-an error correction code that is recorded in each sector of a magnetic disk and used to catch errors in the data...
Following are suggestions for addressing each of the preceding issues.
1. Dirty tape-head. To fix this, run a tape head cleaner through your tape device a few times.
2. Bad tape. The tape might contain bad sectors, or it might be aging, and the coating material is coming off or has stretched due to high heat or use. Because 4/8mm tapes can become unreliable after 50 passes, Overland Storage recommends that you retire your 4/8mm tapes after 30-50 passes. If a tape has defective segments, the drive will attempt to write a data block "NNN" number of times (127 times for 4mm) before it skips a bad tape segment and moves on to the next. In this case, a backup that normally takes 1 gigabyte (GB) can fill a tape. Another sign is a backup that starts taking a lot more time to process than it usually does. There is a life-cycle variation with some brands and types of tapes becoming unreliable earlier than others; for example, DLT tapes can be used hundreds of times. Because it jeopardizes your backup process, Overland does NOT recommend using a tape until it fails. Also note that some brands of tapes seem to be more prone to bad sectors than others when used on different manufacturers tape drives. If a new tape fails, try using another brand or use a tape that is absolutely known to be good and see if it behaves differently.
3. Bad tape drive. A bad tape drive might manifest itself in several ways, including I/O errors, CRC errors, and filling up tape. In one documented case, a technician removed the top of a 4mm tape drive and watched a backup go through an error-free cycle without the tape moving an inch. (This is a good example why it is important to verify a completed backup.) UltraBac has been programmed to accumulate tape drive errors and terminate a backup if more than "NNN" number of errors (the default is 100) occur during a backup.
4. Termination or other cabling problems. Check to make sure your system is using active termination. Only very old SCSI devices use passive termination. Note that every SCSI controller has both internal and external device capability. If your system does not have any external devices, you must either have an active terminator plugged into the controller's external socket or you must deactivate the external SCSI devices option by using the manufacturer's software utility. Also note that almost every disk offers active termination through a jumper setting option on the drive itself. If you are experiencing SCSI errors, first remove all non-disk and tape SCSI devices from the bus experiencing the problem, for example, scanners, CD-ROM drives, and so on. Some of these devices can cause undue noise on the bus, which results in SCSI CRC errors. If problems continue, check each disk device and verify that it has not been jumped to provide active termination unless it is the last internal device on the cable. If you have an external disk cabinet, verify the same, but the cabinet should be terminated with an external active terminator (so that another device can be easily added). If problems persist, remove all nonessential devices except one disk and tape. Swap cables to eliminate this as a problem. Sometimes an external cable will work between the controller and the first device but will not work between the first and second device (due to low quality).
5. Bus-termination power. Normally the SCSI controller provides power to the bus for termination (this is the power source that illuminates the light on the active terminator at the end of the SCSI chain). When a new computer system is delivered from a major vendor, bus power is usually properly applied. However, a serious problem can exist or be created because new disk drives are also usually jumpered to apply bus-termination power when pulled new out of the box. If BOTH the SCSI controller and a disk drive apply bus-termination power, intermittent SCSI problems might occur during backup. From what Overland has seen, double bus-termination power is a leading cause of backup problems when upgrading to a high-performance backup program, even though normal disk activity does not seem to suffer or the standard Windows NT backup applet works okay. Unfortunately, the only way to eliminate this as a problem is by examining each internal and external disk for its jumper settings.
6. A problem with the SCSI card on the computer itself. If you are receiving Errors 1111 or 1117, check the SCSI configuration. Termination is quite often the problem with 1111 and 1117 errors especially if these errors do not occur on backup and DO occur on verify.
a. What type SCSI controller do you have?
b. What is the physical configuration, internal and external?
c. Check for proper termination power, termination (Active, NOT Passive).
d. Check your cabling. If possible, swap cables and retest.
e. Check the card itself. Run diagnostics and switch cards if possible.
f. Try disabling Wide-Negotiation, and lower the transfer rate on the SCSI controller.