I think providing some advice on formatting would be useful. I'm thinking that we could at least update this FAQ question to request text, and further SQL for test data.
Formatting
When possible do not include pictures where text would otherwise do, and even better than text is the necessary SQ...
Anyone want to look into it? The questions aren't total shit in that someone else can use them, and the presence is probably mostly harmless but I'm fairly sure they're the same person. The questions are just to similar not to be sufficient evidence of a sock.
I mean, a lot of the questions seem questionable. As if they were asked by a script kiddie from the edge of the autistic spectrum. Perhaps they're NSA. Asking questions about parking, zipcodes, and wifi hotspots. I suspect it's Snowden's replacement.
We've got two questions on essentially ranges with WHERE (as in BETWEEN) -- basic stuff. And, both use boroid in the sample. Both are by new users. Both of the new users have 'C.' as their last name.
It currently has 4 upvotes and one comment from @TomV (thank you) and was wondering what my next steps might be.
Should I migrate the question from SF to DBA.SE and hope for the best or should I put some bounty on the question over on SF? What would the "oldies" recommend?
Well @JamesLupolt 's response would basically be the solution, if the windows server's IP address is the lowest in the range of IP addresses assigned to all the other SQL Server IP addresses.
Suppose network interface 1 (net1) and 2 (net2) are both connected to the Internet. I want application 1 (app1) to use net1, and application 2 (app2) to use net2. How can it be done under Windows?
If you need a concrete example of the above situation, consider:
A laptop with Windows 7
net1 = A...
@hot2use A bounty on SF seems worthwhile. You question has low views, and there are only a small number of bounty questions there right now. And yes the bounty offer here still stands.
@TomV you wrote in the transcript that "Meaning, if you could instruct the OS of your SSRS server to route your outgoing connection to SQL over a specific outgoing ip address/NIC SSRS wouldn't notice "
The issues is that I could have potentially 1 Server with 15 SQL Server instances. Each of those 15 SQL Server instances could be hosting a Reporting Services instance
@TomV I can have a windows server 168.xxx.xxx.70 with a SQL Server instance on it 168.xxx.xxx.71 which itself has SSRS installed. I can then have another 14 instances (168.xxx.xxx.72 - 168.xxx.xxx.85) all running on the same Windows server. So I have one windows server with 15 SQL Server instances and all of them could be providing SSRS Services.
The issue is when the Reporting Services connect to a different SQL Server say the 162.xxx.xxx.52 SQL Server instance.
It's the same. The SQL Server instance on .72 with the master, model and msdb database is also home to the SSRS instance with the ReportServer$APPL1 databases.
BUT (and that is a big but) the SSRS/SQL Instance has no USER databases holding data that the SSRS instance will retrieve data from. The data for this standalone SSRS/SQL instance (on .72) is retrieved from a different SQL Server instance (on .52).
That's why I currently (need) have two firewall rules from the .72 instance, because I am unsure if the SSRS connection will go out from the SQL/SSRS instance specific IP address .72 (to .52 on 1433) or from the server address .70 (to .52 on 1433).
Basically we have created a SQL/SSRS instance to house only the SSRS relevant database and functionality, but the data retrieved for the reports could be anywhere (in the example on the .52 server).
That is one of the restrictions I have in our environment "....I am not interested in feedback suggesting that having SQL Server and SSRS on the same server would resolve my issues. I know that and would gladly do it but for the third-party software required to run together with the SSRS components."
I agree that my question is one real MF of a question and through the discussion here, I am seeing potential to tweak my question slightly here and there to increase the quality.
@PaulWhite ... or do you see through my question from an SSRS perspective?
@dezso looking at that answer you edited, I've seen that mentioned before, but does PostgreSQL really run better on *nix than on windows, or is that a myth?
PostgreSQL will definitely run faster on Linux than on Windows (and I say this as one of the guys who wrote the windows port of it..) It is designed for a Unix style architecture, and implements this same architecture on Windows, which means it does a number of things that Windows isn't designed ...
@PaulWhite what is interesting that people are using servers as desktops. In my time a 4 CPU Sun Fire V880 was a giant, with enormous amount of RAM (we had maybe 16 GB)
@AndriyM once I made the thing into a mobile computer, putting all the keyboard, monitor, mouse and UPS on top of the V880 and pushing the whole thing (it had wheels) into the other room
I had a borrowed Texas Instrument for programming BASIC, but I never made it that far. Can remember typing page after page of ASSEMBLER code into a C-64 compiler to create games
@TomV That row goal question probably is a duplicate. The two solutions in the linked Q & A are probably the most general solutions. The IF EXISTS sets a top-level row goal otherwise, and there's very little to be done about it. No query hints allowed, trace flag 4138 ineffective. You might get lucky with a rewrite, but the plan would still be based on an error, and could change at any time.
And I say that having already undone Kin's gold hammer close as dupe :-/
Oh I just noticed the OP has started a bounty on the linked Q & A.
@TomV Well I would have looked into the specifics of his case a bit more, but the scripts don't run. The index defs are missing; the collation isn't specified (it's SQL_Latin1_General_CP1_CI_AS by trial and error) and the same table seems to have been scripted twice (as sometable). Anyway, too hard.
@PaulWhite That's what I suspected, but then I don't get how removing the not exists solves his issue
and indeed I couldn't get the repro to work before I ran out of time
The only thing I can think of is that removing the not exists somehow removes his nested loops (since in the comments he states it still doesn't go parallel)
but we can't verify, and I asked him in the comments to add the resulting plan after that change but he didn't
and a second thing I'm confused about is why he (we?) would get a nested loops join for a not exists, since "not exists" kind of suggests "there is a good chance you won't find any matches" to the optimizer (the way I look at it)
Well as to why my suggestions solves his problem I still don't understand, but I think we never will until we get a working repro or the resulting exec plan
It's not helped by the fact the optimizer checks for a match on source in the NL ASJ, rather than using any index that may be on smalltable. Scanning 952 rows doesn't take very long, unless you do it a lot.
@TheGameiswar could you please provide a working repro for us to play with (hosted on something like pastebin and not dropbox). I'd like to play with this a bit more since I'm still not sure why my suggestions works for you and I'd like to improve my answer after running more tests and hopefully getting some empirical evidence — Tom V26 secs ago
At my old place they still have a replication for intranet access/author permissions, which involves Oracle UCM and additional user tables synchronizing with AD. If it fails (and it does) you have to re-create the basic permissions for all the users.
I found it generally OK when everything was just running along. The restrictions, odd bugs, and corner cases were a pain from time to time, as was upgrading across multiple sites with much scripting and hoping for the best but I really started to dislike it when it broke suddenly and debugging/recovering was necessary in a very hot, live environment. Too much drama.
@TomV No, but how many questions have you seen that say "not working". The typical response is...what's not working/what error message do you get/etc. Just seemed funny to me.
@PaulWhite There are too many errors to document in a comment :)
I was suggesting "test them before you post please" :)
@PaulWhite Hmm, I saw a "move to chat" link, but it only moved the discussion between me and the OP to chat, it didn't move the rest and didn't delete anything
I thought it usually said "continue in chat", so I was wondering what would happen if I clicked the link :)
We're standing up transactional replication here to keep a second database in sync with the primary for N tables. The target is 2 second latency and sometimes we hit it, other times we're very, very far from the mark
My distributor is done running gooderer https://t.co/RC8zMmwgiU
Thanks for helping me out!
Although our developer is quite versed, we found our database to have an user account related calculation that inflated itself.
So keep that in mind when facing similar problems!
Cheers
dns43
We are running SQL Server 2014 SP2 CU1. I've run into an issue where it looks like my temp table should be dropped in my procedure so that I can recreate it but SQL throws an error saying my temp table is still there?
I can of course name the temp table to something else but I'd like to know ...
@hot2use As far as I know, any new connections from SSRS would go through the source address selection rules described here and use the IP address that is closest to the next hop (the default gateway in this case) for connections to the Database Engine on the other server.
So in your case that would be the .70 address, assuming the default gateway is 255.255.255.[something less than 70].
You could try setting SkipAsSource for the .70 address if you want the server to make new outbound connections from .71 instead. I don't know of a way to make that happen for only SSRS though.
Other than the SkipAsSource case or someone changing the gateway or netmask, I can't think of any cases where the connection to the Database Engine would come from .71 instead of .70.
Sorry, I meant to type "assuming the netmask is 255.255.255.xxx and the default gateway is something less than 70."
So I'm surprised both of these rules are necessary:
@billinkc I guess it really depends on how you mean it. I mean thankfully most of us are all Self-D'ers, I mean medically some aren't. But if the self means "on self" well then that's a different matter
I'll always remember a call with a PSS rep about 8 years ago. It was tier one and English was a second language so the way he asked us to send a memory dump on the phone came out "Please. I will need you to take a dump and send it to me"
I am a newbie for administering sql server. I am comfortable with SQL language and making SIS packages. I want to migrate data from SQL Server 2005 to 2012. My question is do I need to worry about the system databases and other objects such as indexes, stored procedures, view, security and per...
Is that really a duplicate? The question is specifically about system databases and the other answer doesn't head there? Don't really care just felt a bit non duplicate to me. Though not a great question all the same.
I also wonder what most folks do. I pretty much NEVER restore master or model or msdb when going up versions. I always migrate via scripts. I'm not saying Max is wrong, I just wonder if I'm in the minority or majority here.
The suggestion for UUIDs is fine as an alternative. +1 But what did you want that subjective "inferior ones like ..."? To piss whoever uses them? -1. (no vote from me, either up or down) — ypercubeᵀᴹ19 secs ago
not for the collection of hats in itself since that's gone after a few weeks, but for the challenges that help understand the site, or clean up the site for many people and the improvement it often brings to the network
Also, if you are using 2016 and notice schema comparisons are slow but improve with the legacy cardinality estimator turned on, you might need to update statistics on system tables github.com/sjh37/….
We were running into a similar situation here (without the insight to try the cardinality estimator switch) and I ran that this morning and things are much better
Paul saved the conversation pretty nicely by switching the topic to random sports
and we were testing if kick muting made the DONK sound to the one being muted IIRC, but that wasn't on me (fortunately, I might have gotten a reputation with the non-resident mods)
I think providing some advice on formatting would be useful. I'm thinking that we could at least update this FAQ question to request text, and further SQL for test data.
Formatting
When possible do not include pictures where text would otherwise do, and even better than text is the necessary SQ...
Suppose network interface 1 (net1) and 2 (net2) are both connected to the Internet. I want application 1 (app1) to use net1, and application 2 (app2) to use net2. How can it be done under Windows?
If you need a concrete example of the above situation, consider:
A laptop with Windows 7
net1 = A...
GATE COMPUTER SCIENCE & INFORMATION TECHNOLOGY, DATABASES QUESTIONS & ANSWERS
http://gate-exam.in/CS/Syllabus/Computer-Science-Information-Technology/Databases
Thanks for helping me out!
Although our developer is quite versed, we found our database to have an user account related calculation that inflated itself.
So keep that in mind when facing similar problems!
Cheers
dns43
We are running SQL Server 2014 SP2 CU1. I've run into an issue where it looks like my temp table should be dropped in my procedure so that I can recreate it but SQL throws an error saying my temp table is still there?
I can of course name the temp table to something else but I'd like to know ...
I am a newbie for administering sql server. I am comfortable with SQL language and making SIS packages. I want to migrate data from SQL Server 2005 to 2012. My question is do I need to worry about the system databases and other objects such as indexes, stored procedures, view, security and per...
