| Nick Elson 2005-07-11, 11:23 am |
| Two or three (or four) observations that may help:
1 - The evaluation of CURRENT TIMESTAMP is done before
any joining or materialization of row data is performed. In
your case you will note that all result set rows have the same
value for this expression. What this means, it is entirely
possible that sa_conn_info( ) will capture (albeit slightly)
newer times for queries submitted during the execution
of your query. [I actually captured this case in one test
here, by simply running the query repeatedly.]
2 - When I run your query once, then change (reverse) the system
clock (by, say, 1 hour) and run it again, the current timestamp
tracks the system clock, but the pre-existing LastReqTime values
do not change. As such I can reproduce your pattern this way.
So it is possible that your system clock reversed itself;
possibly
due to
2b - The more recent entries are only 21 or 22 minutes off. Is it
possible you just had a clock synch update from an extern
source that has not caught up for a while.
3 - I do notice you have numerous 2 day old connections that have
not executed anything recently. That would indicate your server
is running -ti 0 (or some huge number). It may be possible that
setting is affecting the accuracy. It may also be an indication
that you may have been off more a day [or more] on your
system clock at some point.
If this is just a one-time occurrance then it could be some combination of
all of points 1, 2, and 2b. If it usually only off withing a second or two,
I
would suggest observation #1 is common and normal. If this is always
10's of minutes or days off then you may want to look into -ti and other
such switches and maybe submit a bug report.
As to doing your detective work, I would assume LastReqTime is accurate
enough that your most recent values are the most recently active [unless
your system clock is jumping all over the place, all of the time]
"John Davey" <john. davey@automationpart
nership.com> wrote in message
news:42d244c5$1@foru
ms-2-dub...
> SQLAnywhere server v9.0.2.2551 on Windows XP
>
> I am trying to diagnose some odd application behaviour, but in doing so I
> have noticed something a little odd about what the ASA server is reporting
> about my active connections. The attached text file shows that the
> 'lastrequesttime' reported for some of the active connections is actually
> ahead of the current time (as reported in the same query by 'current
> timestamp'). Is this behaviour familiar to anyone? Any suggestions about
> what is going on? I can't at the moment see any reason why this would be
> causing the application problems I am trying to understand, but it doesn't
> help in my 'detective work' to not really know which connections are
> recently active.
>
>
> Thanks,
>
> John Davey
>
>
>
|