Search Results

Search found 14951 results on 599 pages for 'connect by'.

Page 362/599 | < Previous Page | 358 359 360 361 362 363 364 365 366 367 368 369  | Next Page >

• Kerberos: connection from win app running from IIS to SQL failed

  - by Mikhail Kislitsyn

  I have an IIS web-application with Windows authentication and impersonation. This application connects to SQL server. In this case Kerberos works fine. But there is a problem. Web-application runs windows application (not .NET), which also connects to the SQL server. Windows application runs with IIS app user credentials and impersonates current site user to connect to SQL server. scheme: http://i.stack.imgur.com/2cgv7.png When delegation for IIS user is set to "Trust this computer for delegation to any service" everything works fine. But I can't use this type of delegation according to security requirements. When I set delegation to "Specific services" and choose MSSQLSvc SPN, connection from windows application fails with "ANONIMOUS" fault. WireShark shows "KRB5KDC_ERR_BADOPTION" packet. What I'm doing wrong?

  Read the article

• Bash window title not restored after exit

  - by forthrin

  I have a problem with the window title in the Terminal window on OS X: Start Terminal. Window title is "bash" Type "ssh external" to connect to an external server. Window title is "user@external:~" Type "exit". I am now back at my local machine, but the window title still says "user@external:~". How do I make the window title return to "bash", which I assume would be correct since I have logged out of the external server and returned to my local machine? My ~/.bash_profile has the same PS1 value: export PS1='\w$ '

  Read the article

• Is it possible to hide folders/subfolders from users based on permissions?

  - by Uwe Keim

  Having a Windows Server 2008 R2 that has a share with lots of nested folders, I want to be able to only show certain folders to certain AD users/AD user groups. Is it possible to configure the permissions on single folders, so that clients that connect with Windows XP/Windows 7 to the share on the Windows 2008 R2 server only see those folders for which they have "view" permission? Other clients should not see the folders at all in Windows Explorer. I was told that this seems to be a standard feature on Novell networks.

  Read the article

• Fixing SharePoint 2010 Permission Problems on Windows 7

  - by Ricardo Peres

  I had a tough time trying to have SharePoint working perfectly on a Windows 7 development machine that was occasionally disconnected from the Active Directory (when I am home I must connect through a VPN). I mostly had problems with service applications such as User Profile, Managed Metadata, Business Connectivity Services and the like, and all I knew were cryptical messages such as “access denied” or “the service or application pool is not started”. I was sure that both the services and application pools were running under a domain account that had proper permissions on the SQL Server instance, and basically it was a fresh installation. Lots of people are having the same problem, apparently. After banging my head against the wall for several days, I remembered about farm (what I had) versus stand-alone (which I had never tried) installations. Bingo! Here’s what I did: I dropped all SharePoint databases and logins and reinstalled SP from scratch, only this time not in farm mode, but as stand-alone. After the SharePoint Configuration Wizard started, I cancelled it and started the Management Shell. I created the configuration database manually by using the New-SPConfigurationDatabase cmdlet where I specified a local account – something that the Configuration Wizard wouldn’t allow me to do. Then I restarted the Configuration Wizard and everything began working perfectly! Yes, I got some pre-configured service applications and also some content which I didn’t need, but I realized it was possible to drop and recreate everything the way I wanted to. All services and application pools are now running under local accounts, which is fine for my development needs. Really, Microsoft… I hope this will bring light to someone facing the same problems!

  Read the article

• Windows - Delayed Write Failed error on USB hard drive

  - by ndngrd

  I've got a new Verbatim 1.5TB USB hard drive (Samsung HD154UI) and I'm finding myself completely unable to fill it. I'm using Windows XP. Whenever I try to copy a load of files over, it works for some time (will copy over between 20 and 90GB) but eventually stops with an error saying "The specified path is too deep" - the specified is not too deep, there's nothing more than 2 dirs deep that I'm copying. A balloon pops up at the bottom saying "Windows - Delayed Write Failed" telling me the data could not be copied. This wouldn't be too bad if I could just restart the transfer, but after this error has happened I can't write anything else to the disk - including if I eject it and then connect it to another machine. It just seems completely locked. The only way I can unlock it is to delete everything that I was copying to it. I've tried various USB cables and copying from different machines, and the same thing keeps happening.

  Read the article

• How can I force Parallels' networking to obtain an IP through a wireless router?

  - by RLH

  Here is my setup. I have a Macbook, Thunderbolt display and an Ethernet connection plugged into the Thunderbolt display. During the day, most of my network use can (and should) operate across the ethernet associated with my display. However, I also need to be able to connect up to a wireless router. This hasn't been a problem on the Mac OS X side, but the program that I need to run on the router has to obtain an IP address from the wireless access point. Considering my current setup, how can I leave it so that I can access the internet in OS X, yet have my Window 7 instance running in Parallels, get it's assigned IP address from a wireless router that my Mac is also connected to? I've fiddled around with the Parallel's network settings for an hour, and I can't get Parallel's to see the router, even though my Mac is certainly connected to it.

  Read the article

• Personal DNS server and fallback outside home

  - by Jens

  I have my own DNS server at home to access local names, and that is working fine. Then I have my laptop, now obviously my laptop leaves the home now and then, therefore it accesses different nets outside my home, and my DNS server is not accessible there... So I figured that I would just add Google as secondary DNS... But actually, when I do that, then suddenly I can't access my local stuff (at home that is, obviously), like my laptop is getting a quicker response from Google's DNS or something, because it can't find anything on the addresses I use locally. If I then remove the secondary DNS, and keeps my own, then it works fine again... So do I somehow need to seperate what DNS's to use on what nets? I already use sepperate DNS settings when I connect using my 3G modem, but when I use hotspots it seems to use the same settings regardless (at least in the train), also can it differ wired connections?... Is there another solution?

  Read the article

• TFS2010 - Correctly setting up Host Header

  - by Keith Barrows

  We have a TFS2010 install on a Win2008R2 Server running IIS7. I've created a host header for TFS and want to use that instead of machineName:8080. I am getting weird behavior from it now. Every other time I log in I get: TF31002: Unable to connect to this Team Foundation Server: http://web2/tfs. Team Foundation Server Url: http://web2/tfs. Possible reasons for failure include: - The name, port number, or protocol for the Team Foundation Server is incorrect. - The Team Foundation Server is offline. - The password has expired or is incorrect. Technical information (for administrator): The request failed with HTTP status 404: Not Found. I force a reconnect and voila - there it is. Also, connecting to the web site rarely works but connecting via VS2010 works 50% of the time. What do i need to change to stabilize this?

  Read the article

• Wammu - USB Device Name?

  - by Paul

  I'm trying to get to my phone's filesystem through USB in Wammu, but I'm stuck in the configuration wizard when it asks for a USB device name. After about an hour of Internet searching, here are the failed solutions I've already tried, starting with the relevant information returned by lsusb in terminal. lsusb Bus 001 Device 003: ID 12d1:101e Huawei Technologies Co., Ltd. So I tried opening Wammu through sudo wammu in terminal and inputting "/dev/bus/usb/001/003" as the device name, which returns: Error opening device Device /dev/bus/usb/001/003 does not exist! and then "/dev/bus/usb/001/", which returns: Failed to connect to phone Description: Error opening device. Unknown, busy, or no permissions.<br> Function: Init<br> Error code: 2 Another proposed solution was to try "tail -f /var/log/messages" in terminal. But that only returned a "No such file or directory" message. Seemingly relevant dmesg info: [ 4739.716214] usb 1-1: new high-speed USB device number 8 using ehci_hcd [ 4739.854137] scsi9 : usb-storage 1-1:1.0 [ 4740.854416] scsi 9:0:0:0: CD-ROM HUAWEI T Mass Storage 2.31 PQ: 0 ANSI: 2 [ 4740.867051] sr0: scsi-1 drive [ 4740.867806] sr 9:0:0:0: Attached scsi CD-ROM sr0 [ 4740.870464] sr 9:0:0:0: Attached scsi generic sg1 type 5 I don't know why it is coming up as CD-ROM. But there it is. If you haven't noticed already, I'm an absolute beginner when it comes to Linux and terminal. So speaking to me like I'm a three year old is welcome if you can propose a solution. I'm running Ubuntu 12.04 LTS, and the phone is a Huawei U1250. My computer is an Acer Aspire One D250/KAV60. Any help is much appreciated.

  Read the article

• Can I port forward to an established reverse ssh tunnel

  - by Ben Holness

  I have three computers, A, B and C A has initiated a reverse ssh tunnel to B: ssh -nTNx -p 443 -R 22222:localhost:22 [user]@[server] If I log in to B, I can use 'ssh -p 22222 localhost' and I get a login prompt for A. If I try 'ssh -p 22222 [public IP of B]', it doesn't work What I would like to be able to do is have C connect to A without needing to login to B. So from C I could 'ssh -p 22222 [public IP of B]' and I would get the login prompt for A. I am using debian and shorewall and I have a basic understanding of how things work. I have tried various combinations of REDIRECT and DNAT rules, but haven't had any luck. I have tried using the same port (22222) and a different port (forwarding 22223 from C to 22222 on localhost). Any ideas? Cheers, Ben

  Read the article

• Ubuntu 12.04 MySQL 5.5 MyODBC 5.1 or 3.1 query hangs

  - by jorgearr

  I have been able to install Ubuntu 12.04 with LAMP MySQL version 5.5.x It works fine within linux, it allows me to connect from myodbc windows vista or windows 7 I have configured networking access and have been able to access from windows vista using putty and other tcp connections like mysql query browser. I have also configured or disabled ufw firewall and apparmor. The connection works fine until I query data from the tables. It lets me query small amounts of data like: SELECT name FROM users limit 20 but if I do a SELECT * FROM users, it goes on a never-ending loop. This happens even on tables with very few records like 5 or even less. The problems occur with windows because I tried ssh from linux mint and it worked fine. I need to be able to work using MyODBC either 3.51 or 5.1 since my client program is made in VB6 and connects to mysql server via tcp/ip. The server is an HP PROLIANT ML350G6 with Intel Xeon 64 bits. I tried several ubuntu server version (12.04 64bit, 10.10 64bit, 11.04 32bit) and none has worked I even tried CentOS 6.3 and the same. As a reference, it works fine with onother ubuntu server version 6.x on HP Proliant 150 and mysql 5.0.x that is like 7 years old and never updated. Help Please.

  Read the article

• How to automatically restart RRAS service after OpenVPN

  - by JT

  I have OpenVPN set up on a Windows Server 2003 box using a routed configuration. This allows users to connect and access the work LAN subnet. There are remote hosts/services however that are only accessible when used via the work network. To enable access these, I push routes out to the clients to make sure traffic to these destinations goes across the VPN, and NAT the traffic using RRAS. This all works, except: if I restart the OpenVPN service, network traffic stops working until I restart the RRAS service as well. Is there a good way for me to make the RRAS service start/restart after OpenVPN? Are service dependencies the way to go? Obviously I could write a batch file to do this, but I'd like to make the process as bullet-proof and obvious as I can so it doesn't cause problems for other admins.

  Read the article

• monitoring TCP/IP performance on Solaris

  - by Andy Faibishenko

  I am trying to tune a high message traffic system running on Solaris. The architecture is a large number (600) of clients which connect via TCP to a big Solaris server and then send/receive relatively small messages (.5 to 1K payload) at high rates. The goal is to minimize the latency of each message processed. I suspect that the TCP stack of the server is getting overwhelmed by all the traffic. What are some commands/metrics that I can use to confirm this, and in case this is true, what is the best way to alleviate this bottleneck?

  Read the article

• In IIS why do HTTP requests use the host header, and FTP requests do not

  - by Keeno

  So.... In IIS, if you use the in-build FTP you need to combine the FTP host header in the FTP username e.g. www.hello.com|domain/username So, the FTP program gets its "hook" from the username. However, you can connect to the FTP site using www.hello.com:21 over the FTP port. Why then, doesnt the FTP service work the same way as the HTTP service? IIS knows what site to serve back based on the host header after all.... Thanks!

  Read the article

• SSMS hanging without error when connecting to SQL

  - by Rob Farley

  Scary day for me last Thursday. I had gone up to Brisbane, and was due to speak at the Queensland SQL User Group on Thursday night. Unfortunately, disaster struck about an hour beforehand. Nothing to do with the recent floods (although we were meeting in a different location because of them). It was actually down to the fact that I’d been fiddling with my machine to get Virtual Server running on Windows 7, and SQL had finally picked up a setting from then. I could run Management Studio, but it couldn’t connect at all. No error, it just seemed to hang. One of the things you have to do to get Virtual Server installed is to tweak the Group Policy settings. I’d used gpupdate /force to get Windows to pick up the new setting, which allowed me to get Virtual Server running properly, but at the time, SQL was still using the previous settings. Finally when in Brisbane, my machine picked up the new settings, and caused me pain. Dan Benediktson describes the situation. If the SQL client picks up the wrong value out of the GetOverlappedResult API (which is required for various changes in Windows 7 behaviour), then Virtual Server can be installed, but SQL Server won’t allow connections. Yay. Luckily, it’s easy enough to change back using the Group Policy editor (gpedit.msc). Then restarting the machine (again!, as gpupdate /force didn’t cut it either, because SQL had already picked up the value), and finally I could reconnect. On Thursday I simply borrowed another machine for my talk. Today, one of my guys had seen and remembered Dan’s post. Thanks, both of you.

  Read the article

• Feeling Old? Before Middleware, Gamification, and MacBook Airs

  - by ultan o'broin

  Think we're done with green screens in the enterprise apps world? Fusion User Experience Advocate Debra Lilley (@debralilley) drew my attention to this super retro iPad terminal emulator app being used by a colleague to connect to JDE. Yes, before Middleware, this is how you did it. Surely the ultimate in hipster retro coexistence? Mind you, I've had to explain to lots of people I showed this to just what Telnet and IBM AS/400 are (or were). MochaSoft TN5250 Terminal Emulator iPad App This OG way of connecting to apps is a timely reminder not to forget all those legacy apps out there and the UX aspect to adoption and change. If a solution already works well and there's an emotional attachment to it, then the path to upgrade needs to be very clear and have valuable and demonstrable ROI for users and decision makers, a path that spans emotion and business benefits. On a pure usability front, that old school charm of the character-based green glow look 'n' feel could be easily done as a skin, personalizing an application for the user so that they feel comfortable with it. Fun too particularly in the mobile and BYOD space! In fact, there is a thriving retro apps market out there as illustrated by this spiffy lunar lander app (hat tip: John Cartan), part of a whole set of Atari's greatest hits available for iOS. Lunar Lander App And of course, there's the iOS version of Pong. Check out this retro Apple Mac SE/30 too. I actually remember using one of these. I have an Apple Mac Plus somewhere in my parents' house. I tried it out recently, and it actually booted, although all it was good for was playing the onboard games. Looking at all these olde worlde things makes me feel very old, but kinda warm inside too. The latter is a key part of today's applications user experience too.

  Read the article

• Site to site VPN using RRAS from an untrusted network?

  - by DrZaiusApeLord

  Our remote office will be moving to a new space where internet will be provided. They'll be behind a router doing NAT (I do not have admin rights to this router). They will be sharing a printer with the other people on the LAN, but will need VPN to our network for email and file shares. I was thinking of just having them run the windows VPN client and connecting via PPTP like they do when they are off-site, but I have read that multiple PPTP connections from the same NAT'd address to the same destination doesn't work well or at all. I am thinking some kind of site-to-site VPN is needed so there is just one tunnel. Can I just put in a VPN gateway, set it to connect to our RRAS/PPTP server, and have them use it as their default gateway? Perhaps even use the local default gateway for internet traffic. If so, what VPN gateway/device is recommended for this? Or other solutions? Thanks.

  Read the article

• Squid traffic tunneled through VPN

  - by NerdyNick

  So what I'm trying to do is have a Squid Proxy run on 1 machine along side a VPN connection. What I want to happen is all traffic running though the Squad Proxy would run though the VPN for its outbound. ie Desktop - (Squid Proxy - VPN) The goal is to allow my desktop selective tunneling through the VPN. So that Instant Messaging and the like that do not need to run through the VPN can go through my normal traffic. Typically I would go though a SSH Proxy but currently am forced to use VPN to gain entry into the office, and a Squid proxy seemed like it might work out the easiest for what I am needing. EDIT Realize I forgot to actually state what problem I'm running into. I have the Squid setup and verified it works, but once I connect to the VPN. All requests to Squid get accepted but Squid is unable to make the request over the VPN. So the client ends up just sitting there.

  Read the article

• How Can I Disable CRL Checks For A Windows 2008 App Using WinHTTP?

  - by Mike B

  I've got a Windows 2008 server with an app that uses WinHTTP for SSL sessions. The server is isolated from the internet but still tries to connect to CRL distribution points, which leads to some timeouts. Since the server has no access to the internet whatsoever, I'd like to disable CRL checks. I had a similar issue on a Windows 2003 server and resolved it by adjusting the following registry keys: HKEY_LOCAL_MACHINE/System/CurrentControlSet/Services/Http/Parameters/SslBindiongInfo/0.0.0.0:443/DefaultSslCertCheckMode (DWORD=1) HKEY_LOCAL_MACHINE/System/CurrentControlSet/Services/Rasman/PPP/EAP/13/NoRevocationCheck (DWORD = 1) HKEY_LOCAL_MACHINE/System/CurrentControlSet/Services/Rasman/PPP/EAP/13/NoRootRevocationCheck (DWORD = 1) That doesn't seem to be working in 2008. I've also tried disabling the CRL check from IE under Tools Internet Options Advanced. Is there anything else I can try here?

  Read the article

• Comparing the Performance of Visual Studio's Web Reference to a Custom Class

  As developers, we all make assumptions when programming. Perhaps the biggest assumption we make is that those libraries and tools that ship with the .NET Framework are the best way to accomplish a given task. For example, most developers assume that using ASP.NET's Membership system is the best way to manage user accounts in a website (rather than rolling your own user account store). Similarly, creating a Web Reference to communicate with a web service generates markup that auto-creates a proxy class, which handles the low-level details of invoking the web service, serializing parameters, and so on. Recently a client made us question one of our fundamental assumptions about the .NET Framework and Web Services by asking, "Why should we use proxy class created by Visual Studio to connect to a web service?" In this particular project we were calling a web service to retrieve data, which was then sorted, formatted slightly and displayed in a web page. The client hypothesized that it would be more efficient to invoke the web service directly via the HttpWebRequest class, retrieve the XML output, populate an XmlDocument object, then use XSLT to output the result to HTML. Surely that would be faster than using Visual Studio's auto-generated proxy class, right? Prior to this request, we had never considered rolling our own proxy class; we had always taken advantage of the proxy classes Visual Studio auto-generated for us. Could these auto-generated proxy classes be inefficient? Would retrieving and parsing the web service's XML directly be more efficient? The only way to know for sure was to test my client's hypothesis. Read More >

  Read the article

• ecryptfs - decrypt and mount at boot with USB key

  - by Josh McGee

  I have a system running Ubuntu Server as a testbed for some services that I want to get familiar with. I decided to let the installation procedure set up encryption. I knew all along that I would have to decrypt it with the passphrase in order to get the system booted, but I assumed it wouldn't matter since it will only boot once or twice a month. However, my brother has informed me that he is a victim of power outages at the residence where this server is located. This means we have to explain to his girlfriend how to turn on the computer, attach a keyboard, connect a monitor (she just can't understand that she can type to the computer without a display, so whatever) and input the passphrase for us, while we are at work. I have arrived at the conclusion that I should just put together a USB key that can be plugged in before powering on the computer, to avoid all the trouble. Is this possible with ecryptfs? Is there a tutorial or simple list of instructions available so that I can knock this out and focus back on the stuff I care about? EDIT: I am aware that this is possible with LUKS and dm-crypt, but unfortunately the magical encryption that Ubuntu hands you during the installation is only ecryptfs so my question is specific to that.

  Read the article

• ArchBeat Link-o-Rama for 2012-06-01

  - by Bob Rhubart

  Complexity of Social Computing - Is it a Consideration for EA's? | Pat Shepherd blogs.oracle.com Pat Shepherd asks, "Does Enterprise Architecture need to consider Social Computing in its scope?" Who should own the Enterprise Architecture? | Michael Glas blogs.oracle.com "Instead of looking at just who owns the architecture," suggests Michael Glas, "think about what the person/role/organization should do." The Application Architecture Domain | Michael Glas blogs.oracle.com Michael Glas asks—and answers: "As an Enterprise Architect, what do I need to consider when looking at/defining/designing the Application Architecture Domain?" CAP Twelve Years Later: How the "Rules" Have Changed | Eric Brewer www.infoq.com The CAP theorem asserts that any net­worked shared-data system can have only two of three desirable properties. How­ever, by explicitly handling partitions, designers can optimize consistency and availability, thereby achieving some trade-off of all three. Oracle DB with OEM in Amazon Cloud | Dr. Frank Munz www.munzandmore.com Dr. Frank Munz shares a video that screencast that explains "how to create an Oracle DB instance in AWS, how to enable OEM...and how to connect to your cloud instance with a local installation of NetBeans." Sample External Login.jsp page for Oracle Access Manager 11g | Brian Eidelman fusionsecurity.blogspot.com A-Team blogger Brian Eidelman expands on a previous post dealing with configuring OAM 11g to use an externally hosted custom login page. Bay Area Coherence Special Interest Group (BACSIG) Meeting June 7 coherence.oracle.com Date: Thursday, June 7, 2012 Time: 5:30pm – 9:00pm PT Where: Oracle Conference Center Room # 103 350 Oracle Parkway Redwood, Shores, CA Presentations: 6:00 p.m. - Coherence 101, The Evolution of Distributed Caching - Noah Arliss (Oracle) 7:00 p.m. - Optimizing Performance for Oracle Coherence and TopLink Grid at OOCL - Matt Rosen, Leo Limqueco (OOCL) 8:00 p.m. - Oracle Coherence Message Bus - Extreme Performance on Oracle Exalogic - Ballav Bihani (Oracle) Thought for the Day "I can't be left unsupervised." — Ron Wood (Born 06/01/1947 Source: Brainy Quote

  Read the article

• About can't export the performance problems.

  - by kyrathy

  At first,let me describe my environment My VirtualCenter is install on window 2008 ,and theDataBase that VC used is SQL 2008 I really want to ask is ..... When I use vsphere clinet to connect VC.....I got a problem. the performance chart only can show "realtime "...... whatever I only want to view the chart , or I want to export the performance log . when i manually want to export performance, and I select the time to 1 hour ,1 day ,1 month, or from a to b. it showed "No performance data to report for selected objects" only select realtime can export data normally. Before I install Vsphere 4 , I install the SQL 2008 , used the schema in the install CD(I follow the step to create SQL DB for vSphere) Could anybody help me how to solve this problem? And if need any information ,just tell me to provide. Thanks a lot.

  Read the article

• Ubuntu 12.04 problem with E160 huawei - can't detect the device nad freezing system

  - by Matt

  i just installed 12.04, plug in E160 and nothing happend - modem doesnt mount. i found this solution : Ubuntu does not mount some Huawei devices due to bugs, problems etc. See if these work: 1st option: Connect the USB modem. After 10 seconds, type this in a terminal window: lsusb The output will be like this: Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 004 Device 004: ID 12d1:140b Huawei Technologies Co., Ltd. Bus 004 Device 002: ID 413c:3016 Dell Computer Corp. Optical 5-Button Wheel Mouse Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 002 Device 005: ID 0b97:7762 O2 Micro, Inc. Oz776 SmartCard Reader Bus 002 Device 004: ID 413c:8103 Dell Computer Corp. Wireless 350 Bluetooth Bus 002 Device 003: ID 0b97:7761 O2 Micro, Inc. Oz776 1.1 Hub Bus 002 Device 002: ID 413c:a005 Dell Computer Corp. Internal 2.0 Hub Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub The device is a Huawei modem, so let's look at the output. The relevant entry is: Bus 004 Device 004: ID 12d1:140b Huawei Technologies Co., Ltd Hence, you must type: sudo modprobe usbserial vendor=0x12d1 product=0x140b 2nd option Download usb-modeswitch and usb-modeswitch-data packages from packages.ubuntu.com. Install them through the command: sudo dpkg -i usb-modeswitch*.deb 3rd option Try a combination of both. but with no result. Modem still is not detected. I've tried to add new conection but system can't see my device in setup dialogue. Also i noticed that when i open eg. terminal and try to type sth, system freeze for a while.. Thx for help!

  Read the article

• Cross Apply Ambiguity

  - by Dave Ballantyne

  Cross apply (and outer apply)  are a very welcome addition to the TSQL language.  However, today after a few hours of head scratching, I have found an simple issue which could cause big big problems. What would you expect from this statement ? select * from sys.objects b join sys.objects a on a.object_id = object_id No prizes for guessing SQL server errors with “Ambiguous column name 'object_id'”. What would you expect from this statement ? Select * from sys.objects a cross apply( Select * from sys.objects b where b.object_id = object_id) as c Surprisingly, perhaps, the result is a cross join of sys.objects.  Well, what happened there ? If you look at the apply statement, within the where clause, only one of the conditions is qualified with a table name.  This meant that is has be interpreted as “b.object_id = b.object_id” causing the cross apply to have no join the the parent sys.objects table and causing the cross join. The fix is , obviously, simple Select * from sys.objects a cross apply( Select * from sys.objects b where b.object_id = a.object_id) as c So why no “Ambiguous column name ” error ?  I’ve raised a connect item on this issue here.

  Read the article

< Previous Page | 358 359 360 361 362 363 364 365 366 367 368 369  | Next Page >