Home » Dell Manuals » Servers » Dell DX6004S » Manual Viewer

Dell DX6004S DX Object Storage Administration Guide - Page 61

C.3. Back-out Steps

Add to My Manuals
Save this manual to your list of manuals

Page 61 highlights

a single node before starting the others in order to verify that the software runs and that it can communicate with the network. Verify that the first node boots properly then power up the other nodes in the cluster. When a node is first booted, there is a default window of 15 minutes, controlled by the hpStartDelay parameter, before it will begin attempting to check for the absence of other nodes in the cluster. As long as all nodes in the cluster are running within this window, the rapid recovery operations will be avoided. This window only exists immediately following the restart of a node. Any time after this, the recovery operations will commence immediately following the detection of a missing node. C.3. Back-out Steps If it becomes necessary to regress the software version on any node, the software can be reverted to an old version using the same method that was utilized to perform the upgrade. Copyright © 2010 Caringo, Inc. All rights reserved 56 Version 5.0 December 2010

Section	Page
DX Object Storage Administration Guide	1
Table of Contents	3
Chapter 1. Welcome to DX Storage	6
1.1. Overview of DX Storage	6
1.2. Components	6
1.3. About this Document	6
1.3.1. Audience	6
1.3.2. Scope	6
Chapter 2. Introduction to the Admin Console	7
2.1. Accessing the Admin Console	7
2.2. Initial View of the Admin Console	7
2.2.1. Viewing the Cluster Status Page	7
2.2.2. Viewing a Node's Status Page	8
2.3. Printing the Admin Console	8
2.4. Viewing License Information	8
Chapter 3. Managing the Cluster	10
3.1. Viewing the Cluster Status Page	10
3.2. Authenticating Cluster-Wide Actions	11
3.3. Shutting Down or Restarting the Cluster	12
3.4. Viewing Nodes in the Cluster	12
3.5. Searching for Nodes By IP Address	13
3.6. Searching for Nodes by Status	13
3.7. Choosing and Preserving Cluster Settings	13
3.7.1. Enabling Logging	15
3.7.2. Replication	15
3.7.3. Suspend	15
3.7.4. Power	15
3.7.5. Managing Tenants	15
3.8. Cluster Name	15
3.9. Cluster Multicast Address	15
3.10. % Used Indicator	15
Chapter 4. Managing Tenants	17
4.1. Terminology Related to Tenant Security	18
4.2. About the Default Cluster Domain	19
4.3. Security Privileges for Administrative Operations	20
4.4. Rules and Recommendations for Managing Tenants	21
4.5. Domain Naming Rules	21
4.6. Adding, Editing, or Deleting Tenants	22
4.7. Other Cluster Administrator Tasks	24
4.7.1. Using Administrative Override	25
4.7.1.1. Using Override to Delete an Object	25
4.7.1.2. Using Override to GET or APPEND User Lists	25
4.7.1.3. Using Override to Resolve Authorization Specification Issues	27
4.7.2. Working With Inaccessible Objects	28
Chapter 5. Managing Nodes	30
5.1. Viewing the Node Status Page	30
5.1.1. Shutting Down or Restarting a Node	30
5.1.2. Retiring or Identifying a Node	30
5.1.2.1. Retiring a Node or Volume	30
5.1.2.2. Identifying a Volume	31
5.1.3. Errors and Announcements	31
5.1.4. Additional Node Status Information	31
5.1.4.1. Hardware Status Reporting	33
5.2. Displaying Subcluster Information	34
Chapter 6. Configuring the Node	35
6.1. Option Names and Descriptions	35
6.2. Managing DX Storage Administrators and Users	41
6.2.1. Defining CAStor administrators and SNMP Administrators	41
6.2.2. Defining DX Storage Operators	42
6.2.3. Securing the Administrator and Operator Passwords	42
6.3. Managing Content Integrity Settings	43
6.3.1. autoRepOnWrite	43
6.3.2. repPriority	43
6.3.3. autoValidateRead	43
6.4. Managing Other Stream Replication Settings	44
6.4.1. minreps, maxreps, and defreps	44
6.4.2. hpStartDelay	44
6.5. Managing Volumes	44
6.5.1. device	45
6.5.2. policy	46
6.5.3. Specifying Exceptions	46
6.6. Configuring Power Management Settings	46
6.6.1. sleepAfter	46
6.6.2. wakeAfter	46
6.7. Managing Other Settings	46
6.7.1. consolePort	46
6.7.2. domainHeaders	46
6.7.3. loghost	47
6.7.4. timeSource	47
Chapter 7. Managing Volumes	49
7.1. Volume Expiration	49
7.2. Movement Between Nodes	49
7.3. Physical Errors	49
Appendix A. Implementation of Multi-Server Chassis	51
A.1. Configuration Parameters	51
A.1.1. Processes parameter	51
A.1.2. Network Setup parameters	51
A.1.3. Using the vols parameter	52
A.1.4. Using the subcluster parameter	52
A.2. Monitoring and Administration	53
Appendix B. Using SNMP with DX Storage	54
B.1. SNMP Management Information Base (MIB) Reference	54
B.2. Managing DX Storage Nodes	54
B.2.1. Shutdown Action for Nodes	54
B.2.2. Retire Action for Nodes and Volumes	54
B.2.2.1. Single Volumes	55
B.2.2.2. Entire Node	55
B.3. SNMP Tools and Monitoring Systems	55
B.3.1. Open Source Tools	55
B.3.2. SNMP Examples with DX Storage	56
B.3.3. SNMP Action OIDs	56
B.3.3.1. castorShutdownAction	56
B.3.3.2. castorRetireAction	56
B.3.3.3. castorLogLevelAction	56
B.3.3.4. castorSyslogHostAction	56
B.3.3.5. volumeRecoverySuspend	57
B.3.4. Practical SNMP with DX Storage	57
B.3.4.1. Health Monitoring	57
B.3.4.2. Capacity Monitoring	57
B.3.4.3. Client Activity Reporting	58
Appendix C. Upgrading a License or Cluster	59
C.1. Upgrading a License File	59
C.2. Software Upgrade Overview	59
C.2.1. Preparation	59
C.2.2. Upgrade Steps	60
C.2.2.1. Example Shutdown Script	60
C.2.2.2. Cluster Reboot	60
C.3. Back-out Steps	61
Appendix D. Troubleshooting	62
D.1. Restoring Domains and Buckets	62
D.1.1. Recovering a Deleted Domain	62
D.1.2. Recovering a Deleted Bucket	65
D.2. Resolving Duplicate Domain Names in a Mirrored or Disaster Recovery (DR) Cluster	66
D.2.1. Renaming a Domain in its Source Cluster (DR Cluster Conflict Only)	66
D.2.2. Renaming a Domain in a Mirrored or DR Cluster	67
D.3. Using DX Content Router to List Buckets and Objects	69
D.4. Boot Errors	69
D.5. Configuration	70
D.6. Operational Problems	70
Appendix E. Drive Identification API	72
E.1. Overview	72
E.2. Customization Steps	72
Appendix F. Customizing the Admin Console	73

Match case Limit results 1 per page

Version 5.0

December 2010

a single node before starting the others in order to verify that the software runs and that it can

communicate with the network. Verify that the first node boots properly then power up the other

nodes in the cluster.

When a node is first booted, there is a default window of 15 minutes, controlled by the hpStartDelay

parameter, before it will begin attempting to check for the absence of other nodes in the cluster. As

long as all nodes in the cluster are running within this window, the rapid recovery operations will be

avoided. This window only exists immediately following the restart of a node. Any time after this, the

recovery operations will commence immediately following the detection of a missing node.

C.3. Back-out Steps

If it becomes necessary to regress the software version on any node, the software can be reverted

to an old version using the same method that was utilized to perform the upgrade.