StorageTek T10000C Tape Drive
 

PLA-Support>Sun Systems>TAPE>Tape Hardware>SN-TP: SL3000-8500 Library
 

In this Document

Goal

Solution

References

Created from <SR 3-11614316107>

Applies to:

Goal

· In order to implement the OKM Encryption Appliances we needed to set up IP networks into which we connected the T10000C tape drives (as well as the Appliances).

· For one of these sites we used the 3COM switch that arrived with the SL8500s. This is a dumb switch which is incapable of sending messages to our SYSLOG server. We implemented two 3COM switches in that site. One for the local tape LAN and one for the remote key management and ILOM LAN.

· For the other three sites  the distance to the gateway for the management/ILOM LAN was too great. Consequently we implemented Cisco 2960S switches.

· When we plugged the Appliances into the Cisco switch all worked correctly.

· When we plugged the tape drives into the Cisco switch the error messages started being generated. All three sites were exactly the same. We disabled the 8 tape ports on each Cisco to stop the message flood.

Cisco messages:

10:15:23 NOTICE LOCAL7 Sep 10 07:15:22.521: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to down
10:15:25 NOTICE LOCAL7 Sep 10 07:15:24.523: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to up

10:16:23 NOTICE LOCAL7 Sep 10 07:16:22.522: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to down
10:16:25 NOTICE LOCAL7 Sep 10 07:16:24.525: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to up

10:17:23 NOTICE LOCAL7 Sep 10 07:17:22.524: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to down
10:17:25 NOTICE LOCAL7 Sep 10 07:17:24.527: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/2, changed state to up

What is a little strange is that the link is dropped by each tape drive at exact one minute intervals.

Solution

Note - some customers have experienced Ethernet port hang problems when they disabled the Ethernet port reset. A few customers had to run with the reset enabled.

Disable the T10000C Ethernet hardware reset.

Reference:
Bug Number: 19356034/18721078
Synopsis:
Customer requested that 10KC drives not reset the Ethernet hardware when zero packets have been received in 60 sec. This code (3.62.109) puts in a new VOP switch option to turn these resets OFF. Factory default is on because that’s the way it has always been. This replaces Special Build RB361101.

Engineering note:

The original implementation for this Ethernet/VOP switch was put in 3.62.109 (and RC362109) but did not work in that release. It was made to work properly in T10000C release 3.62.111 .
Additional NOTES: Code 3.62.111 sets the "60 second Ethernet RESET" to DISABLED as its initial IPL setting. Release 3.62.109 sets the default to ENABLED and will re-enable it if subsequently loaded after 3.62.111 has been loaded.

To Disable Ethernet reset in T10000C drive running on 3.62.111:

Open a VOP session to the drive
Select Configure -> Drive Data
Select the Maintenance tab
Check the Disabled box in front of "60 seconds Ethernet RESET" parameter
Save/IPL

Verify 60 seconds Ethernet RESET is Disabled after the drive reboot:
Select Retrieve -> View Drive Data
Select the Maintenance tab
Check 60 seconds Ethernet RESET is Disabled

Return the drive to service

References

This solution has no attachment