Search Results

Search found 12964 results on 519 pages for 'pass summit 2013'.

Page 283/519 | < Previous Page | 279 280 281 282 283 284 285 286 287 288 289 290  | Next Page >

  • SQL Server Optimizer Malfunction?

    - by Tony Davis
    There was a sharp intake of breath from the audience when Adam Machanic declared the SQL Server optimizer to be essentially "stuck in 1997". It was during his fascinating "Query Tuning Mastery: Manhandling Parallelism" session at the recent PASS SQL Summit. Paraphrasing somewhat, Adam (blog | @AdamMachanic) offered a convincing argument that the optimizer often delivers flawed plans based on assumptions that are no longer valid with today’s hardware. In 1997, when Microsoft engineers re-designed the database engine for SQL Server 7.0, SQL Server got its initial implementation of a cost-based optimizer. Up to SQL Server 2000, the developer often had to deploy a steady stream of hints in SQL statements to combat the occasionally wilful plan choices made by the optimizer. However, with each successive release, the optimizer has evolved and improved in its decision-making. It is still prone to the occasional stumble when we tackle difficult problems, join large numbers of tables, perform complex aggregations, and so on, but for most of us, most of the time, the optimizer purrs along efficiently in the background. Adam, however, challenged further any assumption that the current optimizer is competent at providing the most efficient plans for our more complex analytical queries, and in particular of offering up correctly parallelized plans. He painted a picture of a present where complex analytical queries have become ever more prevalent; where disk IO is ever faster so that reads from disk come into buffer cache faster than ever; where the improving RAM-to-data ratio means that we have a better chance of finding our data in cache. Most importantly, we have more CPUs at our disposal than ever before. To get these queries to perform, we not only need to have the right indexes, but also to be able to split the data up into subsets and spread its processing evenly across all these available CPUs. Improvements such as support for ColumnStore indexes are taking things in the right direction, but, unfortunately, deficiencies in the current Optimizer mean that SQL Server is yet to be able to exploit properly all those extra CPUs. Adam’s contention was that the current optimizer uses essentially the same costing model for many of its core operations as it did back in the days of SQL Server 7, based on assumptions that are no longer valid. One example he gave was a "slow disk" bias that may have been valid back in 1997 but certainly is not on modern disk systems. Essentially, the optimizer assesses the relative cost of serial versus parallel plans based on the assumption that there is no IO cost benefit from parallelization, only CPU. It assumes that a single request will saturate the IO channel, and so a query would not run any faster if we parallelized IO because the disk system simply wouldn’t be able to handle the extra pressure. As such, the optimizer often decides that a serial plan is lower cost, often in cases where a parallel plan would improve performance dramatically. It was challenging and thought provoking stuff, as were his techniques for driving parallelism through query logic based on subsets of rows that define the "grain" of the query. I highly recommend you catch the session if you missed it. I’m interested to hear though, when and how often people feel the force of the optimizer’s shortcomings. Barring mistakes, such as stale statistics, how often do you feel the Optimizer fails to find the plan you think it should, and what are the most common causes? Is it fighting to induce it toward parallelism? Combating unexpected plans, arising from table partitioning? Something altogether more prosaic? Cheers, Tony.

    Read the article

  • Call For Papers Tips and Tricks

    - by speakjava
    This year's JavaOne session review has just been completed and by now everyone who submitted papers should know whether they were successful or not.  I had the pleasure again this year of leading the review of the 'JavaFX and Rich User Experiences' track.  I thought it would be useful to write up a few comments to help people in future when submitting session proposals, not just for JavaOne, but for any of the many developer conferences that run around the world throughout the year.  This also draws on conversations I recently had with various Java User Group leaders at the Oracle User Group summit in Riga.  Many of these leaders run some of the biggest and most successful Java conferences in Europe. Try to think of a title which will sound interesting.  For example, "Experiences of performance tuning embedded Java for an ARM architecture based single board computer" probably isn't going to get as much attention as "Do you like coffee with your dessert? Java on the Raspberry Pi".  When thinking of the subject and title for your talk try to steer clear of sessions that might be too generic (and so get lost in a group of similar sessions).  Introductory talks are great when the audience is new to a subject, but beware of providing sessions that are too basic when the technology has been around for a while and there are lots of tutorials already available on the web. JavaOne, like many other conferences has a number of fields that need to be filled in when submitting a paper.  Many of these are selected from pull-down lists (like which track the session is applicable to).  Check these lists carefully.  A number of sessions we had needed to be shuffled between tracks when it was thought that the one selected was not appropriate.  We didn't count this against any sessions, but it's always a good idea to try and get the right one from the start, just in case. JavaOne, again like many other conferences, has two fields that describe the session being submitted: abstract and summary.  These are the most critical to a successful submission.  The two fields have different names and that is significant; a frequent mistake people make is to write an abstract for a session and then duplicate it for the summary.  The abstract (at least in the case of JavaOne) is what gets printed in the show guide and is typically what will be used by attendees when deciding what sessions to attend.  This is where you need to sell your session, not just to the reviewers, but also the people who you want in your audience.  Submitting a one line abstract (unless it's a really good one line) is not usually enough to decide whether this is worth investing an hour of conference time.  The abstract typically has a limit of a few hundred characters.  Try to use as many of them as possible to get as much information about your session across.  The summary should be different from the abstract (and don't leave it blank as some people do).  This field is where you can give the reviewers more detail about things like the structure of the talk, possible demonstrations and so on.  As a reviewer I look to this section to help me decide whether the hard-sell of the title and abstract will actually be reflected in the final content.  Try to make this comprehensive, but don't make it excessively long.  When you have to review possibly hundreds of sessions a certain level of conciseness can make life easier for reviewers and help the cause of your session. If you've not made many submissions for talks in the past, or if this is your first, try to give reviewers places to find background on you as a presenter.  Having an active blog and Twitter handle can also help reviewers if they're not sure what your level of expertise is.  Many call-for-papers have places for you to include this type of information.  It's always good to have new and original presenters and presentations for conferences.  Hopefully these tips will help you be successful when you answer the next call-for-papers.

    Read the article

  • Brazil is Hot for Social Media

    - by Mike Stiles
    Today’s guest blog is from Oracle SVP Product Development Reggie Bradford, fresh off a visit to Sao Paulo, Brazil where he spoke at the Dachis Social Business Summit and spent some time getting a personal taste for the astonishing growth of social in Brazil, both in terms of usage and engagement. I knew it was big, but I now have an all-new appreciation for why the Wall Street Journal branded Brazil the “social media capital of the universe.” Brazil has the world’s 5th largest economy, an expanding middle class, an active younger demo market, a connected & outgoing culture, and an ongoing embrace of the social media platforms. According to comScore's 2012 Brazil Digital Future in Focus report, 97% are using social media, and that’s not even taking mobile-only users into account. There were 65 million Facebook users in 2012, spending an average 535 minutes there, up 208%. It’s one of Twitter’s fastest growing markets and the 2nd biggest market for YouTube. Instagram usage has grown over 300% since last year. That by itself is exciting, but look at the opportunity for social marketing brands. 74% of Brazilian social users follow brands on Facebook, and 59% have praised a company on either Twitter or Facebook. A 2011 Oh! Panel study found 81% of social networkers there used social to research new products and 75% went there looking for discounts. B2C eCommerce sales in Brazil is projected to hit $26.9 billion by 2015. I bet I’m not the only one who sees great things ahead, and I was fortunate enough give a keynote ABRADI, an association of leading digital agencies in Brazil with 53 execs from 35 agencies attending. I was also afforded the opportunity to give my impressions of what’s going on in Brazil to Jornal Propoganda & Marketing, one of the most popular publications in Latin America for marketers. I conveyed that especially in an environment like Brazil, where social users are so willing to connect and engage brands, marketers need to back away from the heavy-handed, one-way messaging of old school advertising and move toward genuine relationships and trust-building. To aide in this, organizational and operation changes must be embraced inside the enterprise. We've talked often about the new, tighter partnership forming between the CIO and CMO. If this partnership is not encouraged, fostered and resourced, the increasing amount of time consumers spend on mobile and digital, and the efficiencies and integrations offered by cloud-based software cannot be exploited. These are the kinds of changes that can yield social data that, when combined with enterprise data, helps you come to know your social audiences intimately and predict their needs. Consumers are always connected and need your brand to be accessible at any time, be it for information or customer service. And, of course, all of this is happening quite publicly. The holistic, socially-enable enterprise connects social to customer service systems and all other customer touch points, facilitating the kind of immediate, real-time, gratifying response customers are coming to expect. Social users in Brazil are highly active and clearly willing to meet us as brands more than halfway. Empowering yourself with a social management technology platform will have you set up to maximize this booming social market…from listening & monitoring to engagement to analytics to workflow & automation to globalization & language support. Brands, it’s time to be as social as the great people of Brazil are. Obrigado! @reggiebradfordPhoto: Gualberto107, freedigitalphotos.net

    Read the article

  • Finalists for Community Manager of the Year Announced

    - by Mike Stiles
    For as long as brand social has been around, there’s still an amazing disparity from company to company on the role of Community Manager. At some brands, they are the lead social innovators. At others, the task has been relegated to interns who are at the company temporarily. Some have total autonomy and trust. Others must get chain-of-command permission each time they engage. So what does a premiere “worth their weight in gold” Community Manager look like? More than anyone else in the building, they have the most intimate knowledge of who the customer is. They live on the front lines and are the first to detect problems and opportunities. They are sincere, raving fans of the brand themselves and are trusted advocates for the others. They’re fun to be around. They aren’t salespeople. Give me one Community Manager who’s been at the job 6 months over 5 focus groups any day. Because not unlike in speed dating, they must immediately learn how to make a positive, lasting impression on fans so they’ll want to return and keep the relationship going. They’re informers and entertainers, with a true belief in the value of the brand’s proposition. Internally, they live at the mercy of the resources allocated toward social. Many, whose managers don’t understand the time involved in properly curating a community, are tasked with 2 or 3 too many of them. 63% of CM’s will spend over 30 hours a week on one community. They come to intuitively know the value of the relationships they’re building, even if they can’t always be shown in a bar graph to the C-suite. Many must communicate how the customer feels to executives that simply don’t seem to want to hear it. Some can get the answers fans want quickly, others are frustrated in their ability to respond within an impressive timeframe. In short, in a corporate world coping with sweeping technological changes, amidst business school doublespeak, pie charts, decks, strat sessions and data points, the role of the Community Manager is the most…human. They are the true emotional connection to the real life customer. Which is why we sought to find a way to recognize and honor who they are, what they do, and how well they have defined the position as social grows and integrates into the larger organization. Meet our 3 finalists for Community Manager of the Year. Jeff Esposito with VistaprintJeff manages and heads up content strategy for all social networks and blogs. He also crafts company-wide policies surrounding the social space. Vistaprint won the NEDMA Gold Award for Twitter Strategy in 2010 and 2011, and a Bronze in 2011 for Social Media Strategy. Prior to Vistaprint, Jeff was Media Relations Manager with the Long Island Ducks. He graduated from Seton Hall University with a BA in English and a minor in Classical Studies. Stacey Acevero with Vocus In addition to social management, Stacey blogs at Vocus on influential marketing and social media, and blogs at PRWeb on public relations and SEO. She’s been named one of the #Nifty50 Women in Tech on Twitter 2 years in a row, as well as included in the 15 up-and-coming PR pros to watch in 2012. Carly Severn with the San Francisco BalletCarly drives engagement, widens the fanbase and generates digital content for America’s oldest professional ballet company. Managed properties include Facebook, Twitter, Tumblr, Pinterest, Instagram, YouTube and G+. Prior to joining the SF Ballet, Carly was Marketing & Press Coordinator at The Fitzwilliam Museum at Cambridge, where she graduated with a degree in English. We invite you to join us at the first annual Oracle Social Media Summit November 14 and 15 at the Wynn in Las Vegas where our finalists will be featured. Over 300 top brand marketers, agency executives, and social leaders & innovators will be exploring how social is transforming business. Space is limited and the information valuable, so get more info and get registered as soon as possible at the event site.

    Read the article

  • Zero to Cloud : One stop shop for resources to accelerate your transformation to enterprise private cloud

    - by Anand Akela
    Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} During the Oracle Open World 2012 last week, Oracle introduced "Zero to Cloud" resource center to help you accelerate your transformation journey to enterprise private cloud. To help organizations deploy fully operational, enterprise-grade private cloud environment in as little as half a day, Oracle has brought key content together into this single, user-friendly resource center. Normal 0 false false false EN-US X-NONE X-NONE MicrosoftInternetExplorer4 /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-qformat:yes; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:11.0pt; font-family:"Calibri","sans-serif"; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:"Times New Roman"; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi;} The resource center is launched just as the Oracle Cloud Builder Summit series moves into full swing. Designed for executives, cloud architects, and IT operations professionals, the day-long event series will eventually reach nearly 100 cities around the globe. During this event, an interactive "Zero to Cloud" session will showcase the transformational journey of a fictitious enterprise to the private cloud using the latest solutions from Oracle—including, Oracle Database, Oracle Fusion Middleware, Oracle VM and Oracle Enterprise Manager, as well as Oracle’s full range of engineered systems. The online "Zero to Cloud" resource center includes best practices from Oracle experts and early adopter customers as well as interviews with Oracle development executives responsibly for Oracle’s private cloud solutions and roadmap. It also includes a new self-assessment quiz that can help determine readiness for a successful private cloud deployment. Once you've determined organizational readiness, explore early adopter tips, demos, guides, exclusive white papers and more at the "Zero to Cloud" resource center.

    Read the article

  • ?????????????Fusion Middleware??????????? ?4?

    - by rika.tokumichi
    ??????????OTN????????? ??OTN???????Fusion Middleware???????????????????????????????????????????????? ?OTN????????????????????????????? 2010?3??????????????????? ???????????? Oracle Direct Seminar?????????? ???! ????? ?????????????? OTN???????????????????! ???Fusion Middleware????????????DB?????????????????????????????? ???????????????????? ···?????????????????5???? ????????????????????????????????????? ????????Oracle WebLogic Server? ???????????????! ???? ?Oracle Direct Seminar??????????? ????????? ?????????????????????????????? Oracle Direct Seminar??????????????????? ?????????????????????????????????????? -------------------------------------------- ????????????????????????????Oracle Direct Seminar(????)? ???????????????????????????????????????????????????????? ???????????????????????????????????? ??????????????????????????????????????????? ????????????????/?????????????????????????????????????????????????????????? ????????????????????????????????????????????????????? ?????????????(????)? ?????????????????????????????? ??????????????????????????????????? ?????????????????????????????????? ???????????????????!?DB?Upgrade???????? ?????????????????????????? ???????????????????????????? ??????????????????????????????????????????????????? Oralce Database????????????????????????????????????????????? ???????????????????????????????????????????????????????????????????????????? ?????????????(????)? ????????????????????? ?????????????????????????????????????????????? ?????????(??????????)? ??????????????????????(??????????)? ????????????????????????????????????????????????????? ??????????????????????·??????????????????? ??????????????????????????????? ????????????????????????????????????????????????????????????????????????????????????? ??????????????????????????????????????????OTN???????????????????????????????? ???????????????????????????????????????????????? -------------------------------------------- ????????????????????????????????????????! ???????????????! ?????????????????????????????! >Oracle Direct Seminar?????????? >?????????!????????????OTN???? ????????! ?????????????????????????????????? ????????????????????????????????????????? ???????????????OTN??????????????????????????????????????????????????????????????????????????????????????????? ?????????????????????????? ?Developers Summit????????????????????2?????????·????????? ???? ?????????????????? ???? ??(?????????? ????????????????????)?? ?????? ??(Fusion Middleware?????? Fusion Middleware???????? ??????????????) ??? ????????????????????????????^^ >???????????????????1(?????? ???) >???????????????????2(???????) ???4???????????????????????????????????????????????? ?Oracle VM?Oracle WebLogic Server????????? ??????????????????????????????????!! ???????????? ????????????2????????????????????? ?????????????????? ???OTN?????????????????????????????? ????????????????!! ???????????????????????! >????????Oracle Customer Successes? ?????4??????????50??(???)?????????????????? ?????????????????(by Applications)?????????(by Technology)???(by Industry)???????(by Services)?????????? ??????????????????????????????? ????????!????????????????????! ?????????????????????Fusion Middleware??????????? ?4??????????? ??????????????????????! Fusion Middleware???????????! ?INFORMATION INDEPTH NEWSLETTERS Fusion Middleware Edition???????! ??1?OTN???????Fusion Middleware???????????????? ???????????????????????????????????????????????? ???????????????????????!! ?????????? Oracle/OTN????????????????????????????????????????????????? ????????????????????????????????????? INFORMATION INDEPTH NEWSLETTERS Fusion Middleware Edition??????????????????????? >???????????????????????? ??????????????????????Oracle's Dev2DBA Newsletter?????????????^^ >?Oracle's Dev2DBA Newsletter?????????????????????????????????

    Read the article

  • Is there really a need for encryption to have true wireless security? [closed]

    - by Cawas
    I welcome better key-wording here, both on tags and title. I'm trying to conceive a free, open and secure network environment that would work anywhere, from big enterprises to small home networks of just 1 machine. I think since wireless Access Points are the most, if not only, true weak point of a Local Area Network (let's not consider every other security aspect of having internet) there would be basically two points to consider here: Having an open AP for anyone to use the internet through Leaving the whole LAN also open for guests to be able to easily read (only) files on it, and even a place to drop files on Considering these two aspects, once everything is done properly... What's the most secure option between having that, or having just an encrypted password-protected wifi? Of course "both" would seem "more secure". But it shouldn't actually be anything substantial. That's the question, but I think it may need more elaborating on. If you don't think so, please feel free to skip the next (long) part. Elaborating more on the two aspects ... I've always had the feeling using any kind of the so called "wireless security" methods is actually a bad design. I'm talking mostly about encrypting and pass-phrasing (which are actually two different concepts), since I won't even consider hiding SSID and mac filtering. I understand it's a natural way of thinking. With cable networking nobody can access the network unless they have access to the physical cable, so you're "secure" in the physical way. In a way, encrypting is for wireless what building walls is for the cables. And giving pass-phrases would be adding a door with a key. But the cabling without encryption is also insecure. If someone plugin all the data is right there. So, while I can see the use for encrypting data, I don't think it's a security measure in wireless networks. It's wasting resources for too little gain. I believe we should encrypt only sensitive data regardless of wires. That's already done with HTTPS, so I don't really need to encrypt my torrents, for instance. They're torrents, they are meant to be freely shared! As for using passwords, they should be added to the users, always. Not to wifi. For securing files, truly, best solution is backup. Sure all that doesn't happen that often, but I won't consider the most situations where people just don't care. I think there are enough situations where we actually use passwords on our OS users, so let's go with that in mind. I keep promoting the Fonera concept as an instance. It opens up a free wifi port, if you choose so, and anyone can connect to the internet through that, without having any access to your LAN. It also uses a QoS which will never let your bandwidth drop from that public usage. That's security, and it's open. But it's lacking the second aspect. I'll probably be bashed for promoting the non-usage of WPA 2 with AES or whatever, but I wanted to know from more experienced (super) users out there: what do you think?

    Read the article

  • Gmail and Live are making all messages from my server as spam.

    - by Ryan Kearney
    I'm getting very weird results here. When my server sends an email to my @hotmail or @gmail account, it's marked as spam. When I send email through my server from Outlook to @hotmail, it doesn't get marked as spam, but it still gets marked as spam in gmail. They seem to get through fine on Yahoo though. My servers hostname A record points to an IP address whose PTR record points back to the same domain name. The TXT record has a SPF record in it to allow email to be sent from that servers IP. I moved from a VPS to a Dedicated server when this started to happen. From what I can see, the email headers are identical. Here's one of my email headers that gmail marks as spam. Some fields were repalced. MYGMAILACCOUNT is the email address of the account the email was addressed to. USER is the name of the account on the system it was sent from HOSTNAME is the servers FQDN IPADDR is the IP Address of the Hostname MYDOMAIN is my domain name Delivered-To: MYGMAILACCOUNT Received: by 10.220.77.82 with SMTP id f18cs263483vck; Sat, 27 Feb 2010 23:58:02 -0800 (PST) Received: by 10.150.16.4 with SMTP id 4mr3886702ybp.110.1267343881628; Sat, 27 Feb 2010 23:58:01 -0800 (PST) Return-Path: <USER@HOSTNAME> Received: from HOSTNAME (HOSTNAME [IPADDR]) by mx.google.com with ESMTP id 17si4604419yxe.134.2010.02.27.23.58.01; Sat, 27 Feb 2010 23:58:01 -0800 (PST) Received-SPF: pass (google.com: best guess record for domain of USER@HOSTNAME designates IPADDR as permitted sender) client-ip=IPADDR; Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of USER@HOSTNAME designates IPADDR as permitted sender) smtp.mail=USER@HOSTNAME Received: from USER by HOSTNAME with local (Exim 4.69) (envelope-from <USER@HOSTNAME>) id 1Nle2K-0000t8-Bd for MYGMAILACCOUNT; Sun, 28 Feb 2010 02:57:36 -0500 To: Ryan Kearney <MYGMAILACCOUNT> Subject: [Email Subject] MIME-Version: 1.0 Content-type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit From: webmaster@MYDOMAIN Message-Id: <E1Nle2K-0000t8-Bd@HOSTNAME> Sender: <USER@HOSTNAME> Date: Sun, 28 Feb 2010 02:57:36 -0500 X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - HOSTNAME X-AntiAbuse: Original Domain - gmail.com X-AntiAbuse: Originator/Caller UID/GID - [503 500] / [47 12] X-AntiAbuse: Sender Address Domain - HOSTNAME Anyone have any ideas as to why all mail leaving my server gets marked as spam? EDIT: I already used http://www.mxtoolbox.com/SuperTool.aspx to check if my servers IP's are blacklisted and they are in fact not. That's what I thought at first, but it isn't the case. Update Mar 1, 2010 I received the following email from Microsoft Thank you for writing to Windows Live Hotmail Domain Support. My name is * and I will be assisting you today. We have identified that messages from your IP are being filtered based on the recommendations of the SmartScreen filter. This is the spam filtering technology developed and operated by Microsoft and is built around the technology of machine learning. It learns to recognize what is and isn't spam. In short, we filter incoming emails that look like spam. I am not able to go into any specific details about what these filters specifically entail, as this would render them useless. E-mails from IPs are filtered based upon a combination of IP reputation and the content of individual emails. The reputation of an IP is influenced by a number of factors. Among these factors, which you as a sender can control, are: The IP's Junk Mail Reporting complaint rate The frequency and volume in which email is sent The number of spam trap account hits The RCPT success rate So I'm guessing it has to do with the fact that I got an IP address with little or no history in sending email. I've confirmed that I'm not on any blacklists. I'm guessing it's one of those things that will work itself out in a month or so. I'll post when I hear more.

    Read the article

  • Varnish POST problem "9 FetchError c backend write error: 11" for application/x-www-form-urlencoded content

    - by ompap
    Cutting a longish story short, we have managed to get a more precise error out of Varnishlog. Varnishlog tells us that we are sending a 31 TxRequest - POST 31 TxHeader - Content-Type: application/x-www-form-urlencoded but we are getting 9 FetchError c backend write error: 11 31 BackendClose - [backend name] 9 VCL_call c error 9 VCL_return c deliver 9 Length c 488 9 VCL_call c deliver 9 VCL_return c deliver 9 TxProtocol c HTTP/1.1 9 TxStatus c 503 We still do not know what this is exactly, but apparently Content-Type: application/x-www-form-urlencoded is not getting through as it should. Help still needed, please! Original message below. The title was "Varnish not letting Joomla users to log in - 503 guru meditation error", but I changed it to get more attention to the problem and not to the symptoms. Hello, We have a production site for a local newspaper which is currently behind an Apache reverse proxy, basicly the site on one server and the other being reserved as a reverse proxy only (well, there is more but that has no relevance here). Apache as a reverse proxy works, but could be faster. We want to change the reverse proxy to use Varnish instead of Apache on an Ubuntu 10.4 Server. The Varnish is version 2.10 installed directly from Ubuntu repos. Ubuntu 10.4 uses PHP 5.3.2. For anonymous surfers the site works wonderfully with Varnish. So far we can get very good speed out of Varnish, we just have a few problems with logging in or out. The big one is, that the users cannot log in: they get a Varnish 503 error page every time. The logs do not reveal the cause. It feels as if the request would never leave Varnish. So we are merely guessing - not a strong starting point. We have gone through what has been suggested on various plces on the web. We have increased the timeouts to backend xxx { .host = "xxx.xx"; .port = "http"; .connect_timeout = 60s; .first_byte_timeout = 60s; .between_bytes_timeout = 60s; } but we seem to get the 503 guru error page much faster than that, as in approx. 5 seconds. We have increased the Varnish headers size to 128 in daemon. In vcl_recv we have if (req.http.Authenticate || req.http.Authorization) { return(pass); } and in vcl_fetch ## auhtentication handling if (req.http.Authenticate || req.http.Authorization) { return(pass); } We do not strip cookies. We have tried to make sure that error pages are not cached. As said above, we cannot see anything in the backend Apache logs, apparently it never gets asked for Joomla user authentication. Varnish does not seem to get much mentioning with connection to Joomla. (We cannot dump Joomla, that selection has been done and we just have to live with what we have been given) Has anyone a working Varnish - Joomla combination? Thanks for reading. Please help. We need some hints - desperately. Any suggestions? ompap

    Read the article

  • IIS 7.5 on Windows Server 2008 R2 refusing to create PASSIVE MODE FTP connections

    - by Campbell
    I'm attempting to get an FTP client written in perl to transfer files from an IIS 7.5 FTP server using passive mode. I've configured the FTP server as per instructions and have also configured Windows Firewall to allow this type of traffic. I have validated that the firewall is behaviong correctly by checking to ensure there are no blocked packets in the logs. I have verified the that FTP control channel is being opened on Port 21. I believe the client is being told by IIS which port to connect on for passive mode and IIS is refusing to allow this connection. The perl log looks like: C:\cygwin\Perl\lib\FMT>perl FTPTest.pl Net::FTP>>> Net::FTP(2.77) Net::FTP>>> Exporter(5.64_01) Net::FTP>>> Net::Cmd(2.29) Net::FTP>>> IO::Socket::INET(1.31) Net::FTP>>> IO::Socket(1.31) Net::FTP>>> IO::Handle(1.28) Net::FTP=GLOB(0x20abac0)<<< 220 Microsoft FTP Service Net::FTP=GLOB(0x20abac0)>>> USER ftpuser Net::FTP=GLOB(0x20abac0)<<< 331 Password required for ftpuser. Net::FTP=GLOB(0x20abac0)>>> PASS .... Net::FTP=GLOB(0x20abac0)<<< 230 User logged in. Net::FTP=GLOB(0x20abac0)>>> CWD /Logs Net::FTP=GLOB(0x20abac0)<<< 250 CWD command successful. Net::FTP=GLOB(0x20abac0)>>> PASV Net::FTP=GLOB(0x20abac0)<<< 227 Entering Passive Mode (xx,xxx,xxx,xxx,160,41). Net::FTP=GLOB(0x20abac0)>>> RETR filename.txt Can't use an undefined value as a symbol reference at C:/Utilities/strawberryper l/perl/lib/Net/FTP/dataconn.pm line 54. IIS logs look as follows: 2010-10-02 17:40:06 xx.xxx.xx.xx - yy.y.yy.yy ControlChannelOpened - - 0 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a - - 2010-10-02 17:40:06 xx.xxx.xx.xx - yy.y.yy.yy USER ftpuser 331 0 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a - - 2010-10-02 17:40:06 xx.xxx.xx.xx MACHINENAME\ftpuser yy.y.yy.yy PASS *** 230 0 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a / - 2010-10-02 17:40:06 xx.xxx.xx.xx MACHINENAME\ftpuser yy.y.yy.yy CWD /Logs 250 0 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a /Logs - 2010-10-02 17:40:06 xx.xxx.xx.xx MACHINENAME\ftpuser yy.y.yy.yy PASV - 227 0 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a - - 2010-10-02 17:40:27 - MACHINENAME\ftpuser zz.z.zz.zzz 41001 DataChannelClosed - - 64 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a - - 2010-10-02 17:40:27 xx.xxx.xx.xx MACHINENAME\ftpuser yy.y.yy.yy ControlChannelClosed - - 64 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a - - 2010-10-02 17:40:27 xx.xxx.xx.xx MACHINENAME\ftpuser yy.y.yy.yy RETR filename.txt 550 1236 0 27a48c9b-9dce-4770-8bcf-fc89f2569b1a filename.txt - We've managed to see this issue with other FTP clients also, I don't think its something funny in Perl. I've been informed that this works fine in the IIS 6 FTP server. I'm wondering if there is something we're missing here.

    Read the article

  • Site-to-site VPN using MD5 instead of SHA and getting regular disconnection

    - by Steven
    We are experiencing some strange behavior with a site-to-site IPsec VPN that goes down about every week for 30 minutes (Iam told 30 minutes exactly). I don't have access to the logs, so it's difficult to troubleshoot. What is also strange is that the two VPN devices are set to use SHA hash algorithm but apparently end up agreeing to use MD5. Does anybody have a clue? or is this just insufficient information? Edit: Here is an extract of the log of one of the two VPN devices, which is a Cisco 3000 series VPN concentrator. 27981 03/08/2010 10:02:16.290 SEV=4 IKE/41 RPT=16120 xxxxxxxx IKE Initiator: New Phase 1, Intf 2, IKE Peer xxxxxxxx local Proxy Address xxxxxxxx, remote Proxy Address xxxxxxxx, SA (L2L: 1A) 27983 03/08/2010 10:02:56.930 SEV=4 IKE/41 RPT=16121 xxxxxxxx IKE Initiator: New Phase 1, Intf 2, IKE Peer xxxxxxxx local Proxy Address xxxxxxxx, remote Proxy Address xxxxxxxx, SA (L2L: 1A) 27986 03/08/2010 10:03:35.370 SEV=4 IKE/41 RPT=16122 xxxxxxxx IKE Initiator: New Phase 1, Intf 2, IKE Peer xxxxxxxx local Proxy Address xxxxxxxx, remote Proxy Address xxxxxxxx, SA (L2L: 1A) [… same continues for another 15 minutes …] 28093 03/08/2010 10:19:46.710 SEV=4 IKE/41 RPT=16140 xxxxxxxx IKE Initiator: New Phase 1, Intf 2, IKE Peer xxxxxxxx local Proxy Address xxxxxxxx, remote Proxy Address xxxxxxxx, SA (L2L: 1A) 28096 03/08/2010 10:20:17.720 SEV=5 IKE/172 RPT=1291 xxxxxxxx Group [xxxxxxxx] Automatic NAT Detection Status: Remote end is NOT behind a NAT device This end IS behind a NAT device 28100 03/08/2010 10:20:17.820 SEV=3 IKE/134 RPT=79 xxxxxxxx Group [xxxxxxxx] Mismatch: Configured LAN-to-LAN proposal differs from negotiated proposal. Verify local and remote LAN-to-LAN connection lists. 28103 03/08/2010 10:20:17.820 SEV=4 IKE/119 RPT=1197 xxxxxxxx Group [xxxxxxxx] PHASE 1 COMPLETED 28104 03/08/2010 10:20:17.820 SEV=4 AUTH/22 RPT=1031 xxxxxxxx User [xxxxxxxx] Group [xxxxxxxx] connected, Session Type: IPSec/LAN- to-LAN 28106 03/08/2010 10:20:17.820 SEV=4 AUTH/84 RPT=39 LAN-to-LAN tunnel to headend device xxxxxxxx connected 28110 03/08/2010 10:20:17.920 SEV=5 IKE/25 RPT=1291 xxxxxxxx Group [xxxxxxxx] Received remote Proxy Host data in ID Payload: Address xxxxxxxx, Protocol 0, Port 0 28113 03/08/2010 10:20:17.920 SEV=5 IKE/24 RPT=88 xxxxxxxx Group [xxxxxxxx] Received local Proxy Host data in ID Payload: Address xxxxxxxx, Protocol 0, Port 0 28116 03/08/2010 10:20:17.920 SEV=5 IKE/66 RPT=1290 xxxxxxxx Group [xxxxxxxx] IKE Remote Peer configured for SA: L2L: 1A 28117 03/08/2010 10:20:17.930 SEV=5 IKE/25 RPT=1292 xxxxxxxx Group [xxxxxxxx] Received remote Proxy Host data in ID Payload: Address xxxxxxxx, Protocol 0, Port 0 28120 03/08/2010 10:20:17.930 SEV=5 IKE/24 RPT=89 xxxxxxxx Group [xxxxxxxx] Received local Proxy Host data in ID Payload: Address xxxxxxxx, Protocol 0, Port 0 28123 03/08/2010 10:20:17.930 SEV=5 IKE/66 RPT=1291 xxxxxxxx Group [xxxxxxxx] IKE Remote Peer configured for SA: L2L: 1A 28124 03/08/2010 10:20:18.070 SEV=4 IKE/173 RPT=17330 xxxxxxxx Group [xxxxxxxx] NAT-Traversal successfully negotiated! IPSec traffic will be encapsulated to pass through NAT devices. 28127 03/08/2010 10:20:18.070 SEV=4 IKE/49 RPT=17332 xxxxxxxx Group [xxxxxxxx] Security negotiation complete for LAN-to-LAN Group (xxxxxxxx) Responder, Inbound SPI = 0x56a4fe5c, Outbound SPI = 0xcdfc3892 28130 03/08/2010 10:20:18.070 SEV=4 IKE/120 RPT=17332 xxxxxxxx Group [xxxxxxxx] PHASE 2 COMPLETED (msgid=37b3b298) 28131 03/08/2010 10:20:18.750 SEV=4 IKE/41 RPT=16141 xxxxxxxx Group [xxxxxxxx] IKE Initiator: New Phase 2, Intf 2, IKE Peer xxxxxxxx local Proxy Address xxxxxxxx, remote Proxy Address xxxxxxxx, SA (L2L: 1A) 28135 03/08/2010 10:20:18.870 SEV=4 IKE/173 RPT=17331 xxxxxxxx Group [xxxxxxxx] NAT-Traversal successfully negotiated! IPSec traffic will be encapsulated to pass through NAT devices.

    Read the article

  • fatal error occured while trying to sysprep the machine windows 8.1

    - by Mick
    I try do sysprep in Windows 8.1 I have create unattend.xml <settings pass="oobeSystem"> <component name="Microsoft-Windows-International-Core" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35" language="neutral" versionScope="nonSxS" xmlns:wcm="http://schemas.microsoft.com/WMIConfig/2002/State" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <InputLocale>en-US</InputLocale> <SystemLocale>en-US</SystemLocale> <UILanguage>en-US</UILanguage> <UILanguageFallback>en-US</UILanguageFallback> <UserLocale>en-US</UserLocale> </component> <component name="Microsoft-Windows-Shell-Setup" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35" language="neutral" versionScope="nonSxS" xmlns:wcm="http://schemas.microsoft.com/WMIConfig/2002/State" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <OEMInformation> <Manufacturer>XYZ</Manufacturer> <SupportURL>http://www.XYZ.com</SupportURL> </OEMInformation> <OOBE> <HideEULAPage>true</HideEULAPage> <NetworkLocation>Work</NetworkLocation> <ProtectYourPC>1</ProtectYourPC> </OOBE> <UserAccounts> <AdministratorPassword> <Value>XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX</Value> <PlainText>false</PlainText> </AdministratorPassword> <LocalAccounts> <LocalAccount wcm:action="add"> <Password> <Value>XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX</Value> <PlainText>false</PlainText> </Password> <Description>Admin</Description> <DisplayName>Admin</DisplayName> <Group>Administrators</Group> <Name>Admin</Name> </LocalAccount> </LocalAccounts> </UserAccounts> <WindowsFeatures> <ShowWindowsMediaPlayer>false</ShowWindowsMediaPlayer> <ShowMediaCenter>false</ShowMediaCenter> </WindowsFeatures> <RegisteredOrganization>XXXXXXXXXXXXXXXXXXXXX</RegisteredOrganization> <RegisteredOwner>XXXXXXXXXXXXXXXXXXXXXXX</RegisteredOwner> <TimeZone>Central European Standard Time</TimeZone> <ShowWindowsLive>false</ShowWindowsLive> </component> </settings> <settings pass="specialize"> <component name="Microsoft-Windows-Shell-Setup" processorArchitecture="amd64" publicKeyToken="31bf3856ad364e35" language="neutral" versionScope="nonSxS" xmlns:wcm="http://schemas.microsoft.com/WMIConfig/2002/State" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <RegisteredOrganization>XXXXXXXXXXXXXXXXXXXXXXXXX</RegisteredOrganization> <RegisteredOwner>XXXXXXXXXXXXXXXXXXXXXXXXXX</RegisteredOwner> <ProductKey>XXXXXXXXXXXXXXXXXXXXXXXXXXXXX</ProductKey> </component> </settings> And then I run sysprep.exe /oobe /generalize /shutdown I see this error: fatal error occurred while trying to sysprep the machine

    Read the article

  • what is the best mid/high-end class audio/music creation audio sound card?

    - by Chris
    Hello, I have a computershop myself, and I repair computers. But one of the things I really don't know (yet) is the performace od audio cards for music creation with midi. I have searched and searched and came up with some good reviews, but after browsing for a couple of hours I could't see the trees trough the forrest :-D (it's a dutch expression) At one moment I thought the M-Audio - Delta 1010LT would be a good PCIe card, later on I read that this card was released years ago. (but that could be false information) Also any personal expierence would be great, but not necessairy. I have searched a few cards, and I hope someone can help me make a choice for a friend of mine. He's buget is between $100 and $350 I know there are audio cards from $ 500 - $1850,- this is just too expensive. The following specs are crucial: ASIO Midi Mic in minimal 5.1, 7.1 recommended it's not for airplay, but just to compose music at home. using Ableton and midi keyboard. 1. M-Audio - Delta 1010LT: 8 x 8 analog I/O 2 mic preamps or line inputs S/PDIF digital I/O (coaxial) with 2-channel PCM SCMS copy protection control digital I/O supports surround-encoded AC-3 and DTS pass-through 1 x 1 MIDI I/O directly drive up to 7.1 surround (bass management software included) software controlled 36-bit internal DSP digital mixing/routing +4dbu/-10dBV operation individually switched in software word clock I/O for sample accurate device synchronization 2. RME HDSP 9632: * Stereo Analog Ein- und Ausgang, symmetrisch*, 24-Bit/192kHz, > 110 dB SNR * Optionale Erweiterungsboards mit je 4 symmetrischen Ein- und Ausgängen * Alle analogen I/Os voll 192 kHz-fähig, also keine Reduzierung der Kanalzahl * 1 x ADAT Digital In/Out, 96 kHz-fähig (S/MUX) * 1 x SPDIF Digital In/Out, 192 kHz-fähig * 1 x Breakout Kabel für koaxialen SPDIF-Betrieb* * Also bis zu 16 Ein-und Ausgänge gleichzeitig nutzbar! * 1 x Stereo Kopfhörerausgang, parallel zum analogen Ausgang, aber eigene Pegelanpassung * 1 x MIDI I/O für 16 Kanäle Hi-Speed MIDI über Breakout Kabel * DIGICheck, RMEs einzigartiges Meter- und Analysetool mit Spectral Analyser, Professionelle Level Meter 2/8/16-Kanalig, Vector Audio Scope und diversen weiteren Analysefunktionen * HDSP Meter Bridge: Frei skalierbare Levelmeter mit Peak- und RMS Berechnung in Hardware * TotalMix: 512-Kanal Mischer mit 40 Bit interner Auflösung 3. EMU 1212M (1212 M) PCIe: * Top kwaliteit convertors 24-bit/192kHz convertors. * Hardware gestuurde effecten. * DSP zero-latency hardware mixen en monitoring. * Analoge en digitale I/O plus MIDI. * EMU Production Tools Software Bundle - Cakewalk SONAR , Steinberg Cubase LE, Ableton Live E-MU Edition **EMU 1212M PCI-e inputs/outputs:** * 2 balanced jack inputs. * 2 balanced jack outputs. * 24-bit/192kHz ADAT I/O. * 24-bit/192kHz Coaxiale S/PDif I/O switchable to AES/EBU. * MIDI I/O. 4. M-Audio Audiophile 192: - Up to 24-bit/192kHz audio - 2 balanced analog inputs (1/4” TRS) - 2 balanced analog outputs (1/4” TRS) - S/PDIF digital I/O (coaxial RCA connectors) with 2-channel PCM - SCMS copy protection control - Digital I/O supports surround-encoded AC-3 and DTS pass-through - Direct hardware input monitoring via separate balanced 1/4” TRS monitor outputs - Software routing of inputs and outputs - Digital I/O can be routed to/from external effects - 16-channel MIDI I/O - ASIO, WDM, GSIF 2 and Core Audio driver support for compatibility with most applications - 64-bit driver support for Windows - PCI 2.2 compatibility - Apple G5 compatible - Incompatible exceptions - Includes Ableton Live Lite music production software, so you can make music right away - Works with other Delta cards Technical Specifcations: - Compatibility - ASIO - WDM - GSIF 2 - Core Audio

    Read the article

  • phpMyAdmin setup issues

    - by EquinoX
    I am trying to follow the tutorial here to setup the user and pass. It says there that "this section is only applicable if your MySQL server is running with --skip-show-database". First question is, how do I check if MySQl server is running with --skip-show-database? Is there any way I can access phpMyAdmin SQL query window without logging in? Otherwise I'd have to execute this SQL from command line. I am also getting this: Cannot load mcrypt extension. Please check your PHP configuration. I have added mcrypt.so to php.ini and doing the following command proves that I have it. [root@DT html]# rpm -qa | grep mcrypt mcrypt-2.6.8-1.el5 php-mcrypt-5.3.5-1.1.w5 libmcrypt-2.5.8-4.el5.centos [root@DT html]# php -v PHP 5.3.5 (cli) (built: Feb 19 2011 13:10:09) Copyright (c) 1997-2010 The PHP Group Zend Engine v2.3.0, Copyright (c) 1998-2010 Zend Technologies Now when I go to phpinfo() and search for mcrypt it can find it inside the Configure Command row ('--with-mcrypt=shared,/usr'). So, what to do next?. UPDATE: I didn't put extension=mcrypt.so in php.ini as it will complain the following: PHP Warning: Module 'mcrypt' already loaded in Unknown on line 0 Here's my nginx.conf: #user nobody; worker_processes 2; #error_log logs/error.log; #error_log logs/error.log notice; #error_log logs/error.log info; #pid logs/nginx.pid; events { worker_connections 1024; } http { include mime.types; default_type application/octet-stream; #log_format main '$remote_addr - $remote_user [$time_local] "$request" ' # '$status $body_bytes_sent "$http_referer" ' # '"$http_user_agent" "$http_x_forwarded_for"'; #access_log logs/access.log main; sendfile on; #tcp_nopush on; #keepalive_timeout 0; keepalive_timeout 65; gzip on; server { listen 80; root /usr/share/nginx/html; server_name localhost; #charset koi8-r; #access_log logs/host.access.log main; location / { #root html; index index.html index.htm; } #error_page 404 /404.html; # redirect server error pages to the static page /50x.html # error_page 500 502 503 504 /50x.html; location = /50x.html { #root html; } # proxy the PHP scripts to Apache listening on 127.0.0.1:80 # #location ~ \.php$ { # proxy_pass http://127.0.0.1; #} # pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000 # location ~ \.php$ { #root /usr/local/nginx/html; fastcgi_pass 127.0.0.1:9000; fastcgi_index index.php; fastcgi_param SCRIPT_FILENAME /usr/share/nginx/html$fastcgi_script _name; include fastcgi_params; } # deny access to .htaccess files, if Apache's document root # concurs with nginx's one location ~ /\.ht { deny all; } } # another virtual host using mix of IP-, name-, and port-based configuration # #server { # listen 8000; # listen somename:8080; # server_name somename alias another.alias; # location / { # root html; # index index.html index.htm; # } #} # HTTPS server # #server { # listen 443; # server_name localhost; # ssl on; # ssl_certificate cert.pem; # ssl_certificate_key cert.key; # ssl_session_timeout 5m; # ssl_protocols SSLv2 SSLv3 TLSv1; # ssl_ciphers ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP; # ssl_prefer_server_ciphers on; # location / { # root html; # index index.html index.htm; # } #} }

    Read the article

  • Nginx Multiple Domains

    - by showFocus
    I am trying to add a second virtual host to nginx. When i go to the new domain it redirects to the old one. I have tried restarting Nginx, rebooting the server. Has anyone come across this before, care to share? File: nginx.conf ### user www-data www-data; worker_processes 4; events { worker_connections 1024; } http { include mime.types; default_type application/octet-stream; sendfile on; tcp_nopush on; tcp_nodelay off; keepalive_timeout 5; gzip on; gzip_comp_level 2; gzip_proxied any; gzip_types text/plain text/css application/x-javascript text/xml application/xml application/xml+rss text/javascript; include /usr/local/nginx/sites-enabled/*; } File: ../sites-enabled/domain1.co.uk server { listen 80; server_name www.domain1.co.uk; rewrite ^/(.*) http://domain1.co.uk/$1 permanent; } server { listen 80; server_name domain1.co.uk; access_log /home/me/public_html/domain1.co.uk/log/access.log; error_log /home/me/public_html/domain1.co.uk/log/error.log; location / { root /home/me/public_html/domain1.co.uk/public/; index index.php index.html; # WordPress supercache & permalinks. include /usr/local/nginx/conf/wordpress_params.super_cache; } # pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000 location ~ \.php$ { fastcgi_pass 127.0.0.1:9000; fastcgi_index index.php; include /usr/local/nginx/conf/fastcgi_params; fastcgi_param SCRIPT_FILENAME /home/me/public_html/domain1.co.uk/public/$fastcgi_script_name; } } File: ../sites-enabled/domain2.co.uk server { listen 80; server_name www.domain2.co.uk; rewrite ^/(.*) http://domain2.co.uk/$1 permanent; } server { listen 80; server_name domain2.co.uk; access_log /home/me/public_html/domain2.co.uk/log/access.log; error_log /home/me/public_html/domain2.co.uk/log/error.log; location / { root /home/me/public_html/domain2.co.uk/public/; index index.php index.html; # Basic version of WordPress parameters, supporting nice permalinks. # include /usr/local/nginx/conf/wordpress_params.regular; # Advanced version of WordPress parameters supporting nice permalinks and WP Super Cache plugin include /usr/local/nginx/conf/wordpress_params.super_cache; } # pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000 # location ~ \.php$ { fastcgi_pass 127.0.0.1:9000; fastcgi_index index.php; include /usr/local/nginx/conf/fastcgi_params; fastcgi_param SCRIPT_FILENAME /home/me/public_html/domain2/public/$fastcgi_script_name; } }

    Read the article

  • Using jQuery and SPServices to Display List Items

    - by Bil Simser
    I had an interesting challenge recently that I turned to Marc Anderson’s wonderful SPServices project for. If you haven’t already seen or used SPServices, please do. It’s a jQuery library that does primarily two things. First, it wraps up all of the SharePoint web services in a nice little AJAX wrapper for use in JavaScript. Second, it enhances the form editing of items in SharePoint so you’re not hacking up your List Form pages. My challenge was simple but interesting. The user wanted to display a SharePoint item page (DispForm.aspx, which already had some customization on it to display related items via this blog post from Codeless Solutions for SharePoint) but launch from an external application using the value of one of the fields in the SharePoint list. For simplicity let’s say my list is a list of customers and the related list is a list of orders for that customer. It would look something like this (click on the item to see the full image): Your first thought might be, that’s easy! Display the customer information using a DataView Web Part and filter the item using a query string to match the customer number. However there are a few problems with this idea: You’ll need to build a custom page and then attach that related orders view to it. This is a bit of a problem because the solution from Codeless Solutions relies on the Title field on the page to be displayed. On a custom page you would have to recreate all of the elements found on the DispForm.aspx page so the related view would work. The DataView Web Part doesn’t look *exactly* like what the out of the box display form page does. Not a huge problem and can be overcome with some CSS style overrides but still, more work. A DVWP showing a single record doesn’t have the same toolbar that you would using the DispForm.aspx. Not a show-stopper and you can rebuild the toolbar but it’s going to potentially require code and then there’s the security trimming, etc. that you have to get right. DVWPs are not automatically updated if you add a column to the list like DispForm.aspx is. Work, work, work. For these reasons I thought it would be easier to take the already existing (modified) DispForm.aspx page and just add some jQuery magic to the page to find the item. Why do we need to find it? DispForm.aspx relies on a querystring parameter called “ID” which then displays whatever that item ID number is in the list. Trouble is, when you’re coming in from an external app via a link, you don’t know what that internal ID is (and frankly shouldn’t). I don’t like exposing internal SharePoint IDs to the outside world for the same reason I don’t do it with database IDs. They’re internal and while it’s find to use on the site itself you don’t want external links using it. It’s volatile and can change (delete one item then re-add it back with the same data and watch any ID references break). The next thought might be to call a SharePoint web service with a CAML query to get the item ID number using some criteria (in this case, the customer number). That’s great if you have that ability but again we had an existing application we were just adding a link to. The last thing I wanted to do was to crack open the code on that sucker and start calling web services (primarily because it’s Java, but really I’m a lazy geek). However if you’re doing this and have access to call a web service that would be an option. Back to this problem, how do I a) find a SharePoint List Item based on some field value other than ID and b) make it low impact so I can just construct a URL to it? That’s where jQuery and SPServices came to the rescue. After spending a few hours of emails back and forth with Marc and a couple of phone calls (and updating jQuery to the latest version, duh!) it was a simple answer. First we need a reference to a) jQuery b) SPServices and c) our script. I just dropped a Content Editor Web Part, the Swiss Army Knives of Web Parts, onto the DispForm.aspx page and added these lines: <script type="text/javascript" src="http://intranet/JavaScript/jquery-1.4.2.min.js"></script> <script type="text/javascript" src="http://intranet/JavaScript/jquery.SPServices-0.5.3.min.js"></script> <script type="text/javascript" src="http://intranet/JavaScript/RedirectToID.js"> </script> Update it to point to where you keep your scripts located. I prefer to keep them all in Document Libraries as I can make changes to them without having to remote into the server (and on a multiple web front end, that’s just a PITA), it provides me with version control of sorts, and it’s quick to add new plugins and scripts. Now we can look at our RedirectToID.js script. This invokes the SPServices Library to call the GetListItems method of the Lists web service and then rewrites the URL to DispForm.aspx to use the correct SharePoint ID (the internal one). $(document).ready(function(){ var queryStringValues = $().SPServices.SPGetQueryString(); var id = queryStringValues["ID"]; if(id == "0") { var customer = queryStringValues["CustomerNumber"]; var query = "<Query><Where><Eq><FieldRef Name='CustomerNumber'/><Value Type='Text'>" + customer + "</Value></Eq></Where></Query>"; var url = window.location; $().SPServices({ operation: "GetListItems", listName: "Customers", async: false, CAMLQuery: query, completefunc: function (xData, Status) { $(xData.responseXML).find("[nodeName=z:row]").each(function(){ id = $(this).attr("ows_ID"); url = $().SPServices.SPGetCurrentSite() + "/Lists/Customers/DispForm.aspx?ID=" + id; window.location = url; }); } }); } }); What’s happening here? Line 3: We call SPServices.SPGetQueryString to get an array of query string values (a handy function in the library as I had 15 lines of code to do this which is now gone). Line 4: Extract the ID value from the query string Line 6: If we pass in “0” it means we’re looking up a field value. This allows DispForm.aspx to work like normal with SharePoint lists but lookup our values when invoked. Why ID at all? DispForm.aspx doesn’t work unless you pass in something and “0” is a *magic* number that will invoke the page but not lookup a value in the database. Line 8-15: Extract the CustomerNumber query string value, build a CAML query to find it then call the GetListitems method using SPServices Line 16: Process the results in our completefunc to iterate over all the rows (there should only be one) and extract the real ID of the item Line 17-20: Build a new URL based on the site (using a call to SPGetCurrentSite) and append our real ID to redirect to the DispForm.aspx page As you can see, it dynamically creates a CAML query for the call to the web service using the passed in value. You could even make this generic to take in different query strings, one for the field name to search for and the other for the value to find. That way it could be used for any field you want. For example you could bring up the correct item on the DispForm.aspx page based on customer name with something like this: http://myserver/Lists/Customers/DispForm.aspx?ID=0&FilterId=CustomerName&FilterValue=Sony Use your imagination. Some people would opt for building a custom page with a DVWP but if you want to leverage all the functionality of DispForm.aspx this might come in handy if you don’t want to rely on internal SharePoint IDs.

    Read the article

  • javafx tableview get selected data from ObservableList

    - by user3717821
    i am working on a javafx project and i need your help . while i am trying to get selected data from table i can get selected data from normal cell but can't get data from ObservableList inside tableview. code for my database: -- phpMyAdmin SQL Dump -- version 4.0.4 -- http://www.phpmyadmin.net -- -- Host: localhost -- Generation Time: Jun 10, 2014 at 06:20 AM -- Server version: 5.1.33-community -- PHP Version: 5.4.12 SET SQL_MODE = "NO_AUTO_VALUE_ON_ZERO"; SET time_zone = "+00:00"; /*!40101 SET @OLD_CHARACTER_SET_CLIENT=@@CHARACTER_SET_CLIENT */; /*!40101 SET @OLD_CHARACTER_SET_RESULTS=@@CHARACTER_SET_RESULTS */; /*!40101 SET @OLD_COLLATION_CONNECTION=@@COLLATION_CONNECTION */; /*!40101 SET NAMES utf8 */; -- -- Database: `test` -- -- -------------------------------------------------------- -- -- Table structure for table `customer` -- CREATE TABLE IF NOT EXISTS `customer` ( `col0` int(11) NOT NULL, `col1` varchar(255) DEFAULT NULL, `col2` int(11) DEFAULT NULL, PRIMARY KEY (`col0`) ) ENGINE=InnoDB DEFAULT CHARSET=latin1; -- -- Dumping data for table `customer` -- INSERT INTO `customer` (`col0`, `col1`, `col2`) VALUES (12, 'adasdasd', 231), (22, 'adasdasd', 231), (212, 'adasdasd', 231); /*!40101 SET CHARACTER_SET_CLIENT=@OLD_CHARACTER_SET_CLIENT */; /*!40101 SET CHARACTER_SET_RESULTS=@OLD_CHARACTER_SET_RESULTS */; /*!40101 SET COLLATION_CONNECTION=@OLD_COLLATION_CONNECTION */; my javafx codes: import java.sql.Connection; import java.sql.DriverManager; import java.sql.ResultSet; import java.sql.SQLException; import java.util.Map; import javafx.application.Application; import javafx.beans.property.SimpleStringProperty; import javafx.beans.value.ChangeListener; import javafx.beans.value.ObservableValue; import javafx.collections.FXCollections; import javafx.collections.ObservableList; import javafx.event.ActionEvent; import javafx.event.EventHandler; import javafx.scene.Scene; import javafx.scene.control.Button; import javafx.scene.control.TableCell; import javafx.scene.control.TableColumn; import javafx.scene.control.TableColumn.CellDataFeatures; import javafx.scene.control.TablePosition; import javafx.scene.control.TableView; import javafx.scene.control.TableView.TableViewSelectionModel; import javafx.scene.control.cell.ChoiceBoxTableCell; import javafx.scene.control.cell.TextFieldTableCell; import javafx.scene.layout.BorderPane; import javafx.stage.Stage; import javafx.util.Callback; import javafx.util.StringConverter; class DBConnector { private static Connection conn; private static String url = "jdbc:mysql://localhost/test"; private static String user = "root"; private static String pass = "root"; public static Connection connect() throws SQLException{ try{ Class.forName("com.mysql.jdbc.Driver").newInstance(); }catch(ClassNotFoundException cnfe){ System.err.println("Error: "+cnfe.getMessage()); }catch(InstantiationException ie){ System.err.println("Error: "+ie.getMessage()); }catch(IllegalAccessException iae){ System.err.println("Error: "+iae.getMessage()); } conn = DriverManager.getConnection(url,user,pass); return conn; } public static Connection getConnection() throws SQLException, ClassNotFoundException{ if(conn !=null && !conn.isClosed()) return conn; connect(); return conn; } } public class DynamicTable extends Application{ Object newValue; //TABLE VIEW AND DATA private ObservableList<ObservableList> data; private TableView<ObservableList> tableview; //MAIN EXECUTOR public static void main(String[] args) { launch(args); } //CONNECTION DATABASE public void buildData(){ tableview.setEditable(true); Callback<TableColumn<Map, String>, TableCell<Map, String>> cellFactoryForMap = new Callback<TableColumn<Map, String>, TableCell<Map, String>>() { @Override public TableCell call(TableColumn p) { return new TextFieldTableCell(new StringConverter() { @Override public String toString(Object t) { return t.toString(); } @Override public Object fromString(String string) { return string; } }); } }; Connection c ; data = FXCollections.observableArrayList(); try{ c = DBConnector.connect(); //SQL FOR SELECTING ALL OF CUSTOMER String SQL = "SELECT * from CUSTOMer"; //ResultSet ResultSet rs = c.createStatement().executeQuery(SQL); /********************************** * TABLE COLUMN ADDED DYNAMICALLY * **********************************/ for(int i=0 ; i<rs.getMetaData().getColumnCount(); i++){ //We are using non property style for making dynamic table final int j = i; TableColumn col = new TableColumn(rs.getMetaData().getColumnName(i+1)); if(j==1){ final ObservableList<String> logLevelList = FXCollections.observableArrayList("FATAL", "ERROR", "WARN", "INFO", "INOUT", "DEBUG"); col.setCellFactory(ChoiceBoxTableCell.forTableColumn(logLevelList)); tableview.getColumns().addAll(col); } else{ col.setCellValueFactory(new Callback<CellDataFeatures<ObservableList,String>,ObservableValue<String>>(){ public ObservableValue<String> call(CellDataFeatures<ObservableList, String> param) { return new SimpleStringProperty(param.getValue().get(j).toString()); } }); tableview.getColumns().addAll(col); } if(j!=1) col.setCellFactory(cellFactoryForMap); System.out.println("Column ["+i+"] "); } /******************************** * Data added to ObservableList * ********************************/ while(rs.next()){ //Iterate Row ObservableList<String> row = FXCollections.observableArrayList(); for(int i=1 ; i<=rs.getMetaData().getColumnCount(); i++){ //Iterate Column row.add(rs.getString(i)); } System.out.println("Row [1] added "+row ); data.add(row); } //FINALLY ADDED TO TableView tableview.setItems(data); }catch(Exception e){ e.printStackTrace(); System.out.println("Error on Building Data"); } } @Override public void start(Stage stage) throws Exception { //TableView Button showDataButton = new Button("Add"); showDataButton.setOnAction(new EventHandler<ActionEvent>() { public void handle(ActionEvent event) { ObservableList<String> row = FXCollections.observableArrayList(); for(int i=1 ; i<=3; i++){ //Iterate Column row.add("asdasd"); } data.add(row); //FINALLY ADDED TO TableView tableview.setItems(data); } }); tableview = new TableView(); buildData(); //Main Scene BorderPane root = new BorderPane(); root.setCenter(tableview); root.setBottom(showDataButton); Scene scene = new Scene(root,500,500); stage.setScene(scene); stage.show(); tableview.getSelectionModel().selectedItemProperty().addListener(new ChangeListener() { @Override public void changed(ObservableValue observableValue, Object oldValue, Object newValue) { //Check whether item is selected and set value of selected item to Label if (tableview.getSelectionModel().getSelectedItem() != null) { TableViewSelectionModel selectionModel = tableview.getSelectionModel(); ObservableList selectedCells = selectionModel.getSelectedCells(); TablePosition tablePosition = (TablePosition) selectedCells.get(0); Object val = tablePosition.getTableColumn().getCellData(newValue); System.out.println("Selected Value " + val); System.out.println("Selected row " + newValue); } } }); } } please help me..

    Read the article

  • AutoMapper MappingFunction from Source Type of NameValueCollection

    - by REA_ANDREW
    I have had a situation arise today where I need to construct a complex type from a source of a NameValueCollection.  A little while back I submitted a patch for the Agatha Project to include REST (JSON and XML) support for the service contract.  I realized today that as useful as it is, it did not actually support true REST conformance, as REST should support GET so that you can use JSONP from JavaScript directly meaning you can query cross domain services.  My original implementation for POX and JSON used the POST method and this immediately rules out JSONP as from reading, JSONP only works with GET Requests. This then raised another issue.  The current operation contract of Agatha and one of its main benefits is that you can supply an array of Request objects in a single request, limiting the about of server requests you need to make.  Now, at the present time I am thinking that this will not be the case for the REST imlementation but will yield the benefits of the fact that : The same Request objects can be used for SOAP and RST (POX, JSON) The construct of the JavaScript functions will be simpler and more readable It will enable the use of JSONP for cross domain REST Services The current contract for the Agatha WcfRequestProcessor is at time of writing the following: [ServiceContract] public interface IWcfRequestProcessor { [OperationContract(Name = "ProcessRequests")] [ServiceKnownType("GetKnownTypes", typeof(KnownTypeProvider))] [TransactionFlow(TransactionFlowOption.Allowed)] Response[] Process(params Request[] requests); [OperationContract(Name = "ProcessOneWayRequests", IsOneWay = true)] [ServiceKnownType("GetKnownTypes", typeof(KnownTypeProvider))] void ProcessOneWayRequests(params OneWayRequest[] requests); }   My current proposed solution, and at the very early stages of my concept is as follows: [ServiceContract] public interface IWcfRestJsonRequestProcessor { [OperationContract(Name="process")] [ServiceKnownType("GetKnownTypes", typeof(KnownTypeProvider))] [TransactionFlow(TransactionFlowOption.Allowed)] [WebGet(UriTemplate = "process/{name}/{*parameters}", BodyStyle = WebMessageBodyStyle.WrappedResponse, ResponseFormat = WebMessageFormat.Json)] Response[] Process(string name, NameValueCollection parameters); [OperationContract(Name="processoneway",IsOneWay = true)] [ServiceKnownType("GetKnownTypes", typeof(KnownTypeProvider))] [WebGet(UriTemplate = "process-one-way/{name}/{*parameters}", BodyStyle = WebMessageBodyStyle.WrappedResponse, ResponseFormat = WebMessageFormat.Json)] void ProcessOneWayRequests(string name, NameValueCollection parameters); }   Now this part I have not yet implemented, it is the preliminart step which I have developed which will allow me to take the name of the Request Type and the NameValueCollection and construct the complex type which is that of the Request which I can then supply to a nested instance of the original IWcfRequestProcessor  and work as it should normally.  To give an example of some of the urls which you I envisage with this method are: http://www.url.com/service.svc/json/process/getweather/?location=london http://www.url.com/service.svc/json/process/getproductsbycategory/?categoryid=1 http://www.url.om/service.svc/json/process/sayhello/?name=andy Another reason why my direction has gone to a single request for the REST implementation is because of restrictions which are imposed by browsers on the length of the url.  From what I have read this is on average 2000 characters.  I think that this is a very acceptable usage limit in the context of using 1 request, but I do not think this is acceptable for accommodating multiple requests chained together.  I would love to be corrected on that one, I really would but unfortunately from what I have read I have come to the conclusion that this is not the case. The mapping function So, as I say this is just the first pass I have made at this, and I am not overly happy with the try catch for detecting types without default constructors.  I know there is a better way but for the minute, it escapes me.  I would also like to know the correct way for adding mapping functions and not using the anonymous way that I have used.  To achieve this I have used recursion which I am sure is what other mapping function use. As you do have to go as deep as the complex type is. public static object RecurseType(NameValueCollection collection, Type type, string prefix) { try { var returnObject = Activator.CreateInstance(type); foreach (var property in type.GetProperties()) { foreach (var key in collection.AllKeys) { if (String.IsNullOrEmpty(prefix) || key.Length > prefix.Length) { var propertyNameToMatch = String.IsNullOrEmpty(prefix) ? key : key.Substring(property.Name.IndexOf(prefix) + prefix.Length + 1); if (property.Name == propertyNameToMatch) { property.SetValue(returnObject, Convert.ChangeType(collection.Get(key), property.PropertyType), null); } else if(property.GetValue(returnObject,null) == null) { property.SetValue(returnObject, RecurseType(collection, property.PropertyType, String.Concat(prefix, property.PropertyType.Name)), null); } } } } return returnObject; } catch (MissingMethodException) { //Quite a blunt way of dealing with Types without default constructor return null; } }   Another thing is performance, I have not measured this in anyway, it is as I say the first pass, so I hope this can be the start of a more perfected implementation.  I tested this out with a complex type of three levels, there is no intended logical meaning to the properties, they are simply for the purposes of example.  You could call this a spiking session, as from here on in, now I know what I am building I would take a more TDD approach.  OK, purists, why did I not do this from the start, well I didn’t, this was a brain dump and now I know what I am building I can. The console test and how I used with AutoMapper is as follows: static void Main(string[] args) { var collection = new NameValueCollection(); collection.Add("Name", "Andrew Rea"); collection.Add("Number", "1"); collection.Add("AddressLine1", "123 Street"); collection.Add("AddressNumber", "2"); collection.Add("AddressPostCodeCountry", "United Kingdom"); collection.Add("AddressPostCodeNumber", "3"); AutoMapper.Mapper.CreateMap<NameValueCollection, Person>() .ConvertUsing(x => { return(Person) RecurseType(x, typeof(Person), null); }); var person = AutoMapper.Mapper.Map<NameValueCollection, Person>(collection); Console.WriteLine(person.Name); Console.WriteLine(person.Number); Console.WriteLine(person.Address.Line1); Console.WriteLine(person.Address.Number); Console.WriteLine(person.Address.PostCode.Country); Console.WriteLine(person.Address.PostCode.Number); Console.ReadLine(); }   Notice the convention that I am using and that this method requires you do use.  Each property is prefixed with the constructed name of its parents combined.  This is the convention used by AutoMapper and it makes sense. I can also think of other uses for this including using with ASP.NET MVC ModelBinders for creating a complex type from the QueryString which is itself is a NameValueCollection. Hope this is of some help to people and I would welcome any code reviews you could give me. References: Agatha : http://code.google.com/p/agatha-rrsl/ AutoMapper : http://automapper.codeplex.com/   Cheers for now, Andrew   P.S. I will have the proposed solution for a more complete REST implementation for AGATHA very soon. 

    Read the article

  • Improving Partitioned Table Join Performance

    - by Paul White
    The query optimizer does not always choose an optimal strategy when joining partitioned tables. This post looks at an example, showing how a manual rewrite of the query can almost double performance, while reducing the memory grant to almost nothing. Test Data The two tables in this example use a common partitioning partition scheme. The partition function uses 41 equal-size partitions: CREATE PARTITION FUNCTION PFT (integer) AS RANGE RIGHT FOR VALUES ( 125000, 250000, 375000, 500000, 625000, 750000, 875000, 1000000, 1125000, 1250000, 1375000, 1500000, 1625000, 1750000, 1875000, 2000000, 2125000, 2250000, 2375000, 2500000, 2625000, 2750000, 2875000, 3000000, 3125000, 3250000, 3375000, 3500000, 3625000, 3750000, 3875000, 4000000, 4125000, 4250000, 4375000, 4500000, 4625000, 4750000, 4875000, 5000000 ); GO CREATE PARTITION SCHEME PST AS PARTITION PFT ALL TO ([PRIMARY]); There two tables are: CREATE TABLE dbo.T1 ( TID integer NOT NULL IDENTITY(0,1), Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T1 PRIMARY KEY CLUSTERED (TID) ON PST (TID) );   CREATE TABLE dbo.T2 ( TID integer NOT NULL, Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T2 PRIMARY KEY CLUSTERED (TID, Column1) ON PST (TID) ); The next script loads 5 million rows into T1 with a pseudo-random value between 1 and 5 for Column1. The table is partitioned on the IDENTITY column TID: INSERT dbo.T1 WITH (TABLOCKX) (Column1) SELECT (ABS(CHECKSUM(NEWID())) % 5) + 1 FROM dbo.Numbers AS N WHERE n BETWEEN 1 AND 5000000; In case you don’t already have an auxiliary table of numbers lying around, here’s a script to create one with 10 million rows: CREATE TABLE dbo.Numbers (n bigint PRIMARY KEY);   WITH L0 AS(SELECT 1 AS c UNION ALL SELECT 1), L1 AS(SELECT 1 AS c FROM L0 AS A CROSS JOIN L0 AS B), L2 AS(SELECT 1 AS c FROM L1 AS A CROSS JOIN L1 AS B), L3 AS(SELECT 1 AS c FROM L2 AS A CROSS JOIN L2 AS B), L4 AS(SELECT 1 AS c FROM L3 AS A CROSS JOIN L3 AS B), L5 AS(SELECT 1 AS c FROM L4 AS A CROSS JOIN L4 AS B), Nums AS(SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) AS n FROM L5) INSERT dbo.Numbers WITH (TABLOCKX) SELECT TOP (10000000) n FROM Nums ORDER BY n OPTION (MAXDOP 1); Table T1 contains data like this: Next we load data into table T2. The relationship between the two tables is that table 2 contains ‘n’ rows for each row in table 1, where ‘n’ is determined by the value in Column1 of table T1. There is nothing particularly special about the data or distribution, by the way. INSERT dbo.T2 WITH (TABLOCKX) (TID, Column1) SELECT T.TID, N.n FROM dbo.T1 AS T JOIN dbo.Numbers AS N ON N.n >= 1 AND N.n <= T.Column1; Table T2 ends up containing about 15 million rows: The primary key for table T2 is a combination of TID and Column1. The data is partitioned according to the value in column TID alone. Partition Distribution The following query shows the number of rows in each partition of table T1: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T1 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are 40 partitions containing 125,000 rows (40 * 125k = 5m rows). The rightmost partition remains empty. The next query shows the distribution for table 2: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T2 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are roughly 375,000 rows in each partition (the rightmost partition is also empty): Ok, that’s the test data done. Test Query and Execution Plan The task is to count the rows resulting from joining tables 1 and 2 on the TID column: SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; The optimizer chooses a plan using parallel hash join, and partial aggregation: The Plan Explorer plan tree view shows accurate cardinality estimates and an even distribution of rows across threads (click to enlarge the image): With a warm data cache, the STATISTICS IO output shows that no physical I/O was needed, and all 41 partitions were touched: Running the query without actual execution plan or STATISTICS IO information for maximum performance, the query returns in around 2600ms. Execution Plan Analysis The first step toward improving on the execution plan produced by the query optimizer is to understand how it works, at least in outline. The two parallel Clustered Index Scans use multiple threads to read rows from tables T1 and T2. Parallel scan uses a demand-based scheme where threads are given page(s) to scan from the table as needed. This arrangement has certain important advantages, but does result in an unpredictable distribution of rows amongst threads. The point is that multiple threads cooperate to scan the whole table, but it is impossible to predict which rows end up on which threads. For correct results from the parallel hash join, the execution plan has to ensure that rows from T1 and T2 that might join are processed on the same thread. For example, if a row from T1 with join key value ‘1234’ is placed in thread 5’s hash table, the execution plan must guarantee that any rows from T2 that also have join key value ‘1234’ probe thread 5’s hash table for matches. The way this guarantee is enforced in this parallel hash join plan is by repartitioning rows to threads after each parallel scan. The two repartitioning exchanges route rows to threads using a hash function over the hash join keys. The two repartitioning exchanges use the same hash function so rows from T1 and T2 with the same join key must end up on the same hash join thread. Expensive Exchanges This business of repartitioning rows between threads can be very expensive, especially if a large number of rows is involved. The execution plan selected by the optimizer moves 5 million rows through one repartitioning exchange and around 15 million across the other. As a first step toward removing these exchanges, consider the execution plan selected by the optimizer if we join just one partition from each table, disallowing parallelism: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = 1 AND $PARTITION.PFT(T2.TID) = 1 OPTION (MAXDOP 1); The optimizer has chosen a (one-to-many) merge join instead of a hash join. The single-partition query completes in around 100ms. If everything scaled linearly, we would expect that extending this strategy to all 40 populated partitions would result in an execution time around 4000ms. Using parallelism could reduce that further, perhaps to be competitive with the parallel hash join chosen by the optimizer. This raises a question. If the most efficient way to join one partition from each of the tables is to use a merge join, why does the optimizer not choose a merge join for the full query? Forcing a Merge Join Let’s force the optimizer to use a merge join on the test query using a hint: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN); This is the execution plan selected by the optimizer: This plan results in the same number of logical reads reported previously, but instead of 2600ms the query takes 5000ms. The natural explanation for this drop in performance is that the merge join plan is only using a single thread, whereas the parallel hash join plan could use multiple threads. Parallel Merge Join We can get a parallel merge join plan using the same query hint as before, and adding trace flag 8649: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN, QUERYTRACEON 8649); The execution plan is: This looks promising. It uses a similar strategy to distribute work across threads as seen for the parallel hash join. In practice though, performance is disappointing. On a typical run, the parallel merge plan runs for around 8400ms; slower than the single-threaded merge join plan (5000ms) and much worse than the 2600ms for the parallel hash join. We seem to be going backwards! The logical reads for the parallel merge are still exactly the same as before, with no physical IOs. The cardinality estimates and thread distribution are also still very good (click to enlarge): A big clue to the reason for the poor performance is shown in the wait statistics (captured by Plan Explorer Pro): CXPACKET waits require careful interpretation, and are most often benign, but in this case excessive waiting occurs at the repartitioning exchanges. Unlike the parallel hash join, the repartitioning exchanges in this plan are order-preserving ‘merging’ exchanges (because merge join requires ordered inputs): Parallelism works best when threads can just grab any available unit of work and get on with processing it. Preserving order introduces inter-thread dependencies that can easily lead to significant waits occurring. In extreme cases, these dependencies can result in an intra-query deadlock, though the details of that will have to wait for another time to explore in detail. The potential for waits and deadlocks leads the query optimizer to cost parallel merge join relatively highly, especially as the degree of parallelism (DOP) increases. This high costing resulted in the optimizer choosing a serial merge join rather than parallel in this case. The test results certainly confirm its reasoning. Collocated Joins In SQL Server 2008 and later, the optimizer has another available strategy when joining tables that share a common partition scheme. This strategy is a collocated join, also known as as a per-partition join. It can be applied in both serial and parallel execution plans, though it is limited to 2-way joins in the current optimizer. Whether the optimizer chooses a collocated join or not depends on cost estimation. The primary benefits of a collocated join are that it eliminates an exchange and requires less memory, as we will see next. Costing and Plan Selection The query optimizer did consider a collocated join for our original query, but it was rejected on cost grounds. The parallel hash join with repartitioning exchanges appeared to be a cheaper option. There is no query hint to force a collocated join, so we have to mess with the costing framework to produce one for our test query. Pretending that IOs cost 50 times more than usual is enough to convince the optimizer to use collocated join with our test query: -- Pretend IOs are 50x cost temporarily DBCC SETIOWEIGHT(50);   -- Co-located hash join SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (RECOMPILE);   -- Reset IO costing DBCC SETIOWEIGHT(1); Collocated Join Plan The estimated execution plan for the collocated join is: The Constant Scan contains one row for each partition of the shared partitioning scheme, from 1 to 41. The hash repartitioning exchanges seen previously are replaced by a single Distribute Streams exchange using Demand partitioning. Demand partitioning means that the next partition id is given to the next parallel thread that asks for one. My test machine has eight logical processors, and all are available for SQL Server to use. As a result, there are eight threads in the single parallel branch in this plan, each processing one partition from each table at a time. Once a thread finishes processing a partition, it grabs a new partition number from the Distribute Streams exchange…and so on until all partitions have been processed. It is important to understand that the parallel scans in this plan are different from the parallel hash join plan. Although the scans have the same parallelism icon, tables T1 and T2 are not being co-operatively scanned by multiple threads in the same way. Each thread reads a single partition of T1 and performs a hash match join with the same partition from table T2. The properties of the two Clustered Index Scans show a Seek Predicate (unusual for a scan!) limiting the rows to a single partition: The crucial point is that the join between T1 and T2 is on TID, and TID is the partitioning column for both tables. A thread that processes partition ‘n’ is guaranteed to see all rows that can possibly join on TID for that partition. In addition, no other thread will see rows from that partition, so this removes the need for repartitioning exchanges. CPU and Memory Efficiency Improvements The collocated join has removed two expensive repartitioning exchanges and added a single exchange processing 41 rows (one for each partition id). Remember, the parallel hash join plan exchanges had to process 5 million and 15 million rows. The amount of processor time spent on exchanges will be much lower in the collocated join plan. In addition, the collocated join plan has a maximum of 8 threads processing single partitions at any one time. The 41 partitions will all be processed eventually, but a new partition is not started until a thread asks for it. Threads can reuse hash table memory for the new partition. The parallel hash join plan also had 8 hash tables, but with all 5,000,000 build rows loaded at the same time. The collocated plan needs memory for only 8 * 125,000 = 1,000,000 rows at any one time. Collocated Hash Join Performance The collated join plan has disappointing performance in this case. The query runs for around 25,300ms despite the same IO statistics as usual. This is much the worst result so far, so what went wrong? It turns out that cardinality estimation for the single partition scans of table T1 is slightly low. The properties of the Clustered Index Scan of T1 (graphic immediately above) show the estimation was for 121,951 rows. This is a small shortfall compared with the 125,000 rows actually encountered, but it was enough to cause the hash join to spill to physical tempdb: A level 1 spill doesn’t sound too bad, until you realize that the spill to tempdb probably occurs for each of the 41 partitions. As a side note, the cardinality estimation error is a little surprising because the system tables accurately show there are 125,000 rows in every partition of T1. Unfortunately, the optimizer uses regular column and index statistics to derive cardinality estimates here rather than system table information (e.g. sys.partitions). Collocated Merge Join We will never know how well the collocated parallel hash join plan might have worked without the cardinality estimation error (and the resulting 41 spills to tempdb) but we do know: Merge join does not require a memory grant; and Merge join was the optimizer’s preferred join option for a single partition join Putting this all together, what we would really like to see is the same collocated join strategy, but using merge join instead of hash join. Unfortunately, the current query optimizer cannot produce a collocated merge join; it only knows how to do collocated hash join. So where does this leave us? CROSS APPLY sys.partitions We can try to write our own collocated join query. We can use sys.partitions to find the partition numbers, and CROSS APPLY to get a count per partition, with a final step to sum the partial counts. The following query implements this idea: SELECT row_count = SUM(Subtotals.cnt) FROM ( -- Partition numbers SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1 ) AS P CROSS APPLY ( -- Count per collocated join SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals; The estimated plan is: The cardinality estimates aren’t all that good here, especially the estimate for the scan of the system table underlying the sys.partitions view. Nevertheless, the plan shape is heading toward where we would like to be. Each partition number from the system table results in a per-partition scan of T1 and T2, a one-to-many Merge Join, and a Stream Aggregate to compute the partial counts. The final Stream Aggregate just sums the partial counts. Execution time for this query is around 3,500ms, with the same IO statistics as always. This compares favourably with 5,000ms for the serial plan produced by the optimizer with the OPTION (MERGE JOIN) hint. This is another case of the sum of the parts being less than the whole – summing 41 partial counts from 41 single-partition merge joins is faster than a single merge join and count over all partitions. Even so, this single-threaded collocated merge join is not as quick as the original parallel hash join plan, which executed in 2,600ms. On the positive side, our collocated merge join uses only one logical processor and requires no memory grant. The parallel hash join plan used 16 threads and reserved 569 MB of memory:   Using a Temporary Table Our collocated merge join plan should benefit from parallelism. The reason parallelism is not being used is that the query references a system table. We can work around that by writing the partition numbers to a temporary table (or table variable): SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   CREATE TABLE #P ( partition_number integer PRIMARY KEY);   INSERT #P (partition_number) SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1;   SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals;   DROP TABLE #P;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; Using the temporary table adds a few logical reads, but the overall execution time is still around 3500ms, indistinguishable from the same query without the temporary table. The problem is that the query optimizer still doesn’t choose a parallel plan for this query, though the removal of the system table reference means that it could if it chose to: In fact the optimizer did enter the parallel plan phase of query optimization (running search 1 for a second time): Unfortunately, the parallel plan found seemed to be more expensive than the serial plan. This is a crazy result, caused by the optimizer’s cost model not reducing operator CPU costs on the inner side of a nested loops join. Don’t get me started on that, we’ll be here all night. In this plan, everything expensive happens on the inner side of a nested loops join. Without a CPU cost reduction to compensate for the added cost of exchange operators, candidate parallel plans always look more expensive to the optimizer than the equivalent serial plan. Parallel Collocated Merge Join We can produce the desired parallel plan using trace flag 8649 again: SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: One difference between this plan and the collocated hash join plan is that a Repartition Streams exchange operator is used instead of Distribute Streams. The effect is similar, though not quite identical. The Repartition uses round-robin partitioning, meaning the next partition id is pushed to the next thread in sequence. The Distribute Streams exchange seen earlier used Demand partitioning, meaning the next partition id is pulled across the exchange by the next thread that is ready for more work. There are subtle performance implications for each partitioning option, but going into that would again take us too far off the main point of this post. Performance The important thing is the performance of this parallel collocated merge join – just 1350ms on a typical run. The list below shows all the alternatives from this post (all timings include creation, population, and deletion of the temporary table where appropriate) from quickest to slowest: Collocated parallel merge join: 1350ms Parallel hash join: 2600ms Collocated serial merge join: 3500ms Serial merge join: 5000ms Parallel merge join: 8400ms Collated parallel hash join: 25,300ms (hash spill per partition) The parallel collocated merge join requires no memory grant (aside from a paltry 1.2MB used for exchange buffers). This plan uses 16 threads at DOP 8; but 8 of those are (rather pointlessly) allocated to the parallel scan of the temporary table. These are minor concerns, but it turns out there is a way to address them if it bothers you. Parallel Collocated Merge Join with Demand Partitioning This final tweak replaces the temporary table with a hard-coded list of partition ids (dynamic SQL could be used to generate this query from sys.partitions): SELECT row_count = SUM(Subtotals.cnt) FROM ( VALUES (1),(2),(3),(4),(5),(6),(7),(8),(9),(10), (11),(12),(13),(14),(15),(16),(17),(18),(19),(20), (21),(22),(23),(24),(25),(26),(27),(28),(29),(30), (31),(32),(33),(34),(35),(36),(37),(38),(39),(40),(41) ) AS P (partition_number) CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: The parallel collocated hash join plan is reproduced below for comparison: The manual rewrite has another advantage that has not been mentioned so far: the partial counts (per partition) can be computed earlier than the partial counts (per thread) in the optimizer’s collocated join plan. The earlier aggregation is performed by the extra Stream Aggregate under the nested loops join. The performance of the parallel collocated merge join is unchanged at around 1350ms. Final Words It is a shame that the current query optimizer does not consider a collocated merge join (Connect item closed as Won’t Fix). The example used in this post showed an improvement in execution time from 2600ms to 1350ms using a modestly-sized data set and limited parallelism. In addition, the memory requirement for the query was almost completely eliminated  – down from 569MB to 1.2MB. The problem with the parallel hash join selected by the optimizer is that it attempts to process the full data set all at once (albeit using eight threads). It requires a large memory grant to hold all 5 million rows from table T1 across the eight hash tables, and does not take advantage of the divide-and-conquer opportunity offered by the common partitioning. The great thing about the collocated join strategies is that each parallel thread works on a single partition from both tables, reading rows, performing the join, and computing a per-partition subtotal, before moving on to a new partition. From a thread’s point of view… If you have trouble visualizing what is happening from just looking at the parallel collocated merge join execution plan, let’s look at it again, but from the point of view of just one thread operating between the two Parallelism (exchange) operators. Our thread picks up a single partition id from the Distribute Streams exchange, and starts a merge join using ordered rows from partition 1 of table T1 and partition 1 of table T2. By definition, this is all happening on a single thread. As rows join, they are added to a (per-partition) count in the Stream Aggregate immediately above the Merge Join. Eventually, either T1 (partition 1) or T2 (partition 1) runs out of rows and the merge join stops. The per-partition count from the aggregate passes on through the Nested Loops join to another Stream Aggregate, which is maintaining a per-thread subtotal. Our same thread now picks up a new partition id from the exchange (say it gets id 9 this time). The count in the per-partition aggregate is reset to zero, and the processing of partition 9 of both tables proceeds just as it did for partition 1, and on the same thread. Each thread picks up a single partition id and processes all the data for that partition, completely independently from other threads working on other partitions. One thread might eventually process partitions (1, 9, 17, 25, 33, 41) while another is concurrently processing partitions (2, 10, 18, 26, 34) and so on for the other six threads at DOP 8. The point is that all 8 threads can execute independently and concurrently, continuing to process new partitions until the wider job (of which the thread has no knowledge!) is done. This divide-and-conquer technique can be much more efficient than simply splitting the entire workload across eight threads all at once. Related Reading Understanding and Using Parallelism in SQL Server Parallel Execution Plans Suck © 2013 Paul White – All Rights Reserved Twitter: @SQL_Kiwi

    Read the article

  • Wishful Thinking: Why can't HTML fix Script Attacks at the Source?

    - by Rick Strahl
    The Web can be an evil place, especially if you're a Web Developer blissfully unaware of Cross Site Script Attacks (XSS). Even if you are aware of XSS in all of its insidious forms, it's extremely complex to deal with all the issues if you're taking user input and you're actually allowing users to post raw HTML into an application. I'm dealing with this again today in a Web application where legacy data contains raw HTML that has to be displayed and users ask for the ability to use raw HTML as input for listings. The first line of defense of course is: Just say no to HTML input from users. If you don't allow HTML input directly and use HTML Encoding (HttyUtility.HtmlEncode() in .NET or using standard ASP.NET MVC output @Model.Content) you're fairly safe at least from the HTML input provided. Both WebForms and Razor support HtmlEncoded content, although Razor makes it the default. In Razor the default @ expression syntax:@Model.UserContent automatically produces HTML encoded content - you actually have to go out of your way to create raw HTML content (safe by default) using @Html.Raw() or the HtmlString class. In Web Forms (V4) you can use:<%: Model.UserContent %> or if you're using a version prior to 4.0:<%= HttpUtility.HtmlEncode(Model.UserContent) %> This works great as a hedge against embedded <script> tags and HTML markup as any HTML is turned into text that displays as HTML but doesn't render the HTML. But it turns any embedded HTML markup tags into plain text. If you need to display HTML in raw form with the markup tags rendering based on user input this approach is worthless. If you do accept HTML input and need to echo the rendered HTML input back, the task of cleaning up that HTML is a complex task. In the projects I work on, customers are frequently asking for the ability to post raw HTML quite frequently.  Almost every app that I've built where there's document content from users we start out with text only input - possibly using something like MarkDown - but inevitably users want to just post plain old HTML they created in some other rich editing application. See this a lot with realtors especially who often want to reuse their postings easily in multiple places. In my work this is a common problem I need to deal with and I've tried dozens of different methods from sanitizing, simple rejection of input to custom markup schemes none of which have ever felt comfortable to me. They work in a half assed, hacked together sort of way but I always live in fear of missing something vital which is *really easy to do*. My Wishlist Item: A <restricted> tag in HTML Let me dream here for a second on how to address this problem. It seems to me the easiest place where this can be fixed is: In the browser. Browsers are actually executing script code so they have a lot of control over the script code that resides in a page. What if there was a way to specify that you want to turn off script code for a block of HTML? The main issue when dealing with HTML raw input isn't that we as developers are unaware of the implications of user input, but the fact that we sometimes have to display raw HTML input the user provides. So the problem markup is usually isolated in only a very specific part of the document. So, what if we had a way to specify that in any given HTML block, no script code could execute by wrapping it into a tag that disables all script functionality in the browser? This would include <script> tags and any document script attributes like onclick, onfocus etc. and potentially also disallow things like iFrames that can potentially be scripted from the within the iFrame's target. I'd like to see something along these lines:<article> <restricted allowscripts="no" allowiframes="no"> <div>Some content</div> <script>alert('go ahead make my day, punk!");</script> <div onfocus="$.getJson('http://evilsite.com/')">more content</div> </restricted> </article> A tag like this would basically disallow all script code from firing from any HTML that's rendered within it. You'd use this only on code that you actually render from your data only and only if you are dealing with custom data. So something like this:<article> <restricted> @Html.Raw(Model.UserContent) </restricted> </article> For browsers this would actually be easy to intercept. They render the DOM and control loading and execution of scripts that are loaded through it. All the browser would have to do is suspend execution of <script> tags and not hookup any event handlers defined via markup in this block. Given all the crazy XSS attacks that exist and the prevalence of this problem this would go a long way towards preventing at least coded script attacks in the DOM. And it seems like a totally doable solution that wouldn't be very difficult to implement by vendors. There would also need to be some logic in the parser to not allow an </restricted> or <restricted> tag into the content as to short-circuit the rstricted section (per James Hart's comment). I'm sure there are other issues to consider as well that I didn't think of in my off-the-back-of-a-napkin concept here but the idea overall seems worth consideration I think. Without code running in a user supplied HTML block it'd be pretty hard to compromise a local HTML document and pass information like Cookies to a server. Or even send data to a server period. Short of an iFrame that can access the parent frame (which is another restriction that should be available on this <restricted> tag) that could potentially communicate back, there's not a lot a malicious site could do. The HTML could still 'phone home' via image links and href links potentially and basically say this site was accessed, but without the ability to run script code it would be pretty tough to pass along critical information to the server beyond that. Ahhhh… one can dream… Not holding my breath of course. The design by committee that is the W3C can't agree on anything in timeframes measured less than decades, but maybe this is one place where browser vendors can actually step up the pressure. This is something in their best interest to reduce the attack surface for vulnerabilities on their browser platforms significantly. Several people commented on Twitter today that there isn't enough discussion on issues like this that address serious needs in the web browser space. Realistically security has to be a number one concern with Web applications in general - there isn't a Web app out there that is not vulnerable. And yet nothing has been done to address these security issues even though there might be relatively easy solutions to make this happen. It'll take time, and it's probably not going to happen in our lifetime, but maybe this rambling thought sparks some ideas on how this sort of restriction can get into browsers in some way in the future.© Rick Strahl, West Wind Technologies, 2005-2012Posted in ASP.NET  HTML5  HTML  Security   Tweet !function(d,s,id){var js,fjs=d.getElementsByTagName(s)[0];if(!d.getElementById(id)){js=d.createElement(s);js.id=id;js.src="//platform.twitter.com/widgets.js";fjs.parentNode.insertBefore(js,fjs);}}(document,"script","twitter-wjs"); (function() { var po = document.createElement('script'); po.type = 'text/javascript'; po.async = true; po.src = 'https://apis.google.com/js/plusone.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(po, s); })();

    Read the article

  • IIS 7&rsquo;s Sneaky Secret to Get COM-InterOp to Run

    - by David Hoerster
    Originally posted on: http://geekswithblogs.net/DavidHoerster/archive/2013/06/17/iis-7rsquos-sneaky-secret-to-get-com-interop-to-run.aspxIf you’re like me, you don’t really do a lot with COM components these days.  For me, I’ve been ‘lucky’ to stay in the managed world for the past 6 or 7 years. Until last week. I’m running a project to upgrade a web interface to an older COM-based application.  The old web interface is all classic ASP and lots of tables, in-line styles and a bunch of other late 90’s and early 2000’s goodies.  So in addition to updating the UI to be more modern looking and responsive, I decided to give the server side an update, too.  So I built some COM-InterOp DLL’s (easily through VS2012’s Add Reference feature…nothing new here) and built a test console line app to make sure the COM DLL’s were actually built according to the COM spec.  There’s a document management system that I’m thinking of whose COM DLLs were not proper COM DLLs and crashed and burned every time .NET tried to call them through a COM-InterOp layer. Anyway, my test app worked like a champ and I felt confident that I could build a nice façade around the COM DLL’s and wrap some functionality internally and only expose to my users/clients what they really needed. So I did this, built some tests and also built a test web app to make sure everything worked great.  It did.  It ran fine in IIS Express via Visual Studio 2012, and the timings were very close to the pure Classic ASP calls, so there wasn’t much overhead involved going through the COM-InterOp layer. You know where this is going, don’t you? So I deployed my test app to a DEV server running IIS 7.5.  When I went to my first test page that called the COM-InterOp layer, I got this pretty message: Retrieving the COM class factory for component with CLSID {81C08CAE-1453-11D4-BEBC-00500457076D} failed due to the following error: 80040154 Class not registered (Exception from HRESULT: 0x80040154 (REGDB_E_CLASSNOTREG)). It worked as a console app and while running under IIS Express, so it must be permissions, right?  I gave every account I could think of all sorts of COM+ rights and nothing, nada, zilch! Then I came across this question on Experts Exchange, and at the bottom of the page, someone mentioned that the app pool should be running to allow 32-bit apps to run.  Oh yeah, my machine is 64-bit; these COM DLL’s I’m using are old and are definitely 32-bit.  I didn’t check for that and didn’t even think about that.  But I went ahead and looked at the app pool that my web site was running under and what did I see?  Yep, select your app pool in IIS 7.x, click on Advanced Settings and check for “Enable 32-bit Applications”. I went ahead and set it to True and my test application suddenly worked. Hope this helps somebody out there from pulling out your hair.

    Read the article

  • SQL Azure Reporting Limited CTP Arrived

    - by Shaun
    It’s about 3 months later when I registered the SQL Azure Reporting CTP on the Microsoft Connect after TechED 2010 China. Today when I checked my mailbox I found that the SQL Azure team had just accepted my request and sent the activation code over to me. So let’s have a look on the new SQL Azure Reporting.   Concept The SQL Azure Reporting provides cloud-based reporting as a service, built on SQL Server Reporting Services and SQL Azure technologies. Cloud-based reporting solutions such as SQL Azure Reporting provide many benefits, including rapid provisioning, cost-effective scalability, high availability, and reduced management overhead for report servers; and secure access, viewing, and management of reports. By using the SQL Azure Reporting service, we can do: Embed the Visual Studio Report Viewer ADO.NET Ajax control or Windows Form control to view the reports deployed on SQL Azure Reporting Service in our web or desktop application. Leverage the SQL Azure Reporting SOAP API to manage and retrieve the report content from any kinds of application. Use the SQL Azure Reporting Service Portal to navigate and view the reports deployed on the cloud. Since the SQL Azure Reporting was built based on the SQL Server 2008 R2 Reporting Service, we can use any tools we are familiar with, such as the SQL Server Integration Studio, Visual Studio Report Viewer. The SQL Azure Reporting Service runs as a remote SQL Server Reporting Service just on the cloud rather than on a server besides us.   Establish a New SQL Azure Reporting Let’s move to the windows azure deveploer portal and click the Reporting item from the left side navigation bar. If you don’t have the activation code you can click the Sign Up button to send a requirement to the Microsoft Connect. Since I already recieved the received code mail I clicked the Provision button. Then after agree the terms of the service I will select the subscription for where my SQL Azure Reporting CTP should be provisioned. In this case I selected my free Windows Azure Pass subscription. Then the final step, paste the activation code and enter the password of our SQL Azure Reporting Service. The user name of the SQL Azure Reporting will be generated by SQL Azure automatically. After a while the new SQL Azure Reporting Server will be shown on our developer portal. The Reporting Service URL and the user name will be shown as well. We can reset the password from the toolbar button.   Deploy Report to SQL Azure Reporting If you are familiar with SQL Server Reporting Service you will find this part will be very similar with what you know and what you did before. Firstly we open the SQL Server Business Intelligence Development Studio and create a new Report Server Project. Then we will create a shared data source where the report data will be retrieved from. This data source can be SQL Azure but we can use local SQL Server or other database if it opens the port up. In this case we use a SQL Azure database located in the same data center of our reporting service. In the Credentials tab page we entered the user name and password to this SQL Azure database. The SQL Azure Reporting CTP only available at the North US Data Center now so that the related SQL Server and hosted service might be better to select the same data center to avoid the external data transfer fee. Then we create a very simple report, just retrieve all records from a table named Members and have a table in the report to list them. In the data source selection step we choose the shared data source we created before, then enter the T-SQL to select all records from the Member table, then put all fields into the table columns. The report will be like this as following In order to deploy the report onto the SQL Azure Reporting Service we need to update the project property. Right click the project node from the solution explorer and select the property item. In the Target Server URL item we will specify the reporting server URL of our SQL Azure Reporting. We can go back to the developer portal and select the reporting node from the left side, then copy the Web Service URL and paste here. But notice that we need to append “/reportserver” after pasted. Then just click the Deploy menu item in the context menu of the project, the Visual Studio will compile the report and then upload to the reporting service accordingly. In this step we will be prompted to input the user name and password of our SQL Azure Reporting Service. We can get the user name from the developer portal, just next to the Web Service URL in the SQL Azure Reporting page. And the password is the one we specified when created the reporting service. After about one minute the report will be deployed succeed.   View the Report in Browser SQL Azure Reporting allows us to view the reports which deployed on the cloud from a standard browser. We copied the Web Service URL from the reporting service main page and appended “/reportserver” in HTTPS protocol then we will have the SQL Azure Reporting Service login page. After entered the user name and password of the SQL Azure Reporting Service we can see the directories and reports listed. Click the report will launch the Report Viewer to render the report.   View Report in a Web Role with the Report Viewer The ASP.NET and Windows Form Report Viewer works well with the SQL Azure Reporting Service as well. We can create a ASP.NET Web Role and added the Report Viewer control in the default page. What we need to change to the report viewer are Change the Processing Mode to Remote. Specify the Report Server URL under the Server Remote category to the URL of the SQL Azure Reporting Web Service URL with “/reportserver” appended. Specify the Report Path to the report which we want to display. The report name should NOT include the extension name. For example my report was in the SqlAzureReportingTest project and named MemberList.rdl then the report path should be /SqlAzureReportingTest/MemberList. And the next one is to specify the SQL Azure Reporting Credentials. We can use the following class to wrap the report server credential. 1: private class ReportServerCredentials : IReportServerCredentials 2: { 3: private string _userName; 4: private string _password; 5: private string _domain; 6:  7: public ReportServerCredentials(string userName, string password, string domain) 8: { 9: _userName = userName; 10: _password = password; 11: _domain = domain; 12: } 13:  14: public WindowsIdentity ImpersonationUser 15: { 16: get 17: { 18: return null; 19: } 20: } 21:  22: public ICredentials NetworkCredentials 23: { 24: get 25: { 26: return null; 27: } 28: } 29:  30: public bool GetFormsCredentials(out Cookie authCookie, out string user, out string password, out string authority) 31: { 32: authCookie = null; 33: user = _userName; 34: password = _password; 35: authority = _domain; 36: return true; 37: } 38: } And then in the Page_Load method, pass it to the report viewer. 1: protected void Page_Load(object sender, EventArgs e) 2: { 3: ReportViewer1.ServerReport.ReportServerCredentials = new ReportServerCredentials( 4: "<user name>", 5: "<password>", 6: "<sql azure reporting web service url>"); 7: } Finally deploy it to Windows Azure and enjoy the report.   Summary In this post I introduced the SQL Azure Reporting CTP which had just available. Likes other features in Windows Azure, the SQL Azure Reporting is very similar with the SQL Server Reporting. As you can see in this post we can use the existing and familiar tools to build and deploy the reports and display them on a website. But the SQL Azure Reporting is just in the CTP stage which means It is free. There’s no support for it. Only available at the North US Data Center. You can get more information about the SQL Azure Reporting CTP from the links following SQL Azure Reporting Limited CTP at MSDN SQL Azure Reporting Samples at TechNet Wiki You can download the solutions and the projects used in this post here.   Hope this helps, Shaun All documents and related graphics, codes are provided "AS IS" without warranty of any kind. Copyright © Shaun Ziyan Xu. This work is licensed under the Creative Commons License.

    Read the article

  • Print SSRS Report / PDF automatically from SQL Server agent or Windows Service

    - by Jeremy Ramos
    Originally posted on: http://geekswithblogs.net/JeremyRamos/archive/2013/10/22/print-ssrs-report--pdf-from-sql-server-agent-or.aspxI have turned the Web upside-down to find a solution to this considering the least components and least maintenance as possible to achieve automated printing of an SSRS report. This is for the reason that we do not have a full software development team to maintain an app and we have to minimize the support overhead for the support team.Here is my setup:SQL Server 2008 R2 in Windows Server 2008 R2PDF format reports generated by SSRS Reports subscriptions to a Windows File ShareNetwork printerColoured reports with logo and brandingI have found and tested the following solutions to no avail:ProsConsCalling Adobe Acrobat Reader exe: "C:\Program Files (x86)\Adobe\Reader 11.0\Reader\acroRd32.exe" /n /s /o /h /t "C:\temp\print.pdf" \\printserver\printername"Very simple optionAdobe Acrobat reader requires to launch the GUI to send a job to a printer. Hence, this option cannot be used when printing from a service.Calling Adobe Acrobat Reader exe as a process from a .NET console appA bit harder than above, but still a simple solutionSame as cons abovePowershell script(Start-Process -FilePath "C:\temp\print.pdf" -Verb Print)Very simple optionUses default PDF client in quiet mode to Print, but also requires an active session.    Foxit ReaderVery simple optionRequires GUI same as Adobe Acrobat Reader Using the Reporting Services Web service to run and stream the report to an image object and then passed to the printerQuite complexThis is what we're trying to avoid  After pulling my hair out for two days, testing and evaluating the above solutions, I ended up learning more about printers (more than ever in my entire life) and how printer drivers work with PostScripts. I then bumped on to a PostScript interpreter called GhostScript (http://www.ghostscript.com/) and then the solution starts to get clearer and clearer.I managed to achieve a solution (maybe not be the simplest but efficient enough to achieve the least-maintenance-least-components goal) in 3-simple steps:Install GhostScript (http://www.ghostscript.com/download/) - this is an open-source PostScript and PDF interpreter. Printing directly using GhostScript only produces grayscale prints using the laserjet generic driver unless you save as BMP image and then interpret the colours using the imageInstall GSView (http://pages.cs.wisc.edu/~ghost/gsview/)- this is a GhostScript add-on to make it easier to directly print to a Windows printer. GSPrint automates the above  PDF -> BMP -> Printer Driver.Run the GSPrint command from SQL Server agent or Windows Service:"C:\Program Files\Ghostgum\gsview\gsprint.exe" -color -landscape -all -printer "printername" "C:\temp\print.pdf"Command line options are here: http://pages.cs.wisc.edu/~ghost/gsview/gsprint.htmAnother lesson learned is, since you are calling the script from the Service Account, it will not necessarily have the Printer mapped in its Windows profile (if it even has one). The workaround to this is by adding a local printer as you normally would and then map this printer to the network printer. Note that you may need to install the Printer Driver locally in the server.So, that's it! There are many ways to achieve a solution. The key thing is how you provide the smartest solution!

    Read the article

  • ParallelWork: Feature rich multithreaded fluent task execution library for WPF

    - by oazabir
    ParallelWork is an open source free helper class that lets you run multiple work in parallel threads, get success, failure and progress update on the WPF UI thread, wait for work to complete, abort all work (in case of shutdown), queue work to run after certain time, chain parallel work one after another. It’s more convenient than using .NET’s BackgroundWorker because you don’t have to declare one component per work, nor do you need to declare event handlers to receive notification and carry additional data through private variables. You can safely pass objects produced from different thread to the success callback. Moreover, you can wait for work to complete before you do certain operation and you can abort all parallel work while they are in-flight. If you are building highly responsive WPF UI where you have to carry out multiple job in parallel yet want full control over those parallel jobs completion and cancellation, then the ParallelWork library is the right solution for you. I am using the ParallelWork library in my PlantUmlEditor project, which is a free open source UML editor built on WPF. You can see some realistic use of the ParallelWork library there. Moreover, the test project comes with 400 lines of Behavior Driven Development flavored tests, that confirms it really does what it says it does. The source code of the library is part of the “Utilities” project in PlantUmlEditor source code hosted at Google Code. The library comes in two flavors, one is the ParallelWork static class, which has a collection of static methods that you can call. Another is the Start class, which is a fluent wrapper over the ParallelWork class to make it more readable and aesthetically pleasing code. ParallelWork allows you to start work immediately on separate thread or you can queue a work to start after some duration. You can start an immediate work in a new thread using the following methods: void StartNow(Action doWork, Action onComplete) void StartNow(Action doWork, Action onComplete, Action<Exception> failed) For example, ParallelWork.StartNow(() => { workStartedAt = DateTime.Now; Thread.Sleep(howLongWorkTakes); }, () => { workEndedAt = DateTime.Now; }); Or you can use the fluent way Start.Work: Start.Work(() => { workStartedAt = DateTime.Now; Thread.Sleep(howLongWorkTakes); }) .OnComplete(() => { workCompletedAt = DateTime.Now; }) .Run(); Besides simple execution of work on a parallel thread, you can have the parallel thread produce some object and then pass it to the success callback by using these overloads: void StartNow<T>(Func<T> doWork, Action<T> onComplete) void StartNow<T>(Func<T> doWork, Action<T> onComplete, Action<Exception> fail) For example, ParallelWork.StartNow<Dictionary<string, string>>( () => { test = new Dictionary<string,string>(); test.Add("test", "test"); return test; }, (result) => { Assert.True(result.ContainsKey("test")); }); Or, the fluent way: Start<Dictionary<string, string>>.Work(() => { test = new Dictionary<string, string>(); test.Add("test", "test"); return test; }) .OnComplete((result) => { Assert.True(result.ContainsKey("test")); }) .Run(); You can also start a work to happen after some time using these methods: DispatcherTimer StartAfter(Action onComplete, TimeSpan duration) DispatcherTimer StartAfter(Action doWork,Action onComplete,TimeSpan duration) You can use this to perform some timed operation on the UI thread, as well as perform some operation in separate thread after some time. ParallelWork.StartAfter( () => { workStartedAt = DateTime.Now; Thread.Sleep(howLongWorkTakes); }, () => { workCompletedAt = DateTime.Now; }, waitDuration); Or, the fluent way: Start.Work(() => { workStartedAt = DateTime.Now; Thread.Sleep(howLongWorkTakes); }) .OnComplete(() => { workCompletedAt = DateTime.Now; }) .RunAfter(waitDuration);   There are several overloads of these functions to have a exception callback for handling exceptions or get progress update from background thread while work is in progress. For example, I use it in my PlantUmlEditor to perform background update of the application. // Check if there's a newer version of the app Start<bool>.Work(() => { return UpdateChecker.HasUpdate(Settings.Default.DownloadUrl); }) .OnComplete((hasUpdate) => { if (hasUpdate) { if (MessageBox.Show(Window.GetWindow(me), "There's a newer version available. Do you want to download and install?", "New version available", MessageBoxButton.YesNo, MessageBoxImage.Information) == MessageBoxResult.Yes) { ParallelWork.StartNow(() => { var tempPath = System.IO.Path.Combine( Environment.GetFolderPath(Environment.SpecialFolder.ApplicationData), Settings.Default.SetupExeName); UpdateChecker.DownloadLatestUpdate(Settings.Default.DownloadUrl, tempPath); }, () => { }, (x) => { MessageBox.Show(Window.GetWindow(me), "Download failed. When you run next time, it will try downloading again.", "Download failed", MessageBoxButton.OK, MessageBoxImage.Warning); }); } } }) .OnException((x) => { MessageBox.Show(Window.GetWindow(me), x.Message, "Download failed", MessageBoxButton.OK, MessageBoxImage.Exclamation); }); The above code shows you how to get exception callbacks on the UI thread so that you can take necessary actions on the UI. Moreover, it shows how you can chain two parallel works to happen one after another. Sometimes you want to do some parallel work when user does some activity on the UI. For example, you might want to save file in an editor while user is typing every 10 second. In such case, you need to make sure you don’t start another parallel work every 10 seconds while a work is already queued. You need to make sure you start a new work only when there’s no other background work going on. Here’s how you can do it: private void ContentEditor_TextChanged(object sender, EventArgs e) { if (!ParallelWork.IsAnyWorkRunning()) { ParallelWork.StartAfter(SaveAndRefreshDiagram, TimeSpan.FromSeconds(10)); } } If you want to shutdown your application and want to make sure no parallel work is going on, then you can call the StopAll() method. ParallelWork.StopAll(); If you want to wait for parallel works to complete without a timeout, then you can call the WaitForAllWork(TimeSpan timeout). It will block the current thread until the all parallel work completes or the timeout period elapses. result = ParallelWork.WaitForAllWork(TimeSpan.FromSeconds(1)); The result is true, if all parallel work completed. If it’s false, then the timeout period elapsed and all parallel work did not complete. For details how this library is built and how it works, please read the following codeproject article: ParallelWork: Feature rich multithreaded fluent task execution library for WPF http://www.codeproject.com/KB/WPF/parallelwork.aspx If you like the article, please vote for me.

    Read the article

  • Enhanced REST Support in Oracle Service Bus 11gR1

    - by jeff.x.davies
    In a previous entry on REST and Oracle Service Bus (see http://blogs.oracle.com/jeffdavies/2009/06/restful_services_with_oracle_s_1.html) I encoded the REST query string really as part of the relative URL. For example, consider the following URI: http://localhost:7001/SimpleREST/Products/id=1234 Now, technically there is nothing wrong with this approach. However, it is generally more common to encode the search parameters into the query string. Take a look at the following URI that shows this principle http://localhost:7001/SimpleREST/Products?id=1234 At first blush this appears to be a trivial change. However, this approach is more intuitive, especially if you are passing in multiple parameters. For example: http://localhost:7001/SimpleREST/Products?cat=electronics&subcat=television&mfg=sony The above URI is obviously used to retrieve a list of televisions made by Sony. In prior versions of OSB (before 11gR1PS3), parsing the query string of a URI was more difficult than in the current release. In 11gR1PS3 it is now much easier to parse the query strings, which in turn makes developing REST services in OSB even easier. In this blog entry, we will re-implement the REST-ful Products services using query strings for passing parameter information. Lets begin with the implementation of the Products REST service. This service is implemented in the Products.proxy file of the project. Lets begin with the overall structure of the service, as shown in the following screenshot. This is a common pattern for REST services in the Oracle Service Bus. You implement different flows for each of the HTTP verbs that you want your service to support. Lets take a look at how the GET verb is implemented. This is the path that is taken of you were to point your browser to: http://localhost:7001/SimpleREST/Products/id=1234 There is an Assign action in the request pipeline that shows how to extract a query parameter. Here is the expression that is used to extract the id parameter: $inbound/ctx:transport/ctx:request/http:query-parameters/http:parameter[@name="id"]/@value The Assign action that stores the value into an OSB variable named id. Using this type of XPath statement you can query for any variables by name, without regard to their order in the parameter list. The Log statement is there simply to provided some debugging info in the OSB server console. The response pipeline contains a Replace action that constructs the response document for our rest service. Most of the response data is static, but the ID field that is returned is set based upon the query-parameter that was passed into the REST proxy. Testing the REST service with a browser is very simple. Just point it to the URL I showed you earlier. However, the browser is really only good for testing simple GET services. The OSB Test Console provides a much more robust environment for testing REST services, no matter which HTTP verb is used. Lets see how to use the Test Console to test this GET service. Open the OSB we console (http://localhost:7001/sbconsole) and log in as the administrator. Click on the Test Console icon (the little "bug") next to the Products proxy service in the SimpleREST project. This will bring up the Test Console browser window. Unlike SOAP services, we don't need to do much work in the request document because all of our request information will be encoded into the URI of the service itself. Belore the Request Document section of the Test Console is the Transport section. Expand that section and modify the query-parameters and http-method fields as shown in the next screenshot. By default, the query-parameters field will have the tags already defined. You just need to add a tag for each parameter you want to pass into the service. For out purposes with this particular call, you'd set the quer-parameters field as follows: <tp:parameter name="id" value="1234" /> </tp:query-parameters> Now you are ready to push the Execute button to see the results of the call. That covers the process for parsing query parameters using OSB. However, what if you have an OSB proxy service that needs to consume a REST-ful service? How do you tell OSB to pass the query parameters to the external service? In the sample code you will see a 2nd proxy service called CallREST. It invokes the Products proxy service in exactly the same way it would invoke any REST service. Our CallREST proxy service is defined as a SOAP service. This help to demonstrate OSBs ability to mediate between service consumers and service providers, decreasing the level of coupling between them. If you examine the message flow for the CallREST proxy service, you'll see that it uses an Operational branch to isolate processing logic for each operation that is defined by the SOAP service. We will focus on the getProductDetail branch, that calls the Products REST service using the HTTP GET verb. Expand the getProduct pipeline and the stage node that it contains. There is a single Assign statement that simply extracts the productID from the SOA request and stores it in a local OSB variable. Nothing suprising here. The real work (and the real learning) occurs in the Route node below the pipeline. The first thing to learn is that you need to use a route node when calling REST services, not a Service Callout or a Publish action. That's because only the Routing action has access to the $oubound variable, especially when invoking a business service. The Routing action contains 3 Insert actions. The first Insert action shows how to specify the HTTP verb as a GET. The second insert action simply inserts the XML node into the request. This element does not exist in the request by default, so we need to add it manually. Now that we have the element defined in our outbound request, we can fill it with the parameters that we want to send to the REST service. In the following screenshot you can see how we define the id parameter based on the productID value we extracted earlier from the SOAP request document. That expression will look for the parameter that has the name id and extract its value. That's all there is to it. You now know how to take full advantage of the query parameter parsing capability of the Oracle Service Bus 11gR1PS2. Download the sample source code here: rest2_sbconfig.jar Ubuntu and the OSB Test Console You will get an error when you try to use the Test Console with the Oracle Service Bus, using Ubuntu (or likely a number of other Linux distros also). The error (shown below) will state that the Test Console service is not running. The fix for this problem is quite simple. Open up the WebLogic Server administrator console (usually running at http://localhost:7001/console). In the Domain Structure window on the left side of the console, select the Servers entry under the Environment heading. The select the Admin Server entry in the main window of the console. By default, you should be viewing the Configuration tabe and the General sub tab in the main window. Look for the Listen Address field. By default it is blank, which means it is listening on all interfaces. For some reason Ubuntu doesn't like this. So enter a value like localhost or the specific IP address or DNS name for your server (usually its just localhost in development envirionments). Save your changes and restart the server. Your Test Console will now work correctly.

    Read the article

< Previous Page | 279 280 281 282 283 284 285 286 287 288 289 290  | Next Page >