Error: "Data error (cyclic redundancy check)" - According to Veritas (TechNote ID: 191896 link below), the "Cyclic Redundancy Check" (CRC) error is generic in nature and can be caused by numerous factors. The following list contains the most common reasons for this error and potential ways to resolve the problem:
1. Contaminated read/write heads of the tape device. Check with the hardware manufacturer for proper cleaning techniques.
2. Bad media. Replace the media. Try a new tape that is certified by the hardware manufacturer.
3. Tape driver. Load the appropriate Backup Exec tape driver.
4. SCSI controller wide negotiation enabled. Use the manufacturer's SCSI setup program to disable wide negotiation on the SCSI controller card. SCSI controller wide negotiation. If the device is a wide (68 pin) SCSI device then wide negotiation may and should be used. If the device is narrow )50 pin) SCSI device, disable wide negotiation.
5. SCSI controller transfer rate is too fast. Use the manufacturer's SCSI setup program to lower the SCSI transfer rate. Check with the controller and backup device manufacturer fir the proper configuration for SCSI transfer rate.
6. SCSI controller synchronous negotiation enabled. Use the manufacturer's SCSI setup program to disable synchronous negotiation on the SCSI controller card. Check with the controller and backup device manufacturer fir the proper configuration for SCSI synchronous negotiation.
7. Incorrect termination or bad cables. Verify the SCSI cable is good and for proper SCSI termination. DO NOT MIX Passive and Active termination.
8. Confirm the tape drive is functioning properly. Check with the tape drive manufacturer for diagnostic software to test the condition of the tape drive hardware.
9. General SCSI problems. Isolate the tape drive on its own controller card.
As per Veritas: "This is a generic event logged for all storage device errors. The event was reported in the application log because the Backup Exec job engine received the specified error when issuing the specified command to the specified drive. This is most often indicative of a hardware failure. Depending on the type of error, it may indicate that the drive needs to be cleaned or that the media is no longer reliable. Other errors may be resolved by stopping the Backup Exec services cycling power on the drive and restarting the services". See "Veritas Support Document ID: 253616" for more details.
- Error: "When accessing a new tape of a multivolume partition, the current block size is incorrect" - I received this error because the block size of my tape drive was set incorrectly. Setting it back to the default fixed the issue.
See the link to "Veritas Support Document ID: 190750" for additional information on this issue.
I have seen this error occur on two different servers at different times. The guilty device turned out to be the SCSI controller card by way of faulty drivers. Prior to doing this, I thought my (SCSI) card was bad, or the tape drive had finally played out. What finally led me to believe it was the card was due to the drive being able to re-label tapes and read its information prior to backup failures. After loading new drivers for my SCSI card, backup operations are working again.
Error: "The requested could not be performed because of an I/O device error." - The problem may be resolved by isolating the tape device on an IDE controller in the computer. For example, if the tape and a CD-ROM are connected to the same IDE controller on the motherboard, move the CD-ROM to a secondary IDE controller.
Note: This can also be used to resolve other hardware problems (including Cyclic Redundancy Check (CRC) errors) encountered in backup and restore operations with Backup Exec. See also Veritas TechNote ID: 191087 link below.