If a user requests a medium via the UNKNOWN medium name or the /NEW qualifier and no media are currently in the available pool, an error message to that effect is issued.

2.4 Common job sequences

Described below are the messages and responses that you will need to make for the common MEDIA jobs that can occur on your system. The two most common operations are:

• Standard backup jobs created using the BCKMGR_MAINT.COM procedure.
• Medium jobs created using the MEDIA job commands.

2.4.1 Responding to standard Backup Jobs

Once the job starts executing, it will allocate an appropriate drive to mount the medium on. Then you will be prompted with the following OPCOM messages ¹²

to mount and dismount various media.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:00:39.84 %%%%%%%%%%% Request 8, from user SYSTEM on ISE Please mount volume 1802 in device _ISE$MSA0: External label 1802, from library, density 1600 bpi, for writing %%%%%%%%%%% OPCOM 16-OCT-1996 09:03:01.59 %%%%%%%%%%% Request 8 was satisfied.

Here you were requested to mount tape 1802 located in the library for writing (i.e. put a write ring in) in drive MTA1:. After you physically place the tape on the drive MTA1:, the system automatically detects the situation and continues with the job without any direct response on your part.

Once started, the job automatically selects the group of tapes to use. ¹³

This information is then sent to you via an OPCOM message of the following format.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:03:43.77 %%%%%%%%%%% Message from user SYSTEM on ISE BATCH_912, INCREMENTAL: Backup of DISK$ISE0: started, use volumes 1802,1803,1817,1827, on _ISE$MSA0:

The number of tapes selected is defined via the REQUIRED field in the job definition. ¹⁴

This job is an incremental backup of three disk volumes. Each disk is processed by a separate BACKUP command. The first backup operation only took up part of the first reel. The second backup operation can thus continue on the first reel. ¹⁵

At the beginning of the second backup operation the list of volumes that can possibly be used is sent to you via an OPCOM message. In this example the next message appears below.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:11:17.38 %%%%%%%%%%% Message from user SYSTEM on ISE BATCH_912, INCREMENTAL: Backup of DISK$ISE1: started, use volumes 1802,1803,1817,1827, on _ISE$MSA0:

The second backup operations continues to the end of the first reel and requests a second reel to be mounted. The below OPCOM message will inform you of this.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:13:27.43 %%%%%%%%%%% Request 9, from user SYSTEM on ISE Please mount volume 1803 in device _ISE$MSA0: BACKUP requests: Saveset ISE1.BCK, Volume number 02, write ENABLED

After you have mounted the reel 1803 on the tape drive the system automatically detects this and proceeds. The following OPCOM message is sent to indicate this.

%%%%%%%%%%% OPCOM 11-OCT-1996 14:00:02.95 %%%%%%%%%%% Request 9 was satisfied.

The third backup just continues on the second reel. You are sent the following message to indicate that the third backup is starting.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:16:14.77 %%%%%%%%%%% Message from user SYSTEM on ISE BATCH_912, INCREMENTAL: Backup of DISK$ISE2: started, use volumes 1803,1817,1827,1828, on _ISE$MSA0:

When the job is done you are sent the following message.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:29:23.63 %%%%%%%%%%% Message from user SYSTEM on ISE Done with medium 1802

This message indicates that the job completed successfully.

2.4.2 Responding to requests for SCRATCH media

This section describes the OPERATOR response to a request for a scratch media. The difference between the loading of scratch media and any other type of media is that with scratch media a specific volume is not requested. Instead the operator should designate a collection of scratch media and be sure that they meet the characteristics of the request.

The request shown below prompts the operator to mount a scratch medium of type TK50 with a density of 1600 bpi for writing.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:00:39.84 %%%%%%%%%%% Request 221, from user SYSTEM on ISE Please mount volume SCRATC in device ISE$MSA0: Scratch SCRATCH(TK50), density 1600 bpi, for writing %%%%%%%%%%% OPCOM 16-OCT-1996 09:03:01.59 %%%%%%%%%%% Request 8 was satisfied.

The criteria used to determine if the media is acceptable is:

• The type of the SCRATCH entry must be equal to the type of the loaded medium.
• The location (if not BLANK) of the SCRATCH entry must equal the location of the loaded medium.
• The length (if not zero) of the SCRATCH entry must equal the location of the loaded medium.

If a loaded medium is not acceptable, an OPCOM message is generated as shown below.

%%%%%%%%%%% OPCOM 16-OCT-1996 09:03:01.59 %%%%%%%%%%% Scratch medium in ISE$MSA0 rejected. Scratch medium not in available pool.

The following are reasons why a loaded medium can be rejected.

• The medium could not be found in the database.
• The Internal and External names were not identical.
• The medium is not in the available pool.
• The medium is not of the requested type.
• The medium is not from the requested location.
• The medium is not of the requested length.

If an acceptable scratch medium is loaded, the following attributes of the scratch template are copied over into the newly acquired medium.

• All label flags.
• All mount flags & conditions (block, record, density)
• All update flags
• Protection settings (Owner, Access settings, ACL)
• Comment

Any medium acqired by this method is automatically marked as `new'. This implies that the first operation to such a volume must be an initialize operation.

2.4.3 Handling BACKUP error messages

When errors occur during VMS BACKUP save and restore you will receive OPCOM messages indicating the condition.¹⁶ The most common problems are parity failures, wrong volume and write ring missing. In each case, BACKUP will allow you to select an option and REPLY.

A typical OPCOM message is shown below:

%%%%%%%%%%% OPCOM 16-OCT-1996 09:35:12.82 %%%%%%%%%%% Request 10, from user SYSTEM on ISE %BACKUP-W-MOUNTERR, volume 2 on _ISE$MUA0 was not mounted because it is protected against access specify option (QUIT, NEW tape or OVERWRITE tape)

This message indicates that the mounted volume has the correct volume label but the UIC ownership is different from that of the user doing this backup operation. In this case all the backup jobs had been switched over to a different user and thus a different UIC. As such OVERWRITE was selected to instruct BACKUP to use the tape anyway.

$ REPLY/TO=10 OVERWRITE

In response to the REPLY being issued the following OPCOM message is displayed.

OVERWRITE 09:39:11.15, request 10 was completed by operator _MSB$WTA1:

In this example the BACKUP operation continued using the reel that was already on the drive.

The messages can be grouped into two categories: label processing errors and (\em data transfer errors). The previous example was from a label processing error. The response that can be given in each are listed in Table 2-3.

Table 2-3 BACKUP error options

Response	Meaning
Label processing errors
QUIT	You cannot correct the error and you want to terminate the backup job immediately.
NEW	You are going to replace the current mounted reel with a new one and BACKUP is to initialize it.
OVERWRITE	You have visually verified that the tape mounted is indeed the correct tape and you want the system to use it regardless of what the (internal) label on the reel says.
Data transfer errors
QUIT	You cannot correct the error and you want to terminate the backup job immediately.
CONTINUE	You want BACKUP to continue reading or writing to the tape even though there are a lot of read/write failures.
RESTART	You want BACKUP to restart the CURRENT volume (not the entire volume set). You can at this point also replace the volume with a new one if it is needed.

Data transfer errors, are those errors that occur while reading or writing the backup volume. Assume, for example, that you received the following messages during a restore operation.

%%%%%%%%%%% OPCOM 1-JAN-1996 13:02:30.44 %%%%%%%%%%% Message from user OPMANAGER %BACKUP-E-FATALERR, fatal error on MTA1:[]ISE1.BCK; %%%%%%%%%%% OPCOM 1-JAN-1996 13:02:30.44 %%%%%%%%%%% Message from user OPMANAGER %BACKUP-F-MEDOFL, medium is offline %%%%%%%%%%% OPCOM 1-JAN-1996 13:02:30.44 %%%%%%%%%%% Request 10, from user OPMANAGER %BACKUP-I-SPECIFY, specify option (QUIT or CONTINUE)

At this point, you check the drive and find that it has lost its vacuum, just ready the drive and enter the following command:

$ REPLY/TO=10 CONTINUE

You could have selected QUIT if the drive was inoperable and you wanted to terminate the job at this point.

2.4.4 Responding to Media Jobs

Operator requests that originate from batch jobs created using any of the Media Job commands described in the General User Guide section generate two types of requests. One requires a REPLY from the operator, the other can automatically detect the completion.

Several circumstances can arise during a media mount operation that makes the volume identification uncertain (for example, an unlabeled tape). In these cases, a request is sent to you that requires that you respond with a REPLY/TO command when the operation is done. The assumption being, since the system cannot determine if it is the correct volume, then you must. Such a situation produces the sequence below.

%%%%%%%%%%% OPCOM 15-MAR-1996 17:26:29.93 %%%%%%%%%%% Request 17, from user SANAVI Please mount volume 1817 in device _MTA1: (reply required) External label 1817, from cabinet 3b, for readonly $ REPLY/TO=17 17:28:19.19, request 17 was completed by operator _TTA1: %%%%%%%%%%% OPCOM 15-MAR-1996 17:53:18.25 %%%%%%%%%%% Message from user SANAVI Done with media 1817

In most cases, the system can verify that the correct medium is mounted on the drive and the sequence would appear as below. Here you just load the tape on the tape drive. At that point, the system request was automatically satisfied without issuing a REPLY command.

%%%%%%%%%%% OPCOM 15-MAR-1996 17:28:26.67 %%%%%%%%%%% Request 18, from user SANAVI Please mount volume 1817 in device _MTA1: External label 1817, from cabinet 3b, for readonly %%%%%%%%%%% OPCOM 15-MAR-1996 17:29:28.14 %%%%%%%%%%% Request 18 was satisfied. %%%%%%%%%%% OPCOM 15-MAR-1996 17:53:18.25 %%%%%%%%%%% Message from user SANAVI Done with media 1817

These mount requests will not appear at all if the system can determine that the medium that is already loaded on the drive is the required one.

If you are requested to mount a tape for writing and mistakenly leave out the write ring, on a tape or a plastic slider on a cartridge, the system will unload the medium and re-request that the medium be mounted. The following type of OPCOM messages are generated.

%%%%%%%%%%% OPCOM 15-MAR-1996 17:58:19.01 %%%%%%%%%%% Message from user SANAVI Medium 1817 on _MFA1: is write locked, write access is required. %%%%%%%%%%% OPCOM 15-MAR-1996 17:58:22.25 %%%%%%%%%%% Request 19, from user SANAVI Please mount volume 1817 in device _MFA1: External label 1817, from cabinet 3b, for writing.

2.4.5 RDB Backup messages

When performing an RMU backup on RDB files, all messages must be explicitly replied to. No automatic "request satisfied operations" are possible. A typical request for a second volume might be:

%%%%%%%%%%% OPCOM 15-MAR-1996 17:26:29.93 %%%%%%%%%%% Request 171, from user JOHNSON on ISE %RMU-I-READYWRITE, mount volume 2 on _ISE$MSA0: for writing

If the incorrect volume is mounted:

%%%%%%%%%%% OPCOM 15-MAR-1996 17:36:29.93 %%%%%%%%%%% Message from user JOHNSON on ISE %RMU-I-WRNGLBL, This tape was incorrectly labeled. Expected TEST01 - Found TEST03 %%%%%%%%%%% OPCOM 15-MAR-1996 17:37:27.98 %%%%%%%%%%% Request 172, from user JOHNSON on ISE %RMU-I-TAPEDISPW, Specify tape disposition (QUIT, INITIALIZE, RETRY, UNLOAD)

An operator resonse would be:

REPLY/TO=172 UNLOAD UNLOAD 10:48:54.96, request 172 was completed by operator _BOOT$0PA0 %%%%%%%%%%% OPCOM 15-MAR-1996 17:39:30.93 %%%%%%%%%%% Request 173, from user JOHNSON on ISE %RMU-I-READYWRITE, mount volume 2 on _ISE$MSA0: for writing

2.5 Extending volume sets

Whenever the amount of data to be written to a medium exceeds its capacity that medium needs to be extended. Every volume in the MEDIA DATABASE can be expanded to become a volume set. ¹⁷

You as an OPERATOR will see this event when the system asks for a second or subsequent volume and no such volume has been defined. At this point it is up to you to attach a volume and respond to the request. There are two primary sources for such requests. The first is for tapes mounted as ANSI volumes. The second is those that are mounted foreign and being used by VMS BACKUP. Volume sets can contain up to 45 ¹⁸ media.