| 1 | Introduction |
| 1.1 | Prerequisites |
2 | Procedure |
| 2.1 | Clean Up IMM Status |
| 2.2 | Restore System Data Backup |
| 2.3 | Restore User Data Backup |
1 Introduction
This document describes how to restore a backup.
1.1 Prerequisites
This section describes the prerequisites, which must be fulfilled before using the procedure.
1.1.1 Conditions
The following conditions must apply:
- No other backup operation is in progress.
- The name of the backup to restore is known.
- An Ericsson Command-Line Interface (ECLI) session in Exec mode is in progress.
- For the User Data restore operation that involves MySQL data:
2 Procedure
2.1 Clean Up IMM Status
Before restoring a System or User Data backup, make sure that the Information Model Management (IMM) status for DHCP or AAA is cleaned up. The cleanup operation must be performed on all PL nodes.
- Log on to SC node, and then connect to PL node.
# ssh <User Name>@<MIP_OAM_IP>
SC-X:~# ssh <PL hostname>
- Check the status of IPWorks services.
PL-X:~# ipw-ctr status all
- If the DHCP service is NOT running, clean up the DHCP
related IMM status.
PL-X:~# /opt/ipworks/dhcp/scripts/ipworks.dhcpv4 cleanup
- If the AAA service is NOT running, clean up the AAA related
IMM status.
PL-X:~# /opt/ipworks/aaa_radius/csvengine/script/ipworks.aaa_csv_engine cleanup
PL-X:~# /opt/ipworks/aaa_radius/backend/script/ipworks.aaa_backend cleanup
PL-X:~# /opt/ipworks/aaa_radius/stack/script/ipworks.aaa_radius_stack cleanup
PL-X:~# /opt/ipworks/aaa_diameter/scripts/ipworks.aaa cleanup
2.2 Restore System Data Backup
To restore a System Data backup, do the following:
- Navigate to the BrmBackupManager managed object, for example:
>dn ManagedElement=<Node Name>,SystemFunctions=1,BrM=1,BrmBackupManager=SYSTEM_DATA
- Specify the backup to restore:
(BrmBackupManager=SYSTEM_DATA)>BrmBackup=system_20150616
- Start the restore operation:
- Note:
- Be careful to perform the restore operation. Because the restore operation cannot be cancelled after it starts.
(BrmBackup=system_20150616)>restore
The system returns output true or false.
Attention!A cluster reboot is automatically triggered when restoring a System Data backup. The resulting in-service performance impact corresponds to the time required for a cluster to restart after reboot.
- Verify that the backup was successfully restored, for
example:
(BrmBackup=system_20150616)>show -v
BrmBackup=system_20150616 backupDescriptor="version: 1.2\nbackupName: system_20150616\nbackupType: BRM_SYSTEM_DATA\ncreationType: MANUAL\ncreationTime: 2011-02-04 09:17:51\nstorageOwner: LDE_BRF_SYSTEM\nswVersion: ERIC-COREMW_RUNTIME-CXP9020355_2-R1A16|MW|2011-02-04T09:00:40|R1A16|CXP9020355_2|ERIC-COREMW_RUNTIME|ERIC-COREMW_RUNTIME\n" <read-only> backupName="system_20150616" <read-only> brmBackupId="system_20150616" creationTime="2015-06-16T09:17:51" <read-only> creationType=MANUAL <read-only> status=BRM_BACKUP_COMPLETE <read-only> asyncActionProgress=[] <empty> <read-only> <deprecated> progressReport <read-only> actionId=3 <read-only> actionName="RESTORE" <read-only> additionalInfo=[] <empty> <read-only> progressInfo="PERMIT_PHASE is completed" <read-only> progressPercentage=33 <read-only> result=NOT_AVAILABLE <read-only> resultInfo="" <read-only> state=RUNNING <read-only> timeActionCompleted="1970-01-01T13:00:00" <read-only> timeActionStarted="2015-06-16T10:35:52" <read-only> timeOfLastStatusUpdate="2015-06-16T10:35:56" <read-only> swVersion[@1] <read-only> description="ERIC-COREMW_RUNTIME" <read-only> productionDate="2015-05-15T04:53:53" <read-only> productName="ERIC-COREMW_RUNTIME" <read-only> productNumber="CXP9020355_2" <read-only> productRevision="R1A16" <read-only> type="MW" <read-only>- Note:
- When performing a System Data backup restore, the restore progress can be monitored (refer to View Progress Report) until the system reboot is triggered. After the system is rebooted, the progress report is reset to default values.
Proceed with the next step after the system is rebooted successfully.
The system is going down for reboot NOW! Remote host closes the connection to 192.168.10.11.
- Create the link and folder for ss7caf on SC nodes manually.
Execute the following commands on both SC-1 and SC-2:
# ln -s /storage/system/config/ss7caf-ana90137/etc /opt/sign/etc
# ln -s /storage/no-backup/ss7caf-ana90137/log /opt/sign/log
# mkdir -p /opt/sign/cnf_template
- Replace the ipworks_sql.conf file.
Execute the following command on SC-1 and SC-2.
# cp /opt/ipworks/ss/db/ipworks_sql.conf /opt/ipworks/mysql/db/ipworks_sql.conf
- If the backup you restored contains the IPWorks software,
do the following:
- Start MySQL NDB cluster manually on SC-1 or SC-2.
# /etc/init.d/ipworks.mysql start-ndbcluster
- Perform health check.
For information on how to check service status, refer to IPWorks Manual Health Check.
If some of the services cannot work normally, you can choose a right user data to restore (see Section 2.3), then IPWorks will be back to a health status.
- Start MySQL NDB cluster manually on SC-1 or SC-2.
2.3 Restore User Data Backup
To restore a User Data backup:
- For the user data restore that includes MySQL data, additional
steps are required:
- Check drbd role on both SC nodes, make sure that SC-1
is the primary drbd.
SC-1:~ # lde-info -t drbd
Primary
SC-2:~ # lde-info -t drbd
Secondary
- If SC-2 is the primary drbd, reboot SC-2.
SC-2:~ # reboot
After SC-2 reboot, check whether SC-1 changes to the primary drbd.
- Check whether MySQL is working normally on SC node.
SC-X:~ # /usr/local/mysql/bin/mysql -P 3307 --protocol=tcp -hipw_sql
For example:
Welcome to the MySQL monitor. Commands end with ; or \g. Your MySQL connection id is 191 Server version: 5.6.31-ndb-7.4.12-cluster-commercial-advanced MySQL Cluster Server - Advanced Edition (Commercial) Copyright (c) 2000, 2016, Oracle and/or its affiliates. All rights reserved. Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners. Type 'help;' or '\h' for help. Type '\c' to clear the current input statement. mysql>
- Make sure the following file size is NOT 0 on SC nodes.
/cluster/ipworks/mysql-cluster/sqlnode/mysql/user.MYD
If the file size is already 0, consult the next level of maintenance support.
- Check drbd role on both SC nodes, make sure that SC-1
is the primary drbd.
- Navigate to the BrmBackupManager managed object, for example:
>dn ManagedElement=<Node Name>,SystemFunctions=1,BrM=1,BrmBackupManager=USER_DATA
- Specify the backup to restore:
- Restore a backup that does not include MySQL data, for
example:
(BrmBackupManager=USER_DATA)>BrmBackup=testBackup_20150616
- Restore a backup that includes MySQL data, for example:
(BrmBackupManager=USER_DATA)>BrmBackup=ndb_testBackup_20150616
- Restore a backup that does not include MySQL data, for
example:
- Start the restore operation:
- Note:
- Be careful to perform the restore operation. Because the restore operation cannot be cancelled after it starts.
(BrmBackup=testBackup_20150616)>restore
The system returns output true or false.
- Verify that the backup was successfully restored, for
example:
(BrmBackup=testBackup_20150616)>show -v
The following is an example output:
BrmBackup=testBackup_20150616 backupDescriptor="version: 1.2 backupName: testBackup_20150616 backupType: BRM_USER_DATA creationType: MANUAL creationTime: 2015-06-16 06:47:35 storageOwner: IPWorks_BRF swVersion: ERIC-COREMW_RUNTIME-CXP9020355_1-R8C01|MW|2015-05-15T04:53:53|R8C01|CXP9020355_1|ERIC-COREMW_RUNTIME|ERIC-COREMW_RUNTIME " <read-only> backupName="testBackup_20150616" <read-only> brmBackupId="testBackup_20150616" creationTime="2015-06-16T06:47:35" <read-only> creationType=MANUAL <read-only> status=BRM_BACKUP_COMPLETE <read-only> asyncActionProgress <read-only> actionId=3 <read-only> actionName="RESTORE" <read-only> additionalInfo=[] <empty> progressInfo="PERMIT_PHASE is completed" <read-only> progressPercentage=33 <read-only> result=NOT_AVAILABLE <read-only> resultInfo="" <read-only> state=RUNNING <read-only> timeActionCompleted="1970-01-01T06:00:00" <read-only> timeActionStarted="2015-06-16T07:35:52" <read-only> timeOfLastStatusUpdate="2015-06-16T07:35:56" <read-only> progressReport <read-only> actionId=3 <read-only> actionName="RESTORE" <read-only> additionalInfo=[] <empty> progressInfo="PERMIT_PHASE is completed" <read-only> progressPercentage=33 <read-only> result=NOT_AVAILABLE <read-only> resultInfo="" <read-only> state=RUNNING <read-only> timeActionCompleted="1970-01-01T06:00:00" <read-only> timeActionStarted="2015-06-16T07:35:52" <read-only> timeOfLastStatusUpdate="2015-06-16T07:35:56" <read-only> swVersion[@1] description="ERIC-COREMW_RUNTIME" <read-only> productionDate="2015-05-15T04:53:53" <read-only> productName="ERIC-COREMW_RUNTIME" <read-only> productNumber="CXP9020355_1" <read-only> productRevision="R8C01" <read-only> type="MW" <read-only>- Note:
- If the restore failed, try to start MySQL NDB cluster by executing /etc/init.d/ipworks.mysql start-ndbcluster, and then perform the User Data restore again. If the issue still remains, consult the next level support.
- Make the backup take effect.
- Note:
- Check the ENUM/DNS/SS/EPC AAA server:
- During the restore process, the used server will be stopped. After restore, the server will be started.
- But if the server is not used before the restore, the server is still stopped. Start them manually after restore if needed.
For DNS:
To make the DNS record work, update the dnsserver in IPWorks CLI, for example:
SC-1:~ # ipwcli IPWorks>Login: <username> IPWorks> Password:******** Login to server successful. The last successful login time of this user is 2015-06-16 07:45:07 IPWorks> list dnsserver [DnsServer dns1] Partition: active Name: dns1 Address: 169.254.43.15 PrimaryAddress: 169.254.43.15 DnsName: dns1.test.com PrimaryDnsName: dns1.test.com Filename: named.conf AlgServerType: false Status: On 06/16/15 at 07:44:42 server is 'running' ExportNeeded: false IPWorks> update dnsserver dns1 -rebuild true Result of performing an export is: Exported the zone [MasterZone test.com] Exported configuration for [DnsServer dns1] Updated the configuration for 'DNS' server 'dns1'.
For ENUM:
When ENUM uses the Number Portability (NP) function, to make the backup take effect, restart the SS7 stack manually on an SC. For example:
amf-adm lock safSu=PL-3,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm lock-in safSu=PL-3,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm unlock-in safSu=PL-3,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm unlock safSu=PL-3,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm lock safSu=PL-4,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm lock-in safSu=PL-4,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm unlock-in safSu=PL-4,safSg=NWA,safApp=ERIC-ss7caf.core amf-adm unlock safSu=PL-4,safSg=NWA,safApp=ERIC-ss7caf.core
To make the backup take effect, restart the SS7 stack manually as the same as the steps for ENUM.
For Geographic Redundancy:
After restoring the user data backup, the user data MUST be synchronized between two sites before implementing geographic redundancy function. For more information on who to synchronize data between two sites, refer to the section MySQL Replication for Geographic Redundancy Failed On All Sites in IPWorks Troubleshooting Guideline.

Contents