| Title: | Ask the Storage Architecture Group |
| Notice: | Check out our web page at http://www-starch.shr.dec.com |
| Moderator: | SSAG::TERZA N |
| Created: | Wed Oct 15 1986 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 6756 |
| Total number of notes: | 25276 |
Un-formatted RF31 drive mechanics???? RF31, fault LED on. Replacing the drive mechanics module 70-24697-01 solved this. Drive goes ready after a few minutes of calibration. VMS (5.5-1) initialize command fails with; Invalid Media Format. Drive internal diagnostics DRVTST and DRVEXER fail immediately. DKUTIL "GET" command fails with; "All copies of RCT bad". This has happened with two different drive mechanics (Rev J02) and two different drive modules 54-18329-01 (Rev.J02 and E01). Customer starts to get worried, and I am confused; please help! Regards, Espen Arnestad MCS Norway
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 6473.1 | not unformatted, just bad hda. | SUBSYS::VIDIOT::PATENAUDE | Ask your boss for ARRAY's... | Thu Mar 13 1997 07:46 | 12 |
The first thing the drive does on spin up is try to read in the metadata that contains format/rct/fct/ect,,, information. It cannot do that if the heads are off the arms and sitting on the bottom of the HDA. Replace the entire drive and not just the ECM module. Also the factory calibration data was on the original module you removed, so even if the heads are intact, the drive most likely has all the wrong offset information in the new ECM and would not be able to read either. roger. | |||||
| 6473.2 | there are two calibration cycle types | FIEVEL::FILGATE | Bruce Filgate SHR3-2/W4 237-6452 | Thu Mar 13 1997 08:20 | 8 |
A mated set of drive/ECM will calibrate is very short order. Swapping the ECM or the HDA results in an un-mated set and bogus data tables in the ECM. This bogus data will be over written when the ECM detectes that serial number on the HDA has changed, a long recalibration will be initiated which should run about 20 minutes or so. Bruce | |||||
| 6473.3 | A few more suggestions . . . | SUBSYS::BROWN | SCSI and DSSI advice given cheerfully | Thu Mar 13 1997 08:56 | 18 |
First, please confirm that the FCT is OK, and that only the RCT is
unusable. Run DKUTIL, and use the GET command.
Second, see if Roger's theory about the heads being useless is correct.
Run DKUTIL, say "GET /IMF" (ignore media format), then say
"DUMP PBN 0 100 0", then "DUMP PBN 1 100 0". If the EDC on each PBN is
correct, and you don't hear any grinding noises, the heads are working.
Third, check that the ECM is properly set for an RF31. Run PARAMS,
say "STAT CONFIG", and ensure that it really thinks it's an RF31, not
an RF31F or RF72. Check for a reasonable code revision (V255, V256, V257,
V258). Say "STAT LOG", and read the last number on the first line of the
highest-numbered log entry. Then, power-cycle the drive, and see if
a new entry is recorded. Also in PARAMS, try saying "STAT BBR",
to get more information about the RCT. You might need a password.
Finally, try a wild guess. In PARAMS, say "SHOW CALMASK", and
ensure it says FFE0 or FFC0.
| |||||
| 6473.4 | * Problem solved! * | OSLLAV::ESPEN_P | Fri Mar 14 1997 06:10 | 7 | |
Thanks for all inputs. I replaced both the mech.assy and drive module as a pair, new from SLOG, and this solved the problem. Situation did not permit further analyzis of the problem. Have a nice weekend, Regards, | |||||