VAX Rdb IMPORT fails with "%RDO-E-NOIDXRES, unable to IMPORT index...%RDO-F-SYS_REQUEST

[Navipa]

Hi All,

We use CharonVAX OpenVMS 6.1, Rdb/VMS v4.1 as a Test server.

We want to EXPORT and IMPORT. The Export runs fine, but the IMPORT fails with the following errors "Unable IMPORT index...", earlier I tested with "NODATA" option and found it works fine and exported all the tables.  I've executed this $ $SET RMS_DEFAULT/EXTEND_QUANTITY=30000 before EXPORT.

The IMPORT comands is

RDO> IMPORT "TELK1:[BACKUP]TELKOM.RBR" INTO $DB

For the information, the table TELKCC is very old and very large size (75447348 records) compared with other tables. Our Database has 4 uniform and 3 mixed storage areas, and 20 tables, all tabs have indexes. Storage area, indexes, MAP, tables structure are attached as a file here

IMPORTing relation TELKCC
%RDO-E-NOIDXRES, unable to IMPORT index TELKCC_INDEX
%RDO-F-SYS_REQUEST, error from system services request
-SORT-F-SYSERROR, system service error
-RDMS-F-EXQUOTA, exceeded quota

IMPORTing relation TELKCOP
%RDB-F-DB_CORRUPT, database filename is corrupt
-RDMS-F-AREA_CORRUPT, storage area DB_AREA_08:[PROD]DEFAULT_AREA.RDA;1 is corrupt
%RDB-F-DB_CORRUPT, database filename is corrupt
-RDMS-F-AREA_CORRUPT, storage area DB_AREA_08:[PROD]DEFAULT_AREA.RDA;1 is corrupt
EXIT
===================================
The Failed Index detail
Indexes on table TELKCC:
CC_INDEX with column TRU_NO
and column TOR_NO
and column TCOP
and column TKE_NO
no duplicates allowed
type is sorted
Node size 960
Percent fill 70
Store clause: STORE in uniform_04

I appreciate your assistances.

Thanks

Navipa

[Hoff]

Check the Rdb manuals for the minimal process quota and system parameter settings, and raise the settings on this server and this username to at least the documentation-specified values, rebooting and/or re-logging in as required.    This per the -RDMS-F-EXQUOTA, exceeded quota error. 

[Navipa]

Thanks Hoff for your response.

I will look into your suggestion on "exceeded quota...", do you have any idea about "unable to IMPORT index and error from system services request, the details below..

%RDO-E-NOIDXRES, unable to IMPORT index TELKCC_INDEX
%RDO-F-SYS_REQUEST, error from system services request

Thanks

Navipa

[abrsvc]

It looks to me like the EXQUOTA is the reason for the failure rather than the failure being a separate issue.  Resolve the exquota problem and it is likely that the problem will be resolved in its entirety.

Dan

[Hoff]

---

@Navipa wrote:

Thanks Hoff for your response.

I will look into your suggestion on "exceeded quota...", do you have any idea about "unable to IMPORT index and error from system services request, the details below..

%RDO-E-NOIDXRES, unable to IMPORT index TELKCC_INDEX
%RDO-F-SYS_REQUEST, error from system services request

Thanks

Navipa

---

Could you elaborate on your thinking here?   It seems that you might believe that an error due to insufficient quotas is isolated and unrelated to the other errors.   If so, that's incorrect.

Within OpenVMS, quota errors are fundamental, low-level failures, and these errors and limitations can arise in many different parts of OpenVMS that can be used by applications, and the system andthe  application call stacks associated with these many and varied places — and the associated secondary errors — can and do vary.

Oracle Rdb is comparatively resource-intensive, and resource-intensive applications are subject to quota restrictions.  

Beyond outright crashes, incorrect quota settings can also lead to hangs and stalls, and to poor application performance.

FWIW and if you decide to increase quotas past the recommdations, that too also lead to problems.   If the quotas are too high, you can consume all available system resources, and hang the whole system.   Old VAX VMS configurations don't have much in the way of "spare" resources, by present standards.  If you do decide to go past the recommended settings, mantain the ratios across the settings, and watch for cases where system-wide resources — pagefile, non-paged pool, etc — are depleted.  That tends to lead to process or system-wide hangs.

The quota-related and system-parameter-related settings details and recommendations are covered in the Oracle Rdb installation and configuration documentation.

Beyond discussions of quotas and the associated limits, it's also advantageous during general troubleshooting to remove known errors when troubleshooting and debugging, as that can also resolve what are actually secondary errors; collateral damage.    Make one change, see if that works better or worse, then either back out the change or proceeed with the additional necessary debugging or changes.   Making multiple changes at once means you don't know what fixed the problem.   (Troubleshooting, debugging and researching are more often learned, and are unfortunately not skills that are universally taught.)

Hello "Your post has been changed because invalid HTML was found in the message body. The invalid HTML has been removed. Please review the message and submit the message when you are satisfied."  Nice to see you again.

[Navipa · Accepted solution]

Hi Hoff/Dan,
Thanks for your time and suggestions.
This issue has been resolved by creating BOUND Volume and Batch Update argument. The details are in the below thread.

http://community.hpe.com/t5/System-Management/SQL-IMPORT-fails-while-IMPORTing-very-large-table-size-3-9GB-on/td-p/6848396

http://community.hpe.com/t5/System-Management/Rdb-SQL-IMPORT-fails-with-quot-RMS-W-RTB-8224-byte-record-too/td-p/6838720

Thanks
Navipa.