This appendix describes the format and interpretation of messages reported by dmcatadm(8) and dmvoladm(8). If you are uncertain about how to correct these errors, contact your customer service representative.
Messages in this section are divided into the format used for dmcatadm and dmvoladm.
Error messages generated when comparing the CAT database to the daemon database will start with the following phrase:
Bfid bfid - |
The bfid is the bit file ID associated with the message.
The preceding phrase will be completed by one or more of the following phrases:
missing from cat db missing from daemon db for vsn volume_serial_number chunk chunk_number msg1 msg2 |
In the above, msgn can be one of the following:
filesize < 0 chunkoffset < 0 chunklength < 0 zonenumber < 0 chunknumber <0 filesize < chunklength + chunkoffset zonenumber missing or improper vsn filesize != file size in daemon entry (size) no chunk for bytes msg1, msg2 |
In the above, msgn gives the byte range as nnn - nnn
nnn bytes duplicated |
Error messages generated when comparing the VOL database to the CAT database will start with the following phrase:
Vsn vsn |
The vsn is the volume serial number associated with the message.
The preceding phrase will be completed by one or more of the following phrases:
missing eotpos < largest position in cat (3746) eotchunk < largest chunk in cat (443) eotzone < largest zone in cat (77) chunksleft != number of cat chunks (256) dataleft !=sum of cat chunk lengths (4.562104mb) tapesize is bad version is bad blocksize is bad zonesize is bad eotchunk < chunksleft dataleft > datawritten volume is empty but msg1, msg2 |
In the above, msgn can be one of the following:
hfull is on hsparse is on hrsv is on datawritten != 0 eotpos != 1/0 eotchunk != 1 |
volume is not empty but msg1, msg2 |
In the above, msgn is one of the following:
hfree is on version < 4 but msg1, msg2 |
In the above, msgn can be one of the following:
volume contains new chunks hfull is off eotpos !=2/0 |
The following lists the meaning of messages associated with the dmcatadm database.
nnn bytes duplicated | Two or more chunks in the database contain data from the same region of the file. |
for vsn DMF001 chunk 77 chunkoffset < 0 | The chunkoffset value for chunk 77 on volume serial number (VSN) DMF001 is obviously bad because it is less than 0. |
for vsn DMF001 chunk 77 chunklength < 0 | The chunklength value for chunk 77 on VSN DMF001 is obviously bad because it is less than 0. |
for vsn DMF001 chunk 77 chunknumber < 0 | The chunknumber value for chunk 77 on VSN DMF001 is obviously bad because it is less than 0. |
for vsn DMF001 chunk 77 filesize < 0 | The filesize value for chunk 77 on DMF001 is obviously bad because it is less than 0. |
for vsn DMF001 chunk 77 filesize < chunklength + chunkoffset | The value of chunklength plus chunkoffset should be less than or equal to the filesize. Therefore, one or more of these values is wrong. |
for vsn DMF001 chunk 77 missing or improper vsn | The list of volume serial numbers for the chunk is improperly constructed. The list should contain one or more 6-character names separated by colons. |
for vsn DMF001 chunk 77 zonenumber < 0 | The zonenumber value for chunk 77 on DMF001 is obviously bad because it is less than 0. |
for vsn DMF001 chunk 77 zonenumber > chunknumber | Either the zonenumber value or the chunknumber value for chunk 77 on DMF001 is wrong, because the zonenumber is larger than the chunknumber value. (Each zone contains at least two chunks, because the end-of-zone header on the tape counts as a chunk.) |
for vsn DMF001 chunk 77 filesize != file size in daemon entry (nnn) | The filesize value in the chunk entry is different from the file size in the daemon record. If no daemon record was provided, this message indicates that more than one chunk exists for the bfid and that the filesize value is not the same for all the chunks. |
missing from cat db | The daemon entry was not found in the CAT database. |
missing from daemon db | No daemon entry was found for the entry in the CAT database. |
no chunk for bytes nnn - nnn | There is no chunk that contains the specified bytes of the file. |
The following lists the meaning of messages associated with the dmvoladm database.
blocksize is bad | The tapesize field for the tape is not a multiple of 4096. | |
eotpos < largest position in cat (3746) | The position for the EOT descriptor on the tape is less than the largest position of all the chunk entries for the tape. | |
chunksleft != number of cat chunks (256) | The number of chunks referencing the tape in the CAT database does not equal the number of chunks left recorded in the VOL entry for the tape. | |
dataleft != sum of cat chunk lengths (4.562104mb) | The sum of the chunks length for chunks referencing the tape in the CAT database does not equal the dataleft value recorded in the VOL entry for the tape. | |
dataleft > datawritten | The entry shows that more data remains on the tape than was written. | |
eotchunk < chunksleft | The entry shows that more chunks remain on the tape than were written. | |
eotchunk < largest chunk in cat (443) | The chunk number of the EOT descriptor on the tape is less than the largest chunk number of all the chunk entries for the tape. | |
eotzone < largest zone in cat (77) | The zone number of the EOT descriptor on the tape is less than the largest zone number of all the chunk entries for the tape. | |
missing | The volume was found in a chunk entry from the CAT database but is not in the VOL database. | |
tapesize is bad | The tapesize field for the tape is an impossible number. | |
version is bad | The version field for the tape is not 1, 2, or 3 (for a tape still containing data written by an old MSP) or 4 (for a tape written by this MSP). | |
| When a volume is empty, the hfull, hsparse, and hrsv hold flags should be off. | |
| When the hfree hold flag is cleared, the datawritten field is set to 0, the eotpos field is set to 1/0, and the eotchunk is set to 1. The entry is inconsistent and should be checked. | |
volume is not empty but hfree is on | When a volume contains data, the hfree hold flag must be off. | |
volume is not empty and version < 3 but volume contains new chunks | One or more of the chunks associated with this volume were written by the advanced tape MSP, which only writes tapes that have a version value equal to 3. However, the version value of the tape is less than 3. | |
volume is not empty and version < 3 but hfull is off | Tapes containing data with a version value of less than 3 must have hfull set, because the MSP cannot append to the tape. | |
volume is not empty and version < 3 but eotpos != 2/0 | Tapes imported from the old MSP only have one zone of data, so eotpos must be 2/0. | |
zonesize is too small | The zonesize field for the tape is an impossible number. | |