[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
Title: | Oracle Rdb - Still a strategic database for DEC on Alpha AXP! |
Notice: | RDB_60 is archived, please use RDB_70 .. |
Moderator: | NOVA::SMITHI SON |
|
Created: | Fri Mar 18 1994 |
Last Modified: | Fri May 30 1997 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 5118 |
Total number of notes: | 28246 |
4987.0. "%RMU-E-POSITERR while RESTOREing from DISK backup" by ORAREP::NORISQ::SANKARAN () Fri Jan 31 1997 11:02
While doing an RMU/RESTORE/NORECOVERY from a FULL database backup that uses
the /OPTIONS and /DIRECTORY qualifiers, the following ERROR is REPEATEDLY
received after several storage areas are successfully restored:
%RMU-I-LOGRESSST, restored storage area SID$:[000000.SIDDB_PRACTICE]MPSR_HDR_S_I
DX.RDA;1
%RMU-E-POSITERR, error positioning PECOS"dbase_admin password"::DBACK1:[SID.BACK
UPS]SIDDB_F.RBF;19
-RMS-F-SYS, QIO system service request failed
-SYSTEM-F-LINKEXIT, network partner exited
%RMU-I-SPECIFY, specify option (QUIT or CONTINUE)
RMU> continue
%RMU-E-POSITERR, error positioning PECOS"dbase_admin password"::DBACK1:[SID.BACK
UPS]SIDDB_F.RBF;19
-RMS-F-WBE, error on write behind
-SYSTEM-F-LINKEXIT, network partner exited
%RMU-I-SPECIFY, specify option (QUIT or CONTINUE)
RMU> continue
%RMU-E-POSITERR, error positioning PECOS"dbase_admin password"::DBACK1:[SID.BACK
UPS]SIDDB_F.RBF;19
-RMS-F-ISI, invalid internal stream identifier (ISI) value
%RMU-I-SPECIFY, specify option (QUIT or CONTINUE)
RMU> quit
%RMU-F-ABORT, operator requested abort on fatal error
%RMU-F-FATALERR, fatal error on RESTORE
MSCHVS::DBA->
The HELP RMU_ERRORS states as follows on the %RMU-E-POSITERR error:
RMU
RMU_ERRORS
POSITERR
error positioning <str>
Explanation: Tape device rejected the attempt to set the tape
characteristics, or positioning.
User Action: Correct the device or media problem, and reissue
the command.
Please note:
o NO TAPES ARE INVOLVED in our backup/restore operations, only DISKS.
o The error ALWAYS occurs after the same storage area has been restored
(MPSR_HDR_S_IDX.RDA) as shown in the error re-produced here above.
o the OPTIONS file covers only a FEW storage areas; the rest are defaulted to
where the /DIRECTORY qualifier points to
o the restore is across TWO NODES, i.e., the /DIRECTORY points to a location
on a NODE DIFFERENT FROM WHERE THE BACKUP IS LOCATED.
o the initial restore attempted to restore the entire database to a single
location pointed to by /DIRECTORY, but failed because it ran out of disk
space. Hence the addition of the /OPTIONS file.
Any help will be greatly appreciated.
- Jyoti
T.R | Title | User | Personal Name | Date | Lines |
---|
4987.1 | | DUCATI::LASTOVICA | Is it possible to be totally partial? | Fri Jan 31 1997 11:08 | 11 |
| >o the restore is across TWO NODES, i.e., the /DIRECTORY points to a location
> on a NODE DIFFERENT FROM WHERE THE BACKUP IS LOCATED.
I'm not certain that this is supported, but I don't know if it
is documented one way or another.
If you are able to reproduce this with a current supported version
of Rdb (you didn't indicate what version you are using), then please contact
oracle rdb support (it appears that you are on the Digital network) with a
reproducable case (the exact commands needed to reproduce this behaviour
on a system here).
|
4987.2 | | DUCATI::LASTOVICA | Is it possible to be totally partial? | Fri Jan 31 1997 11:09 | 22 |
| btw, I think that you concentrated on the wrong status. The more
interesting one is probably:
LINKEXIT, network partner exited
Facility: SYSTEM, System Services
Explanation: The remote process exited before confirming the logical link
for one of the following reasons:
o A problem with the log file on the remote node
o An unopened remote log file because of a write-locked disk
(on the remote node)
o A premature exit of the process
User Action: Read the remote log or retry the operation.
I think that the remote log file could have some additional
information.
|
4987.3 | VMS and Rdb version details | ORAREP::NORISQ::SANKARAN | | Fri Jan 31 1997 11:21 | 12 |
| VAX/VMS versions:
o on BACKUP node: V5.5-2
o on RESTORE node: V6.1
Rdb versions:
o on BACKUP node: V6.0-05
o on RESTORE node: V6.0-05
|
4987.4 | | DUCATI::LASTOVICA | Is it possible to be totally partial? | Fri Jan 31 1997 11:27 | 5 |
| > o on BACKUP node: V6.0-05
you'll want to upgrade to the current version (6.0A ECO 5 [aka 6.0-15])
to determine if this problem has already been fixed (I've not checked the
release notes).
|
4987.5 | | M5::LWILCOX | Chocolate in January!! | Fri Jan 31 1997 15:57 | 31 |
| Well, when I try this on version 7.0 with the following command:
RMU/RESTORE/DIRE=LONGS"username password"::[LWILCOX] V7.RBF
I get:
%RMU-F-FILACCERR, error parsing file LONGS"username password"::DISK$USR1:[LW
ILCOX]V7.RDB;1
-COSI-I-REMOTE, remote file specification is not allowed
HELP on this errror states:
***************
REMOTE, remote file specification is not allowed
Explanation: A node name was found in the file specification. Node
names cannot be used.
User Action: Use a file name without a node specification.
However, I can restore locally a file that resides remotely, i.e.,
RMU/RESTORE/NEW LONGS"username password"::[LWILCOX]V7.RBF
FWIW. What is the exact command you're using and has this worked in the
past?
Liz
|
4987.6 | | NOVA::STENOISH | | Tue Feb 04 1997 16:37 | 5 |
| RESTORE does not support creation of remote databases. However, you
should be able to read a backup file remotely and create the database
locally.
Jim
|