I would propose the following wording, or some variation there-of:
This question appears to be off-topic because it is about licensing. Questions on licensing should be referred to the vendor, and an answer obtained in writing from them.
I have a question on building a solution.I have a few txt/csv file having huge amount of data approx 30000000000 rows .By size they are 1.5 TB.The tables based on these data are only for lookup,I mean, there will be no update or insert or delete. So please guide me as how to plan and proceed. Sho...
Soemone is very confused about foreign keys and locks.
@ypercube Locks due to FKs are a very common issue (it's part of the tradeoff that FK's bring), and this approach to dealing with it is very common too, much like denormalising. It's done as a cleanup because that helps maintain code separation (avoids having one model deleting records from another's table), but it's not a bad idea to do as you suggest. — Synchro10 mins ago
My guess it's they're using some ORM.
and that they probably mean "deadlocks" when they say "locks"
@ypercube this relational division question is pretty much the canonical one on SO but it's tagged with mysql and postgresql. I had used it the other day to close and oracle version and I overruled because "the top answer won't work in oracle". Does that seem correct? Do you see any improvements we can make to the post to let this apply to all questions?
I'm asking you because you've got an answer on this
@bluefeet and this is one of the main reasons I stopped answering there. 95% of the questions posted are dupes are so unclear they're impossible to answer.
Oh lookie, an opportunity to use the VtC Licensing question on SO
This question appears to be off-topic because it is about licensing. Questions on licensing should be referred to the vendor, and an answer obtained in writing from them. Any answer received here is speculative and liable to open you, your organization and any product you derive from it open to lawsuits — billinkc1 min ago
@bluefeet for some things that's inevitable - complex things like PIVOT. If you point them to a duplicate they'll say "but I don't have a column x" and that sort of thing. So in a lot of those cases they need help solving a specific PIVOT problem, not a generic one.
On the other hand, how many duplicates of "A network-related or instance specific..." do we need?
@AaronBertrand What we really need is a tag for it. Mysql people get error-7230, we should get "a network-related or instance specific" tag and yes, I want it that long
I do confess that I often answered questions because it was easier/lazier than finding the dupe. And finding the right dupe is no picnic even when I'm not being lazy. Especially since many questions are closed as dupes where the answers are no longer the best answers possible on more modern versions (or maybe the accepted answer has actually been terrible all along).
I'm trying to take in ssis 2010 a big image, change the size, and set the byte[] result into a column.
Please see the picture below (sorry for the low quality). What am I doing wrong?
@AaronBertrand finding them is hard - but when we do find them we should attempt to keep them up to date so they can be used as a dup target, you know?
I have some DTS packages in SQL Server 2000. I know that DTS packages does not support in SQL Server 2012.
But, how can I create a new package or equivalent stuff in SQL Server 2012 similar to SQL 2000 DTS?
@JackDouglas I still think having a VtC reason that handles it properly would be a Good Thing™. If a particular question is easily answerable and doesn't put anyone in jeopardy by answering it, then the question likely would not be closed.
So then Licensing is addressed under VtC, Too Localised?
> not relevant to most of our audience. Consider revising your question so that it appeals to a broader audience. As it stands, the question is unlikely to help other users
for me, the close reasons aren't why we close something. They are used when I look at a question and think, man that stinks. How can I make this go away.
and prevent DBA.SE from turning into a steaming pile of carp.
Licensing questions aren't always off-topic, but licensing questions that are off-topic are always off-topic, if you follow me. A lot of problems are "too localized" but they're not closed as such because they're still interesting to solve. I think a "Some licensing questions are off-topic" close reason, with a link to the meta discussion(s) about it, would be sufficient.
If it's an option, I believe you can stop the SQL service, rename the .mdf file and start the service. The database would show up as Recovery Pending and you can delete it then. — ZabadakGalorex2 mins ago
@JackDouglas execution plans. We ask for them all the time, and it's always a pain because they have to upload them to some external site. And often they choose sites which require consumers of those attachments to register.
@JackDouglas except there are many who are behind corporate firewalls. i can't access google drive, dropbox, sky..ahem...onedrive, pastebin, etc. i have SE/SO, github, and box.com
Normally I'd agree with you on onedrive with their direct-link-not-really-direct thing but since I had pushed all of my onenotes onto it and this client blocks it, I'd do unseemly things with a host of barnyard animals to tear down this firewall
I'm trying to take in ssis 2010 a big image, change the size, and set the byte[] result into a column.
Please see the picture below (sorry for the low quality). What am I doing wrong?
HA! Just to throw in a little extra credit for this query, the [Message] column, yeah it's varchar(max) and stores RIDICULOUSLY long amounts of data. I think the max length is somewhere near 69k characters
"Hey, want to attend a brown bag lunch on Apache Spark" Sure, always happy to learn more. (receive invite) Ohhh, it's lunch time in Belgium which means 6 a.m. here
Yikes, indeed. "It ran fine 3 months ago". Yes, and your DB wasnt 16GB with a varchar(max) column that, right now has 679k as its max length (I thought it was ~60 earlier, coffee hadnt take effect yet)
not even sure then, seems to opinionated for me. Well, at least my opinion is that this is going to end badly regardless of the approach taken for their problem..
I want to check SQL server database is modified or not. Reason to check database is modified is for backup(sql database backup) purpose. So I will be doing a full backup first and if any data inserted, modified or any structural or any changes done then I have to identify that there is some chang...
I am new to Databases. Can someone, please, give me a clear and concise understanding of DB table normalization? I keep hearing and reading that tables should not have redundant data. But is it not possible to have a table named "Customers" and a field named "State" and have 'Michigan' repeated f...
There is a SQL file which contains some Transact SQL statements in it and some plain table queries as follows:
IF NOT EXISTS (SELECT * FROM [dbo].[SYSTEM_PROPERTIES] WHERE SYS_PROP = 'ABC')
BEGIN
DECLARE @SYS_PROP_ID INT;
INSERT INTO SYSTEM_PROPERTIES (...,....,...) values ('...','...','...');
S...
@Lamak ANT, another neat tool, was a build thing I used back in 2000 for making builds in java. I spent far too much energy helping on their email list
@Zane in theory what you'd see is a drop in PLE. you'd like to think that your BP is filled with pages already, right? with all the splits you'd keep (roughly) the same number of pages in memory but they'd get flushed to disk and new pages would come in. i think that sounds right
The reason is your identity column
Try this.....
SELECT Fishtype ,
[3] AS [3 Years Old] ,
[4] AS [4 Years Old] ,
[5] AS [5 Years Old]
FROM
(SELECT Fishtype, AvgLength, FishAge_Years
FROM dbo.AverageFishLength)t
PIVOT( SUM(AvgLength)
FOR Fis...
@Zane I thought that the impact would be larger too, however I suspect that a lot of the page splits happened in the middle 70-80% of the workload, and toward the end those pages just finished getting filled up (not every insert leads to a page split, unless you're constantly correcting fillfactor by defragging after every insert). Still, I am surprised that the clustered index was fragmented 99% but the buffer pool usage was only 10% higher.
SQL Server works in mysterious ways. I also found it surprising that BIGINT required such a small amount of extra space, even without compression. I guess it's all relative - in a two-int-column table it would have much more impact.
@AaronBertrand @Zane was talking about whether or not guids as a PK would inflate the number of pages in the buffer pool because of all of the splits. i didnt realize it was in response to your blog post and i generically said that it'd drop PLE because of the number of pages getting flushed out and read back in.
None of these posts are ever one-parters. But if I tried to cram all variables into one post it would take a year to write and nobody would ever read it all.