Check whether the xsync_x2o synchronization module has
generated a core file. If a core file has been generated, it
indicates that an alarm has been raised, and the alarm shows
the path and the name of the core file.

B-14

SJ-20120313112131-024|2012-11-28 (R1.0) ZTE Proprietary and Confidential

Glossary
API
- Application Program Interface
CPU
- Central Processing Unit
DML
- Data Manipulation Language
JDBC
- Java Database Connectivity
ODBC
- Open DataBase Connectivity
OMM
- Operation and Maintenance Management
OMMP
- Operation and Maintenance Management Platform
OS
- Operating System

RDBMS
- Relational Data Base Management System

SQL
- Structured Query Language