HP Archive Backup System for OpenVMS Guide to Operations
HP Archive Backup System for OpenVMS Guide to Operations HP Archive Backup System for OpenVMS Guide to Operations
System Backup to Tape for Oracle Databases 9.11 Troubleshooting Tips 9.11.2 Fatal Internal Error SBT-00008140 12/17/01 09:50:50 MBytes/sec: 1.752 10 SBT-00008140 12/17/01 09:50:50 Bytes/sec : 1752820.870 11 The following describes the entries in Example 9–19: 1. DB block size:-this is the size of blocks that Oracle sends to SBT. 2. I/O block size:-this the size of blocks that SBT writes to the I/O device. You can change this with the logical MDMS$SBT_IO_BLOCK_SIZE which can be specified in an Oracle Recovery Manager script or as a system wide logical. 3. Total I/Os:-total I/Os to write the I/O block size blocks to the I/O device. In this example: 620 total I/Os writing 65024 byte blocks to the I/O device. 4. Total I/O wait:-this is the number of milliseconds that SBT waited while the blocks are being written to the I/O device. 5. Maximum I/O wait:-this is the longest wait that SBT made to write a block to the I/O device. 6. Average I/O wait:-this the average wait of the total I/Os for the total I/O wait time. 7. Total Kbytes:-total Kbytes transferred to the I/O device. It is a truncated value. See total bytes below. 8. Total bytes:-total bytes transferred to the I/O device. 9. Total seconds:-this is the number of seconds that SBT waited while the blocks are being written to the I/O device. This does not have the time that Oracle took to provide the information to write to the I/O device. The value is truncated from the total I/O wait in milliseconds. 10. Mbytes/sec:-Mbytes per second transferred based on total Kbytes and total seconds. 11. Bytes/sec:-bytes per second transferred based on total bytes and total seconds. Note These values are the time that SBT waited for the I/O device to give back control. The values have nothing to do with how long it took Oracle to give information to SBT. If you should ever get an Internal error, you should report it to the customer support center. Example 9–20 shows an example of a fatal internal error. Anyplace that SBT could have received a value that it did not expect, it is captured and reported as a fatal internal error. Example 9–20 Fatal Internal Error Example SBT-00001DB2 12/17/01 13:53:45 Internal error SBT-00001DB2 12/17/01 13:53:45 Extended Status: The invalid archive type, 23 SBT-00001DB2 12/17/01 13:53:57 ABEND (abnormal end ) was passed in sbtend flag 9.11.3 Check ORA_DUMP:SBTIO.LOG for Errors Any time you receive an error in Oracle's Recovery Manager that is related to SBT, you should check ORA_DUMP:SBTIO.LOG for errors. We are limited to 250 characters returned to the Oracle Recovery Manager. Therefore, you may not receive all of the information about the error. However, I am not limited to what I put in the ORA_DUMP:SBTIO.LOG file. So be sure to look in ORA_DUMP:SBTIO.LOG when getting an error message that in not all there. 9-22 System Backup to Tape for Oracle Databases
9.11.4 Using Tape I/O Slaves System Backup to Tape for Oracle Databases 9.11 Troubleshooting Tips Example 9–21 shows an example of a fatal error reported in Oracle's Recovery Manager. Example 9–22 shows what is reported in ORA_DUMP:SBTIO.LOG for the same error. The File:, Function: and line are information for me to troubleshoot the problem if it is a software error. In this case, there were no volumes available. Look for it in the ORA_ DUMP:SBTIO.LOG file. In this case, you can look in the archive to see what the media type, pool, and location were. Example 9–21 Fatal Error in Oracle's Recovery Manager ORA-27028: skgfqcre: sbtbackup returned error ORA-19511: Error received from media manager layer, error text: Fatal media movement error Failed to allocate tape volume MDMS object: RMAN_TAPE_TL875_ARCH System Error: %MDMS-E-NOVOLUMES, no free volumes match selection criteria %MDMS-E-NOVOLUMES, no free volumes match selection criteria %MDMS-I-NOVOLSPOOL, no free volumes in the specified pool were found Example 9–22 Fatal Error in ORA_DUMP:SBTIO.LOG SBT-0000819F 12/17/01 14:10:20 Fatal media movement error SBT-0000819F 12/17/01 14:10:20 Extended Status: Failed to allocate tape volume MDMS object: RMAN_TAPE_TL875_ARCH System Error: %MDMS-E-NOVOLUMES, no free volumes match selection criteria %MDMS-E-NOVOLUMES, no free volumes match selection criteria %MDMS-I-NOVOLSPOOL, no free volumes in the specified pool were found File: WRK$ROOT:[SRC]MDMS_SBT_API_MEDIA.C;1, Function: sbt_media_allocate_volume, Line 416 Failed to allocate volume with attributes: Pool: Media Type: TLZ88M Location: 110281 When using tape I/O slaves, you may not receive the error message from SBT in Oracle's Recovery Manager. Example 9–23 shows the type of error you may get reported for a tape volume being offsite. Example 9–24 shows the tape volume offsite error reported when not using tape I/O slaves. So when troubleshooting, also look in the trace log file ORA_ DUMP:SBTIO.LOG. In both cases the error was reported in the trace log file. Example 9–23 Volume Offsite Error using Tape I/O Slave RMAN-08031: released channel: t1 RMAN-00571: =========================================================== RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS =============== RMAN-00571: =========================================================== RMAN-03006: non-retryable error occurred during execution of command: vali ate RMAN-07004: unhandled exception during command execution on channel t1 RMAN-10035: exception raised in RPC: ORA-00447: fatal error in background process RMAN-10031: ORA-19583 occurred during call to DBMS_BACKUP_RESTORE.RESTORE ACKUPP IECE RMAN> Example 9–24 Volume Offsite Error Not using Tape I/O Slave RMAN-07004: unhandled exception during command execution on channel t1 RMAN-10035: exception raised in RPC: ORA-19507: failed to retrieve sequential fi LE, handle="6sd8vntq_1_1", parms="" ORA-27029: skgfrtrv: sbtrestore returned error ORA-19511: Fatal catalog access error Piece 6sd8vntq_1_1 cannot be restored because volume AHI164 is offsite System Backup to Tape for Oracle Databases 9–23
- Page 113 and 114: Figure 6-5 Domain View Showing Expa
- Page 115 and 116: User Interfaces 6.1 Graphical User
- Page 117 and 118: 6.1.13 Viewing MDMS Audit and Event
- Page 119 and 120: 6.2.1 Syntax Overview User Interfac
- Page 121 and 122: User Interfaces 6.3 User Interface
- Page 123 and 124: 7 Preparing For Disaster Recovery I
- Page 125 and 126: 7.1.2 Backup of MDMS$ROOT Preparing
- Page 127 and 128: 7.2 Prolog and Epilog Procedure Pre
- Page 129 and 130: 7.2.1 Restoring The System Disk To
- Page 131 and 132: 8 Remote Devices 8.1 RDF Installati
- Page 133 and 134: Remote Devices 8.4 Monitoring and T
- Page 135 and 136: 8.4.4 Changing Network Parameters f
- Page 137 and 138: • Free Space is 20 Remote Devices
- Page 139 and 140: Remote Devices 8.5 Controlling Acce
- Page 141: Remote Devices 8.7 RDF Error Messag
- Page 144 and 145: System Backup to Tape for Oracle Da
- Page 146 and 147: System Backup to Tape for Oracle Da
- Page 148 and 149: System Backup to Tape for Oracle Da
- Page 150 and 151: System Backup to Tape for Oracle Da
- Page 152 and 153: System Backup to Tape for Oracle Da
- Page 154 and 155: System Backup to Tape for Oracle Da
- Page 156 and 157: System Backup to Tape for Oracle Da
- Page 158 and 159: System Backup to Tape for Oracle Da
- Page 160 and 161: System Backup to Tape for Oracle Da
- Page 162 and 163: System Backup to Tape for Oracle Da
- Page 166 and 167: System Backup to Tape for Oracle Da
- Page 168 and 169: System Backup to Tape for Oracle Da
- Page 170 and 171: System Backup to Tape for Oracle Da
- Page 172 and 173: System Backup to Tape for Oracle Da
- Page 174 and 175: Virtual Library System (VLS) 10.3 Q
- Page 176 and 177: Architecture 11.1 The Server Proces
- Page 178 and 179: Architecture 11.2 Scheduler Interfa
- Page 180 and 181: Architecture 11.3 Catalogs Example
- Page 182 and 183: Architecture 11.4 Coordinator 11.4.
- Page 184 and 185: Troubleshooting 12.2 Media Manageme
- Page 186 and 187: Troubleshooting 12.2 Media Manageme
- Page 188 and 189: Troubleshooting 12.4 ABS Catalogs 1
- Page 190 and 191: Troubleshooting 12.5 Windows and Un
- Page 192 and 193: Troubleshooting 12.6 RDF (Remote De
- Page 195 and 196: A Configuration Example Getting ABS
- Page 197 and 198: Configuration Example * Magazine -
- Page 199 and 200: Configuration Example Configuring s
- Page 201 and 202: Configuration Example If you comple
- Page 203 and 204: Configuration Example Assist: YES C
- Page 205 and 206: B Migrating from SLS/MDMS V2.X to A
- Page 207 and 208: Migrating from SLS/MDMS V2.X to ABS
- Page 209 and 210: Table B-1 SBK Symbols in ABS Termin
- Page 211 and 212: Migrating from SLS/MDMS V2.X to ABS
- Page 213 and 214: Migrating from SLS/MDMS V2.X to ABS
<strong>System</strong> <strong>Backup</strong> <strong>to</strong> Tape <strong>for</strong> Oracle Databases<br />
9.11 Troubleshooting Tips<br />
9.11.2 Fatal Internal Error<br />
SBT-00008140 12/17/01 09:50:50 MBytes/sec: 1.752 10<br />
SBT-00008140 12/17/01 09:50:50 Bytes/sec : 1752820.870 11<br />
The following describes the entries in Example 9–19:<br />
1. DB block size:-this is the size of blocks that Oracle sends <strong>to</strong> SBT.<br />
2. I/O block size:-this the size of blocks that SBT writes <strong>to</strong> the I/O device. You can change this<br />
with the logical MDMS$SBT_IO_BLOCK_SIZE which can be specified in an Oracle<br />
Recovery Manager script or as a system wide logical.<br />
3. Total I/Os:-<strong>to</strong>tal I/Os <strong>to</strong> write the I/O block size blocks <strong>to</strong> the I/O device. In this example:<br />
620 <strong>to</strong>tal I/Os writing 65024 byte blocks <strong>to</strong> the I/O device.<br />
4. Total I/O wait:-this is the number of milliseconds that SBT waited while the blocks are<br />
being written <strong>to</strong> the I/O device.<br />
5. Maximum I/O wait:-this is the longest wait that SBT made <strong>to</strong> write a block <strong>to</strong> the I/O<br />
device.<br />
6. Average I/O wait:-this the average wait of the <strong>to</strong>tal I/Os <strong>for</strong> the <strong>to</strong>tal I/O wait time.<br />
7. Total Kbytes:-<strong>to</strong>tal Kbytes transferred <strong>to</strong> the I/O device. It is a truncated value. See <strong>to</strong>tal<br />
bytes below.<br />
8. Total bytes:-<strong>to</strong>tal bytes transferred <strong>to</strong> the I/O device.<br />
9. Total seconds:-this is the number of seconds that SBT waited while the blocks are being<br />
written <strong>to</strong> the I/O device. This does not have the time that Oracle <strong>to</strong>ok <strong>to</strong> provide the in<strong>for</strong>mation<br />
<strong>to</strong> write <strong>to</strong> the I/O device. The value is truncated from the <strong>to</strong>tal I/O wait in milliseconds.<br />
10. Mbytes/sec:-Mbytes per second transferred based on <strong>to</strong>tal Kbytes and <strong>to</strong>tal seconds.<br />
11. Bytes/sec:-bytes per second transferred based on <strong>to</strong>tal bytes and <strong>to</strong>tal seconds.<br />
Note<br />
These values are the time that SBT waited <strong>for</strong> the I/O device <strong>to</strong> give back control. The<br />
values have nothing <strong>to</strong> do with how long it <strong>to</strong>ok Oracle <strong>to</strong> give in<strong>for</strong>mation <strong>to</strong> SBT.<br />
If you should ever get an Internal error, you should report it <strong>to</strong> the cus<strong>to</strong>mer support center.<br />
Example 9–20 shows an example of a fatal internal error. Anyplace that SBT could have<br />
received a value that it did not expect, it is captured and reported as a fatal internal error.<br />
Example 9–20 Fatal Internal Error Example<br />
SBT-00001DB2 12/17/01 13:53:45 Internal error<br />
SBT-00001DB2 12/17/01 13:53:45 Extended Status:<br />
The invalid archive type, 23<br />
SBT-00001DB2 12/17/01 13:53:57 ABEND (abnormal end ) was passed<br />
in sbtend flag<br />
9.11.3 Check ORA_DUMP:SBTIO.LOG <strong>for</strong> Errors<br />
Any time you receive an error in Oracle's Recovery Manager that is related <strong>to</strong> SBT, you should<br />
check ORA_DUMP:SBTIO.LOG <strong>for</strong> errors. We are limited <strong>to</strong> 250 characters returned <strong>to</strong> the<br />
Oracle Recovery Manager. There<strong>for</strong>e, you may not receive all of the in<strong>for</strong>mation about the error.<br />
However, I am not limited <strong>to</strong> what I put in the ORA_DUMP:SBTIO.LOG file. So be sure <strong>to</strong> look<br />
in ORA_DUMP:SBTIO.LOG when getting an error message that in not all there.<br />
9-22 <strong>System</strong> <strong>Backup</strong> <strong>to</strong> Tape <strong>for</strong> Oracle Databases