« first day (1902 days earlier)      last day (2974 days later) » 

5:05 AM
If you're going to prefix your column names with Hungarian-type notation, at least make use of it!
0
A: Why is my EXISTS query doing an index scan instead of an index seek?

Paul WhiteIf you want good results from the query optimizer, it pays to be careful about data types. Your variables are typed as datetime2: DECLARE @OrderStartDate datetime2 = '27 feb 2016'; DECLARE @OrderEndDate datetime2 = '28 feb 2016'; But the column these are compared to is typed smalldatetime (a...

 
 
2 hours later…
7:15 AM
@PaulWhite I saw a (new to me) trick at SQLSat yesterday I think you already know, but since I remember you asking somebody how he made that fancy stack trace screenshots (with visual studio) a while back you might not know
Walking the stack with XPerf and WPA, it looked really neat
 
@TomV Yeah, it's handy for an overview, but I rarely use it because I prefer more detail. Different tools for different jobs of course. And Xperf/WPA is easy, so there's that.
 
@PaulWhite I was going to play with it if I get around to it but what details do you miss in WPA?
maybe because it's sampled I can see what you mean, but I was thinking about using it on an environment with high SOS_SCHEDULER_YIELD wait types to see what the beast was doing
 
@TomV It's a summary. Most often I am concerned with the sequence of operations, and details like registers, memory contents etc. at a particular point in the execution.
@TomV Sure. That's what it's aimed at. So it's good to use it for that sort of thing.
None of the tools is perfect on its own though. And the real trick is interpreting what you see, and what to do to fix it.
@TomV That was Martin Smith IIRC.
He was using VS direct.
XPerf/WPA is more the territory of that British dude who's name escapes me for a moment.
Oh Adkin, Chris Adkin.
 
@PaulWhite I'll ask you for support of course
 
Ha. Of course.
My rates are very reasonable.
It's cool these types of tool are available and work with SQL Server.
 
7:32 AM
@JamesLupolt Apparently he even walked past the police office for kicks: pressreader.com/uk/sunday-mirror/20160320/281883002459101
@PaulWhite Yes I suppose you're bleeding money by answering over here :p
 
@TomV Well 0$/hr is reasonable isn't it?
I would hate to have to start paying people to answer their questions.
 
I meant all that lost revenue by helping people for free when you could be billing others
 
@TomV Yes I know, I was being daft.
 
@PaulWhite I'm still slow, I only had 1 coffee
 
That's OK. Everything works better when explained at great length.
And you're excused anyway since (a) I gave no indication I wasn't serious; and (b) it's Sunday.
 
 
1 hour later…
8:45 AM
@PaulWhite You'll regret asking for that, but there you go
my ugly mug in NYC
You'll be wishing you still saw the green triangles on your screen though
@Phil By popular demand ;) It should show up in a while
 
@TomV Ha! Excellent!
 
9:15 AM
@PaulWhite Did you forget your sarcasm tag again? :)
 
@TomV No, I'm genuinely pleased to see the passing of another default avatar.
 
I was just being daft
 
 
1 hour later…
10:32 AM
@TomV You dont look like 30kg rounder as you have said earlier. :)
 
@TomV I went looking for details on how to do that in VS recently but couldn't find anything. Do you know how?
How was SQL Saturday, btw? I guess that was SQL Sat Belgium yesterday?
 
 
4 hours later…
2:41 PM
@JamesLupolt For a first time organization it was fine
much smaller than Utrecht but I guess 180 attendants, and well organized so I don't complain
@JamesLupolt No I haven't taken any stack traces myself, It was incidental in a prez about execution plans where he showed in an XPerf session most time was spent in a function, but the presentation wasn't about walking the stack
I googled during the break and best I found so far was this and some posts on exadata
 
2:58 PM
@TomV ah yes, I know how with xperf but not vs
Going to Adkins' precon at sqlbits btw
 
8 hours ago, by Paul White
@TomV That was Martin Smith IIRC.
It appears Martin knows how to with VS
 
@TomV I'll send him an email
 
 
2 hours later…
5:07 PM
@JSapkota You're too kind :)
 
 
1 hour later…
6:21 PM
rant time
why oh why does a regular editor come up with a nearly unreadable answer?
0
A: Converting standard_conforming_strings from off to on when importing Postgres dump

Md Haidar Ali Khan@MateVarga, As PostgreSQL documentation Here If the configuration parameter standard_conforming_strings is off, then PostgreSQL recognizes backslash escapes in both regular and escape string constants. However, as of PostgreSQL 9.1, the default is on, meaning that backslash escapes are recognized...

 
On the plus side, it's rich in formatting. (It's a good thing, right?)
 
Aren't most of his edits a hunt for unicorn points?
You can't really reject them but they aren't great either
 
6:42 PM
I do reject some of them, others are okay. I'm curious if he manages to get to 500 edits before getting to 2000 rep.
Meaning I'm curious if he's aware that edit suggestions after 500 edits aren't rewarded.
 
"added tags based on the subject matter"
 
7:15 PM
I've only just realised that the principal part of that answer is a citation from the manual page which he's linking to immediately before that part (starting with "If the configuration parameter…" and ending with "…write the string constant with an E."
My issue with that is that the formatting doesn't reflect that's a citation. I think the convention is to use a quote block. Should it be fixed? Should we perhaps suggest he fix it himself first?
Given it a shot:
The principal part of your answer is taken verbatim from the manual page you are linking to. It's conventional to use the blockquote formatting in such cases. Please consider editing your answer to outline the citation. — Andriy M 59 secs ago
 
 
3 hours later…
10:27 PM
This was just migrated from us to SO:
-1
Q: Mongo in PHP is deprecated?

rakibtgMessed-up with the new MongoDB documentation and can't understand how it going to help, any help would be appreciated. As an example, i used to use like this: $con = new MongoClient(); $db = $con->hasan; $friends = $db->friends; $data = [ 'name' => 'Xyz', 'age' => '21', 'sex' =>...

It seemed to me very unclear and I voted it as such, but it had had for votes to migrate already and my last vote actually caused it to migrate, even though it was a different kind of vote.
It's not very nice that we are throwing rubbish over the fence, figuratively speaking, but just so that others can learn from my experience, when a question already has four migrate votes, the fifth is probably going to migrate it regardless of whether it's also a vote to migrate or a different kind.
No idea if this is expected behaviour but this is how it currently works anyway.
 

« first day (1902 days earlier)      last day (2974 days later) »