Document Audience: | INTERNAL |
Document ID: | I0579-1 |
Title: | Systems with Ultra DWIS/SBus Host Adapter slow to a halt under heavy IO loads and the Console displays "SCSI transport failed" error messages. |
Copyright Notice: | Copyright © 2005 Sun Microsystems, Inc. All Rights Reserved |
Update Date: | 2000-06-08 |
---------------------------------------------------------------------
- Sun Proprietary/Confidential: Internal Use Only -
---------------------------------------------------------------------
FIELD INFORMATION NOTICE
(For Authorized Distribution by SunService)
FIN #: I0579-1
Synopsis: Systems with Ultra DWIS/SBus Host Adapter slow to a halt under heavy IO loads and the Console displays "SCSI transport failed" error messages.Create Date: Jun/07/00
Keywords:
Systems with Ultra DWIS/SBus Host Adapter slow to a halt under heavy IO loads and the Console displays "SCSI transport failed" error messages.
Top FIN/FCO Report: Yes
Products Reference: UDWIS/SBus SCSI Adapter
Product Category: Storage / SW Admin
Product Affected:
Systems Affected
------- ---------
Mkt_ID Platform Model Description Serial Number
------ -------- ----- ----------- -------------
- A14 ALL Ultra Enterprise 2 -
- E3000 ALL Ultra Enterprise 3000 -
- E3500 ALL Ultra Enterprise 3500 -
- E4000 ALL Ultra Enterprise 4000 -
- E4500 ALL Ultra Enterprise 4500 -
- E5000 ALL Ultra Enterprise 5000 -
- E5500 ALL Ultra Enterprise 5500 -
- E6000 ALL Ultra Enterprise 6000 -
- E6500 ALL Ultra Enterprise 6500 -
- E10000 ALL Ultra Enterprise 10000 -
X-options Affected
--------- --------
X1065A - - Ultra DWIS/SBus Host Adapter -
- D1000 ALL StorEdge D1000 -
- A1000 ALL StorEdge A1000 -
- A3000 ALL StorEdge A3000 -
- A3500 ALL StorEdge A3500 -
Parts Affected:
Part Number Description Model
----------- ----------- -----
370-2443-01 UDWIS/SBus SPORT-20D ULTRA -
370-2443-02 UDWIS/SBus SPORT-20D ULTRA -
References:
BugId: 4279917, 4318531, 4294710, 4280274, 4237022, 4288995, 4327742,
4284656, 4292114, 4322298, 4318531
Esc#: 522822, 524889, 525094, 525353, 523726, 523767, 525290
Issue Description:
These failures will cause timeouts on the SCSI bus and drastically slow
down the IO operations and eventually hang the system.
The following SCSI error messages were produced under very heavy IO
loads when using the UDWIS SBus Host Adapter connected to the StorEdge
D1000, A1000, A3000, and A3500 in the desktop and/or server
configurations.
WARNING: /sbus@e,0/QLGC,isp@2,10000/sd@a,0 (sd1201):
SCSI transport failed: reason 'incomplete': retrying command
WARNING: /sbus@e,0/QLGC,isp@2,10000/sd@2,0 (sd1194):
SCSI transport failed: reason 'incomplete': retrying command
WARNING: /sbus@2,0/QLGC,isp@1,10000/sd@2,0 (sd1074):
SCSI transport failed: reason 'timeout': retrying command
WARNING: /sbus@2,0/QLGC,isp@1,10000/sd@b,0 (sd1082):
SCSI transport failed: reason 'incomplete': retrying command
WARNING: /sbus@e,0/QLGC,isp@2,10000/sd@b,0 (sd1202):
SCSI transport failed: reason 'tran_err': retrying command
WARNING: /sbus@2,0/QLGC,isp@1,10000/sd@a,0 (sd1081):
SCSI transport failed: reason 'incomplete': retrying command
WARNING: /sbus@2,0/QLGC,isp@1,10000/sd@b,0 (sd1082):
SCSI transport failed: reason 'incomplete': retrying command
The "incomplete", "timeout" and "tran_err" failures occur due to UDWIS
adapter failing SCSI bus arbitration under heavy I/O loads in systems
where heavy SCSI I/O cause the SCSI commands queued in the host
adapter memory to exceed 190.
Implementation:
---
| | MANDATORY (Fully Pro-Active)
---
---
| X | CONTROLLED PRO-ACTIVE (per Sun Geo Plan)
---
---
| | REACTIVE (As Required)
---
Corrective Action:
The following recommendation is provided as a guide line for authorized
Enterprise Services Field Representatives and Enterprise Customers that
may be at risk of encountering the above mentioned problem;
The following solution applies to configurations listed below:
1. 1 D1000 connected to 1 UDWIS host adapter
2. 1 A1000 and A3X00 enclosure connected to 1 UDWIS host adapter.
3. Multiple A1000 and A3X00 enclosures connected to 1 UDWIS host
adapter and the total number of LUNs in such configuration does
not exceed 12.
Solution to above configurations:
Set sd_max_throttle value to 15, so that no more than 190 SCSI commands
will be queued up in UDWIS host adapter memory, even when there are 12
disk drives in a D1000, A1000, or A3X00 disk array. (15 x 12 = 180)
Following are instructions to set a maximum throttle value of 15 SCSI
commands per disk:
Note:
is the RETURN key on keyboard.
is the Esc key.
Statements between braces [....] are comments.
1. At console or command tool window, type:
su
password:[supply root password]
2. Use "vi" editor (or other editor available to you) to add one line
in the /etc/system file.
A sample session is like:
vi /etc/system
[/etc/system file opens on your window]
GA
[you are now in "append" mode at the bottom of the file]
set sd:sd_max_throttle = 15
:wq!
3. Reboot the system for the above change to take effect.
Reboot may take a long time on large systems.
For configurations with multiple A1000s or A3X00s daisy chained to one
UDWIS host adapter and supporting more than 12 LUNs on the SCSI bus,
follow procedure below:
Repeat steps 1 through 3 above, but use a new sd_max_throttle value
determined by dividing 190 by the total number of LUNs on the SCSI
bus.
Comments:
The details provided in the Corrective Action section are a temporary
workaround. A patch will be released in the near future which will
provide the permanent solution.
--------------------------------------------------------------------------
Implementation Footnote:
i) In case of MANDATORY FINs, Enterprise Services will attempt to
contact all affected customers to recommend implementation of
the FIN.
ii) For CONTROLLED PROACTIVE FINs, Enterprise Services mission critical
support teams will recommend implementation of the FIN (to their
respective accounts), at the convenience of the customer.
iii) For REACTIVE FINs, Enterprise Services will implement the FIN as the
need arises.
----------------------------------------------------------------------------
All released FINs and FCOs can be accessed using your favorite network
browser as follows:
SunWeb Access:
--------------
* Access the top level URL of http://sdpsweb.ebay/FIN_FCO/
* From there, select the appropriate link to query or browse the FIN and
FCO Homepage collections.
SunSolve Online Access:
-----------------------
* Access the SunSolve Online URL at http://sunsolve.Corp/
* From there, select the appropriate link to browse the FIN or FCO index.
Supporting Documents:
---------------------
* Supporting documents for FIN/FCOs can be found on Edist. Edist can be
accessed internally at the following URL: http://edist.corp/.
* From there, follow the hyperlink path of "Enterprise Services Documenta-
tion" and click on "FIN & FCO attachments", then choose the appropriate
folder, FIN or FCO. This will display supporting directories/files for
FINs or FCOs.
Internet Access:
----------------
* Access the top level URL of https://infoserver.Sun.COM
--------------------------------------------------------------------------
General:
--------
* Send questions or comments to [email protected]
---------------------------------------------------------------------------