« first day (2765 days earlier)      last day (2398 days later) » 

00:17
100% of languages have something similar, afaik.
well, the languages most people use anyway. =)
There is Moment.js in Javascript, DateTime in perl. Just google for date/time library in <language>
It's very common at this point.
@Wildcard if you need something else to google for, check for Olson DB support, because if it supports the Olson database, it's good and awesome.
 
6 hours later…
06:33
Morning
07:20
Morning and Evening
07:35
@EvanCarroll Thanks. I was looking for some standalone command-line tools. I think dateutils will do extremely nicely: fresse.org/dateutils
07:50
Morning
08:02
Another lovely TiDB question :
1
Q: How to implement Session Timeout in TiDB?

CocoAs far as I know, TiDB does not support session timeout on the database level. But can I use the session ID to implement session timeout in TiDB on the application level? What should I do?

X-P
Some guy in the suggested edit queue is constantly editing list items and other stuff to quoted text. You can't miss it.
08:30
@JackDouglas (Y)
Can we get Postgres 11?
@Colin'tHart yes that's coming up soon
might have another crack at DB2 after that
firebird is also on the list
hey guys
I'm guessing you're running all the non-SQL Server on Linux?
08:32
I made my colleague - boss left a real mess before leaving
he altered tables in one schema
and altered other tables on a mirror schema in another server
due to that there are severe differences between the two schemas
@Colin'tHart yes, though that's exactly what I'm thinking about DB2, maybe I'll find it easier to install on Windows instead
I would need a comparison tool that can check the two schemas and say that on the schema x there are these differences compared to X' and vice versa
I'm doing it manually but that moron, did such a sh## job that it is taking a long time
@JackDouglas Yes, possibly
I'm on postgres
@AndyK Googling postgres schema compare brings up tons of hits
Some of which are questions here or on sister sites
08:37
@Colin'tHart +1
cheers Colin
08:51
Is this worthy of a rollback and real edit? Chris Pop is IMHO doing some half-baked edits that get approved. Opinions?
@hot2use saw nothing wrong with that particular edit; agree others have been problematic
Missing capitalisation of sentence beginnings?
Just edit it again.
Yes that
Spaces where non required.
08:55
OK well fix it up and hopefully they will see what you did.
@Colin'tHart A possibility.
@PaulWhite Ok. Will do.
I've given them an opportunity to go back and review the outcomes of their previous suggested edits.
@PaulWhite edited
09:17
cheers
09:51
@AndyK Get the schema in text (with \d each table or with pg_dump schema-only) and then do a diff on the 2 files.
10:38
Hi all
Hypothetically, if I have two linked servers, both running sql server 2008, and one is upgraded to 2016, will that cause any issues with the link?
this majoritively answers my question, just want to double check experts-exchange.com/questions/28964276/…
@WhatsThePoint I suggest you post a question at our site: dba.stackexchange.com/questions/ask
@ypercubeᵀᴹ thanks, but the question wasn't my own, it was for someone else, I will relay that back to him
@WhatsThePoint Sure. I just don't know the answer. It is easier if it is posted on the site. Many more people will look at it.
Reading the SQL Server page about linked servers (docs.microsoft.com/en-us/sql/relational-databases/…), I guess you may need a different DLL (to connect to 2016 from the 2008)
@ypercubeᵀᴹ yeah that's understandable, someone dropped into the sql room on stackoverflow, thought it just do a quick ask in here before telling him to ask a question
11:00
@Colin'tHart James has added Postgres 11 to db<>fiddle. Any issues blame him not me :p
3
11:10
@WhatsThePoint that link should be flagged as span ;)
2
11:27
span?
11:42
@JackDouglas Great! Will start playing around with new features!
12:00
@PaulWhite yer
@dezso how come?
@JackDouglas Your brother?
13:00
@TomV James is my son :)
he's something like 16 years old
The update will reflect that you need to be 13 years old, unless you live in the EU, which means we require you to be 16. However, there's nothing in the law that says we have to go proactively looking for underage users, which means the policy isn't any different from what we have, essentially. If someone says "I'm only x years old" and mods suspect the user might be underage, they just escalate it to us to investigate. A lack of action on our part is an indication that we looked, and found no grounds to delete the account. More about this in the latest mod newsletterTim Post ♦ May 18 at 12:52
looking at a database where one of the schemas has a period in its name. So you end up with objects that look like This.Schema.Object. Which to the untrained eye looks like you're referencing a database named This. So dumb.
13:25
@ypercubeᵀᴹ cheers
@MaxVernon I wonder, if you had a schema called This.Schema and a view called Object and another schema called This with a view Schema.Object and then you called sp_helptext 'This.Schema.Object', which object it would find first.
@AndriyM nobody knows :-)
13:38
@AndriyM None of them?
Shouldn't you use sp_helptext '[This.Schema].Object' or sp_helptext 'This.[Schema.Object]' ?
@ypercubeᵀᴹ Generally you don't have to, if I'm not mistaken. Perhaps you must use delimiters in cases like this.
consider:
CREATE SCHEMA [.];
GO
CREATE TABLE [.].[.]
(
    [.] int NOT NULL
);

SELECT [.]
FROM [.].[.]
@AndriyM it's always a good idea to use the square brackets, but it's so much easier if you don't need to.
13:55
@MaxVernon In my scenario you must use delimiters, it appears.
CC @ypercubeᵀᴹ ^^^
I just wrote a quick blog post about it.
I found the problem in a SQL Server Agent job that was reporting Database "DataDictionary" not found.
The schema name in question was DataDictionary.ArtifactManagement
so objects were named DataDictionary.ArtifactManagement.AggregationArtifactsTable
@MaxVernon "object good names" = "good object names"?
@AndriyM or more correctly, "stupid brain" :-)
thanks for letting me know! It's fixed now.
@MaxVernon That would be unfair. A little too hasty, perhaps :)
No worries
@AndriyM and I proof-read the article too. Just couldn't for the life of me see that title !
The url has also changed too: sqlserverscience.com/t-sql/…
14:13
@MaxVernon Interestingly, the old one redirects automatically for me
@AndriyM oh, cool. Gotta love WordPress, I guess :-)
I did nothing special, just rewrote the URL.
@Wildcard I would highly suggest learning perl, and not using that. Any idiot can do it. The Begging Perl book is great. And Perl, is probably the best documented language in the world.
@Wildcard and if you knew Perl, you could actually use that stuff inside PostgreSQL which is often faster than pl/pgsql.
@jadarnel27 - I changed the domain name for my blog - does the subscription feature still work?
did it notify you of the recent post?
@EvanCarroll except you need to know perl.
@MaxVernon Yep, it came through just like nothing had changed!
@jadarnel27 sweet. thanks!
14:23
So congrats on the DNS side of things =)
yah, that's always dicey. Although I do host my own DNS servers so I have that measure of control. :-)
Learning Perl is actually very easy. It's so heavily influenced on shell it pretty much just does what you would expect and want in shell.
I'm not sure why people are so scarred of it, or hate it.
@EvanCarroll I suppose I should probably not comment on it. My only experience was a long time ago when a friend showed me it. It gives me nightmares, but then I'm a VB programmer at heart.
go ahead, laugh.
for me it's about as approachable as regex
@MaxVernon You host your own website, host your own DNS. Sounds like you have trust issues =)
2
@jadarnel27 lol. true dat.
control issues too, probably.
14:27
How do I sign up for the Max Vernon cloud service?
lol... I used to sell that.
Max-as-a-Service
@MaxVernon I started with VB/PHP, true story.
@jadarnel27 yep. it worked out pretty well.
Then I picked up perl and stopped doing both of them =)
14:28
@EvanCarroll wow, php -> assembler. That's quite the journey!
@jadarnel27 I stopped when my last customer finally said to me, "I can just host this myself, right?"
and I was like "yah, cloud is just 'someone elses computer'"
Max is not in sales confirmed.
2
=P
My school had a lot of VB programmers. I learned PHP 4 because it was thought to be an open source alternative, and I was big into open source. One of my high school friends got into Perl at that time. He dragged me a long. CPAN was amazing, and still is.
@Wildcard for what it's worth this is what you'd do in perl my $duration = DateTime->now() - DateTime->new(year => 2018,month => 01,day => 02); say for $duration->in_units("years","months","days")';
@EvanCarroll Is my an interjection in that example?
It's how you declare a variable in Perl. our is global, my is local.
My, oh my...
14:38
lol
Actually a pretty good idea. =) technically anyway.
In most ways perl is far better than other more popular languages, namely Python and C and friends. It's pretty clear in Perl what you want the linkage of the variable to be, and mistyping won't implicitly do something horrible (like in Python).
Not 100% sure of it, but I think Perl invented strict mode for a dynamic language, which is now in popular languages like JavaScript.
So for today: I'm modifying a kernel patch for Linux. Or that's the goal. This may be the first one I've done for work.
The goal is amusingly to make Windows secure.
@EvanCarroll wow, and you're going to do that in one day. Excellent!
I have a contact that wants me to patch TRESOR so it downloads a key remotely into a secure string. TRESOR is a patch that does AES encryption by using the CPU's debug registers so the key never has to be stored in RAM. This protects against a cold boot attack (where someone removes the memory and boots it up in another machine).
So the idea is to either hack a hypervisor or the kernel patch to run CURL and download a key to mount LUKS behind Windows replacing its native bitlocker
So you can run windows servers remotely, and you know that if they're removed off site the encryption can't be compromised by James fucking Bond.
@EvanCarroll interesting idea!
I use BitLocker on my machines. For my mobile machines, I use a 16+ digit key at boot that I type into the keyboard to facilitate the SecureBoot key unlock mechanism.
Right but that's still vulnerable to cold boot attacks
so you have to keep your machine physically safe
14:53
true, but that's such a small risk I'm willing to take it. If I step away, I lock my machine every single time. If someone removes ram out of my laptop while I'm in the bathroom, I'm pretty certain I'll notice right away :-)
what I really want security for is when the machine is turned off.
I know no one can boot it except me. And no one can decrypt the drive except me.
yea, I tend to agree it's not a major factor unless you're worried about a kernel-level vulnerability or exactly that -- someone popping the stick in another machine
@EvanCarroll I've always wanted to see that. Spray that sucka with nitrogen and pull it out while the machine is running. That's some cool shit.
Without Secure Lock you can actually do it on the same machine =)
Go with the smallest flavor of Linux 2.4 that supports /proc/mem
Probably takes up 15 megs the rest you can see.
the keys on my machine are stored in the TPM module; I think that makes them secure from that attack.
the pin I enter is used to unlock the TPM
15:22
cya
@WhatsThePoint the expert sexchange network has a dark history, and once was always on the top of virtually any search that involved programming. The only problem was that the answers were not freely available. If I am not mistaken, SO grew out partly from the discontent with it.
@dezso and the name choice that still brings laughs.
16:29
@ypercubeᵀᴹ I always thought it was serious
16:57
May 23 '12 at 18:18, by JNK
did you mean Expert-Sexchange or Experts-Exchange?
 
3 hours later…
19:43
@JackDouglas nice. When I was sixteen I was still cracking video game save files instead of doing something useful
Or having a beer somewhere
 
2 hours later…
21:57
Forgive me as there is not enough information here i'm just having a little freak out and wanted to chat. I tried to do a db restore earlier today (mssql08) on a production server (due to a failed database update from a vendor) and the restore went fine. However sometime later I started getting Integrity failure alerts on the database.
I didnt notice these until hours later. I ran a DBcc checkdb and it came back fine but the alerts persist. I know that I am supposed to do a full restore to address this but that is what I did in the first place and its been too long to roll back. I think the issue was there were trans logs that had not been backup. I mostly think I just need to try DBCC CHECKDB('database_name', REPAIR_REBUILD) but I don't understand the impact fully.
Did you try that stupid script by the cartoon guys?
it's pretty not-shitty.
Will certainly have a look. Thanks.
22:18
@Matt did you give it a shot?
I am trying to get on my machine at work to add the sp and give the health check a shot. Doing it now.
22:30
Apparently I cannot RDP to my system due to a CredSSP bug.... arg.. not my day.
22:43
Finally got in. Running sp_Blitz right now.
@EvanCarroll The most significant thing to come out of that would be Transaction Log Larger than Data File for my db in question. I would have thought a full backup would have cleaned that up.
Is that referring to max size or size on disk?
23:01
The database [Skyline_Data] has a 9 GB transaction log file, larger than the total data file sizes. This may indicate that transaction log backups are not being performed or not performed often enough.
I did a full several hours ago and the agent has been doing trans every hour
23:14
Hmmm... That database is also replicated. It seems replaction is not working either and it might be preventing the log from properly being cleared.
Database Name | Recovery Model | Log Reuse Wait Description
Skyline_Data | FULL | REPLICATION
I deleted the rep publication on that server for that DB. Log should be backed up in 8 minutes so I will let the schedule pick that one up.
Don't think that will fix the issue but its a problem I think I can fix so I will start with that.
Thanks so far for the help. Its getting me going
23:33
Ran another full and the log is still big. There is no Log Reuse Wait Description anymore though for now.
23:55
@EvanCarroll Holy shit. I think the integrity errors stopped. I need to redo my replication but I think everything is going to be alright. Thanks dude.

« first day (2765 days earlier)      last day (2398 days later) »