Feedback on 13-253 Environmental Conditions log page

Ballard, Curtis C (HP Storage) curtis.ballard at hp.com
Wed Apr 16 09:42:15 PDT 2014


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1404162_f.htm">HTML-formatted message</a>

I've been in support of allowing reporting multiple sensors as we have a
number of products that have many sensors.
I'm concerned however that this is getting excessively complex for the
minority use case.  Most products only really need to report one sensor in
standard data.	Products that have multiple sensors already report the low
level detail in private engineering level logs.  It isn't hard to have a
single sensor report an average or worst case as necessary and that is good
enough most of the time.
Unless a really clean method can be found for allowing multiple copies of a
single set of information we should consider whether the complexity is
justified.  The approach Kevin mentioned where only a single sensor is
defined but log subpages can report identical sets of information for other
sensors may be better than trying to squeeze everything in a single page.
Curtis Ballard
Hewlett Packard
HP Storage
Fort Collins, CO
(970) 898-3013
From: owner-t10 at t10.org [mailto:owner-t10 at t10.org] On Behalf Of Gerry Houlder
Sent: Wednesday, April 16, 2014 8:24 AM
To: T10 Reflector
Subject: Re: Feedback on 13-253 Environmental Conditions log page
Ralph, I should have stated "one set of limits", not thresholds. You are
correct that log thresholds are not in play here and we should try harder to
avoid that confusion.
After thinking about the task of adding a separate set of limits for each
temperature and humidity parameter, it could be difficult to come up with
wording to describe which limit parameters are associated with which
temperature/ humidity parameter. A solution to this is to redefine the
grouping so that a temperature reading and the four parameters related to
that temperature reading are grouped into a single log parameter that is 20
bytes (4 byte header, 1 byte reserved, 1 byte temperature, 4 bytes critical
temp limits, 4 bytes operating temp limits, 2 bytes lifetime temp, 2 bytes
temp since power on, 2 bytes reserved to round to a 4 byte boundary). Is this
the direction to go with this?
On Tue, Apr 15, 2014 at 7:34 PM, Ralph Weber
<Ralph.Weber at wdc.com> wrote:
All of the log parameters defined in 13-253r3 are binary lists. Log
thresholds do not apply to this kind of log parameter.
________________________________
From: owner-t10 at t10.org<mailto:owner-t10 at t10.org>
[owner-t10 at t10.org<mailto:owner-t10 at t10.org>] on behalf of Gerry Houlder
[gerry.houlder at seagate.com]
Sent: Tuesday, April 15, 2014 2:30 PM
To: T10 Reflector
Subject: Re: Feedback on 13-253 Environmental Conditions log page
On Tue, Apr 15, 2014 at 1:58 PM, Kevin D Butt
<kdbutt at us.ibm.com> wrote:
Gerry,
I have received some feedback related to the environment conditions log page
proposal http://www.t10.org/cgi-bin/ac.pl?t=d&f=13-253r3.pdf
(1) Using the parameter code from 0000h to 000Fh, we can select a certain
temperature sensor from up to 15 temperature sensors. Using the parameter
code from 0020h to 002Fh, we can select a certain humidity sensor from up to
15 humidity sensors. But there is no way to know which sensors are
valid/active or not.
I think it is useful for the user if we have another parameter code that
returns the valid/active temperature sensors.
GH: The intent that only the number of sensors that are supported will be
listed. For example, if only two temperature sensors are supported then only
parameter 0000h and parameter 0001h will exist in LOG SENSE parameter list
and there will not be any parameter numbers 0002h through 000Fh. This is why
the support requirements say "at least one" or "optional". Anything listed as
optional is not guaranteed to be implemented.
(2) The log page allows us to manage up to 15 sensors for both temperature
and humidity. Each of them can be attached to the different location. For
example, we can check the DRAM temperature and head temperature
independently.
But the Critical Temperature/Relative Humidity Limits, Operating
Temperature/Relative Humidity Limits, Lifetime Temperature/Relative Humidity,
and Temperature/Relative Humidity Since Power On has only one entry. The
criteria of these threshold must be different depending on the target object.
For example, DRAM might have tolerance to 80 degree Celsius but the tape film
might be melted. The hardware developer usually allocate different criteria
to each sensor(refer to http://www.thinkwiki.org/wiki/Thermal_Sensors. When I
work with ThinkPad development team, they have different threshold to each
sensor). So these entries should be assigned to each sensor independently.
GH: discussion to date had indicated that one set of thresholds, applied to
all of the sensors, was sufficient. i will consider this change if discussion
at the next CAP meeting is in favor of the change.
This makes a lot of sense to me.  I suggest that a good solution would be to
add a parameter in Log Page 0Dh (i.e., existing Temperature log page) that is
a list of sensors supported.  Each subpage could be for a different sensor
number (i.e., subpage 01h could be for temperature sensor #1 and humidity
sensor #1; subpage 02h could be for temperature sensor #2 and humidity sensor
#2).  Parameter 0000h of each subpage could be a validity flag.
Alternately to putting the list of supported subpages in the Temperature log
page, subpage FFh could be used to provide the list of supported subpages.
I know this is a major change, but I think it is a much better solution.
Thanks,
Kevin D. Butt
SCSI Architect, Tape Firmware, T10 Standards
Data Protection & Retention
MS 6TYA, 9000 S. Rita Rd., Tucson, AZ 85744
Tel: 520-799-5280
Fax: 520-799-2723 (T/L:321)
Email address: kdbutt at us.ibm.com
http://www-03.ibm.com/servers/storage/



More information about the T10 mailing list