« first day (5161 days earlier)      last day (2 days later) » 

04:27
It's never occurred to me to ask a LLM to write code for me, except as an exercise in curiosity. I mean, I can just write it myself. I do find it extremely useful for answering questions about things I don't have expertise in.
 
2 hours later…
06:03
Wordle 1,342 3/6*

⬛⬛🟩⬛⬛
🟨🟨🟩⬛⬛
🟩🟩🟩🟩🟩
 
3 hours later…
09:19
Wordle 1,342 3/6*

⬛⬛🟩🟨⬛
⬛🟨🟩⬛🟩
🟩🟩🟩🟩🟩
 
3 hours later…
12:35
I have the same problem! How did you fix it? — Louise Moore 5 hours ago
That's the neat part…
12:56
You never gave me back any information so we never got to solve the issue - did you ever do more digging and figure it out?
@Zikato I only converted that non-answer to a comment so the question would be revisited
If it can't be answered now, we should probably delete it
13:20
a whopping opping
Definitely got opped hard by @Zikato
Sep 6, 2024 at 18:53, by Zikato
@SeanGallardy cannot anymore. A teammate removed the DB from the forwarder and the secondary and that fixed it. I'm still interested in what's causing this though.
I guess we'll never know then :/
That doesn't feel like an answer. I'm wondering if I should at least post this as a comment to help the other guy
But I don't want to comment
By chance... were there a lot of database files in that database or were the paths different?
13:27
I cannot even recall which DB it was
September was lifetime ago
forward progress
look at you go
Erik accidentally typing here instead of the OF DMs
13:29
this is like watching congressional testimony
You're not wrong there
Also, don't apply 2019 CU31 if you're running ReadScale or External AGs
what are even either of those things
Nothing anyone should concern themselves with...
13:52
@Zikato You can do that, but it sounds like this question should go
Wordle 1,342 3/6*

⬛⬛⬛🟨🟨
🟩⬛🟩🟩🟩
🟩🟩🟩🟩🟩
@SeanGallardy by "external" you mean non-Microsoft cluster? i.e. Linux?
 
2 hours later…
15:49
@HannahVernon Correct, External = Linux + pacemaker/corosync
16:05
@SeanGallardy Not the final CU after all?
16:20
i want a special prosecutor assigned to this
16:37
Perhaps we'll get a CU31 refresh
Or CU31 R2
Maybe 2019 SP1
not to disappoint anyone, but i was not actually drinking coffee (or anything) during this recording
16:54
Well, that's disappointing
False advertising
perhaps some redemption can be found in the content
i don't really remember what i said
@ErikReasonableRatesDarling 0.o
@PaulWhite Who knows :/
17:13
what are these faces about
Accurate picture of my face made with ascii art
i understand the intent
i'm asking about the motive
I guess just due to many people also sending them to me today
paying it forward
17:35
@ErikReasonableRatesDarling are you ok? I just had an alert
> Eventually you run out of things to do, even in New York City
That's the best quote from the video
17:51
you do have to work pretty hard
 
2 hours later…
20:11
Kendra's issue descriptions for issue #515 are pretty funny
> I would like for someone else to write this code because i am too lazy
> Describe alternatives you've considered
> I considered doing this myself
> Are you ready to build the code for the feature?
> Sadly, not anytime soon
20:53
I don’t blame her
The code was quite tedious
I encountered a helpful error message along the way telling me that FORMAT Is not a window function and does not support the OVER clause
context?
21:19
I see. Looks like someone doesn’t follow my GitHub repo.
Really shows how much you care.
21:45
@ErikReasonableRatesDarling Who is it? Don't leave us hanging
the transcript tells all
22:11
one transcript tells only truth
the other only lies
you must discern which is which if you want to Merge the Code
> I would like QuickieStore results to group by query hash so that queries with literal values can be ranked as a mob rather than individuals
22:48
i did a gr8 job
23:05
I would have closed the issue as due to a lakh of parameterisation
Also, the request is unreasonable given that forced plans ruin query hashes
i may add a note about that somewhere
the request comes from having to deal with a turd party app called looker that can't do parameters being run on a readable secondary that can't have forced parameterization set independent from the primary
so there we are
23:20
local factors then
yeah
but i run into enough similar situations that it's worthwhile to have
Sounds like they should enable trace flag 144 on the secondary
gr8 flag with two 4s
23:38
woah
that is a good number of fours
> Newsequentialid (Histrory/Benefits and Implementation)
a 19 year old typo
Preserved for all eternity or until Microsoft randomly deletes it for no good reason whatsoever
well
i do hope i can read through these articles
i don't think i've come across this blog before
probably explains why i'm so bad at sql programmability & api development
You're kidding
The plan cache series has been often referenced in here and on the main site
maybe i've seen those posts
I suppose it looks different in blog archive format?
23:45
i don't recall ever going through the blog post by post though
There's a huge amount of good information in the archive blogs
Such a shame Microsoft tells search engines not to index those pages
yeah, i only have a few of them bookmarked
I have a whole Edge Vertical Tab Group™ dedicated to them
There's no good reason for forced param not to be settable on secondaries
Trace flags are so awesome, and AGs are so sucky
Also cloud sucks for not supporting TFs (in general)
> I sincerely apologize for the earlier misquote—I got carried away with inference and crossed into invention.
I can't be mad at Grok when it corrects itself like that

« first day (5161 days earlier)      last day (2 days later) »