StoreEver Tape Storage
1748112 Members
3508 Online
108758 Solutions
New Discussion юеВ

Re: SureStore Library 10/588 and DP 5.10 Problems

 
rccmum
Super Advisor

SureStore Library 10/588 and DP 5.10 Problems

Hi All,
I am facing quite peculier and intermittent
problems since couple of days. This library has 8 Quantum DLT7000 drives , Product ID - A4845A Firmware - 2010
HP-UX 11.11
At one time , I see in DP session ->messages
Major : Can't unload exchanger medium ( Unknown Device type specified )

Sometimes , I see BMA timeouts for a drive
The drive changes most of the time . Well the frequency of this timeout is more compared to other errors.

Critical : /dev/rac/picker
Tape Alert [1]: The library mechanism is having difficulty communicationg with the drive.

There are no messages being logged at OS side
neither in syslog nor in dmesg.

Using schgr as medium changer , I can see major number 230 for /dev/rac/picker.

I am suspecting the most probable cause would a hardware issue at library side and with library mechanism . I am not sure how to trace the root cause.
I guess L&TT support this library . Will it help to identify the problem?

I am thinking of getting the library serviced to make sure all components are aligned ( mechanically) , calibrated etc. Does any one has got it done ever?

You feedback would be appretiated and points would be assigned for sure.

14 REPLIES 14
David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems

The tape alert message (which are a bit basic, but somewhat helpful) implies that the library is having issues communicating to the drive over the serial port (which it uses to control loading/unloading of tapes, checking status, etc).

To better diagnose the issue, I'd suggest looking at the FSC (fault system code) log on the 10/588 operator panel. The last 20 FSCs should be logged. They are two byte hex codes.
The journey IS the reward.
rccmum
Super Advisor

Re: SureStore Library 10/588 and DP 5.10 Problems


These are the FSCs recored in the Library

1. # 30D0 REACH 3
2:42:14 9/22
2. #3205 DRV- 2 6
2:42:8 9/22
3. #3095 REACH 3
2:41:55 9/22
4. #30C4 REACH 94
2:41:55 9/22
5. #30C0 CAP 0 1
11:11:7 9/19
6. #3104 NONE 3
2:25:45 9/18
7. #30C7 REACH 3
2:25:44 9/18
8. #30B2 NONE 18
1:55:18 9/18
9. #3219 DRV- 5 3
8:26:51 9/16
10. #309F REACH 3
8:26:50 9/16
11. #30A2 REACH 3
8:26:50 9/16
12. #309D REACH 3
8:26:50 9/16
13. #3204 NONE 1
0:40:34 9/16
14. #3207 DRV- 5 1
0:40:34 9/16
15. #308D REACH 1
0:40:33 9/16
16. #3217 DRV- 0 2
12:23:55 9/14
17. #3216 DRV- 0 2
12:23:54 9/14
18. #3C87 NONE 9
12:02:21 9/14
19. #3108 NONE 4
8:40:15 9/9
20. #30B4 Z 2
15:32:29 8/31

I tried co-relate these FSCs with their timestamp with DP messages , but there is no match.
Whats those FSCs signifies? .Is it possible to decode them in way to know the cause and get it rectified?

Recently we replaced drives 1 3 6 between 8/29 and 9/18

Thanks in advance
David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems

Here's the meaning of the FSC codes you had on the 22nd:

1. #30D0 REACH 3 2:42:14 9/22
DLT Drive Handle Operation Request rejected - hand is full

2. #3205 DRV- 2 6 2:42:8 9/22
The cartridge drive elevator is not up.

3. #3095 REACH 3 2:41:55 9/22
The max or min current was requested an excessive amount of times

4. #30C4 REACH 94 2:41:55 9/22
An A2D conversion error has taken place. The 'overflow' bit was set when the software tried to read from the converter. This error normally can be treated as a warning and can be expected if there are obstructions in the path of the mechanism. -

---

I'm guessing that the library may be having problems operating the drive handle used to unload the tape. I've seen a few cases where the screw holding the handle in place came loose.

For these large libraries, having trained service folks look at it is usually the best choice.

The journey IS the reward.
rccmum
Super Advisor

Re: SureStore Library 10/588 and DP 5.10 Problems

Hi David,
I appretiate your replies, Thanks.

It is clear that none of these entries repeated atleast during 20 entries being logged. This supposed to me as an irrational behaviour of the library . Moreover more "NONE" entries also bothers me and have happened in the recent past. Can you tell what those entries signifies? I heard that there is a FSC dictionary , where I can find it?

I don't have either service or diagnostic manual. Does HP give these manual along with product?
We do have support contract with HP for this library. Can we get these manuals from HP Support?
These manuals would be helpful to assess the problems and come up with right feedback to
HP Support to make them understand the problem and resolve it completly.

I do agree with you to get the library serviced from a trained HP person. As of now , I see you as the HP library expert ( HP Engineering) . I want to get the library serviced in all respects as these problems occuring since very long time as told by my seniors. Can you please advice to ensure the same like mechanical alignements fittings , circuit testings , connections etc...?

Your reply is awaited

Thanks in Advance
David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems

> It is clear that none of these entries repeated atleast during 20 entries being logged.

I should have explained the fsc code display in more detail.

The general format is:



So nearly all of these had multiple occurances, and as a result the log doesn't show the exact time order the events occurred (just the latest time each event occurred).

The best way to troubleshoot reoccuring issues is clear out the FSC log, and examine it after the first failure.

> Moreover more "NONE" entries also bothers me and have happened in the recent past. Can you tell what those entries signifies?

"None" just means that there is no specific location in the library that the event can specifically be tied to, or are caused by user intervention.

From your list, the most recent two were:

> 6. #3104 NONE 3 2:25:45 9/18
> 8. #30B2 NONE 18 1:55:18 9/18

3104: The hand failed retracting to a safe position.

30B2: Access door is open.

> I heard that there is a FSC dictionary , where I can find it?

The FSCs are defined in a file called "fsc.dos" provided by the OEM vendor with each firmware release.

> I don't have either service or diagnostic manual. Does HP give these manual along with product?

The service/diagnostic manuals are labeled "for internal use only" by our OEM supplier, so we are not allowed to publish them to the external web. I believe the same applies to fsc.dos.

> We do have support contract with HP for this library. Can we get these manuals from HP Support?

You can certainly ask them.

> I do agree with you to get the library serviced from a trained HP person. As of now , I see you as the HP library expert ( HP Engineering) . I want to get the library serviced in all respects as these problems occuring since very long time as told by my seniors. Can you please advice to ensure the same like mechanical alignements fittings , circuit testings , connections etc...?

If you have a support contract, then HP support should be providing the necessary troubleshooting, repair, and/or calibration to resolve the issues you are having.

The journey IS the reward.
David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems

I should mention one additional point with regard to FSCs reported by the enterprise libraries such as the 10/588, 10/180, 20/700, etc (all from the same OEM vendor).

An FSC code logged does NOT necessarly mean the library had a failure, or the library is having a significant problem. Some FSC codes can appear in otherwise normal operation.

As an example, the FSC code 3384 means "When scanning down past a cartridge the label could not be read." Because 3 re-trys are done on each unreadable label, unless all retries are logged no action should be taken. A retry on a label barcode can occur if the label is crooked, damaged, scuffed up, etc. Additionally, if the customer uses unlabeled cartridges in their library this error will appear frequently.

I just wanted to make that clear before anyone panics because there are FSC codes logged on their unit.
The journey IS the reward.
David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems


FYI, The STK 9730 (very similar controller to the 10/588) manual appears here:
http://linux.auxio.org/auxio/misc/stk-9730/stk-9730.pdf

Page 67 explains how you use the FSC log.
The journey IS the reward.
rccmum
Super Advisor

Re: SureStore Library 10/588 and DP 5.10 Problems

David,
Thanks a lot for the clarifications and the doc.

What about using L&TT ? Does it support this library? Does the current version of L&TT available on Web has the firmwares ( for DLT 7000 drives etc.) in-built for this library?

Does this library has RMC in -built or needs to be installed ? How do I make use of RMC to monitoring it remotely?





David Ruska
Honored Contributor

Re: SureStore Library 10/588 and DP 5.10 Problems

When LTT first released, the 10/588 product was already discontinued and therefore support was not added for this. The L180/L700 products were supported by LTT, but only very basic operations like firmware update and capturing (but not decoding) the FSC logs.

Also, the 10/588 was typically connected to HP-UX, so the STM diagnostic was the supported host diagnostic tool.

I'm not aware of any remote management for the 10/588 - the L-series products were the first ones from this vendor to add the web interface from what I recall.

However, there is a serial interface (called the CSE port) on the 10/588 that is used by service engineers.
The journey IS the reward.