Search Results

Search found 577 results on 24 pages for 'recurring billing'.

Page 21/24 | < Previous Page | 17 18 19 20 21 22 23 24  | Next Page >

  • Monitoring Events in your BPEL Runtime - RSS Feeds?

    - by Ramkumar Menon
    @10g - It had been a while since I'd tried something different. so here's what I did this week!Whenever our Developers deployed processes to the BPEL runtime, or perhaps when a process gets turned off due to connectivity issues, or maybe someone retired a process, I needed to know. So here's what I did. Step 1: Downloaded Quartz libraries and went through the documentation to understand what it takes to schedule a recurring job. Step 2: Cranked out two components using Oracle JDeveloper. [Within a new Web Project] a) A simple Java Class named FeedUpdater that extends org.quartz.Job. All this class does is to connect to your BPEL Runtime [via opmn:ormi] and fetch all events that occured in the last "n" minutes. events? - If it doesn't ring a bell - its right there on the BPEL Console. If you click on "Administration > Process Log" - what you see are events.The API to retrieve the events is //get the locator reference for the domain you are interested in.Locator l = .... //Predicate to retrieve events for last "n" minutesWhereCondition wc = new WhereCondition(...) //get all those events you needed.BPELProcessEvent[] events = l.listProcessEvents(wc); After you get all these events, write out these into an RSS Feed XML structure and stream it into a file that resides either in your Apache htdocs, or wherever it can be accessed via HTTP.You can read all about RSS 2.0 here. At a high level, here is how it looks like. <?xml version = '1.0' encoding = 'UTF-8'?><rss version="2.0">  <channel>    <title>Live Updates from the Development Environment</title>    <link>http://soadev.myserver.com/feeds/</link>    <description>Live Updates from the Development Environment</description>    <lastBuildDate>Fri, 19 Nov 2010 01:03:00 PST</lastBuildDate>    <language>en-us</language>    <ttl>1</ttl>    <item>      <guid>1290213724692</guid>      <title>Process compiled</title>      <link>http://soadev.myserver.com/BPELConsole/mdm_product/administration.jsp?mode=processLog&amp;processName=&amp;dn=all&amp;eventType=all&amp;eventDate=600&amp;Filter=++Filter++</link>      <pubDate>Fri Nov 19 00:00:37 PST 2010</pubDate>      <description>SendPurchaseOrderRequestService: 3.0 Time : Fri Nov 19 00:00:37                   PST 2010</description>    </item>   ...... </channel> </rss> For writing ut XML content, read through Oracle XML Parser APIs - [search around for oracle.xml.parser.v2] b) Now that my "Job" was done, my job was half done. Next, I wrote up a simple Scheduler Servlet that schedules the above "Job" class to be executed ever "n" minutes. It is very straight forward. Here is the primary section of the code.           try {        Scheduler sched = StdSchedulerFactory.getDefaultScheduler();         //get n and make a trigger that executes every "n" seconds        Trigger trigger = TriggerUtils.makeSecondlyTrigger(n);        trigger.setName("feedTrigger" + System.currentTimeMillis());        trigger.setGroup("feedGroup");                JobDetail job = new JobDetail("SOA_Feed" + System.currentTimeMillis(), "feedGroup", FeedUpdater.class);        sched.scheduleJob(job,trigger);         }catch(Exception ex) {            ex.printStackTrace();            throw new ServletException(ex.getMessage());        } Look up the Quartz API and documentation. It will make this look much simpler.   Now that both components were ready, I packaged the Application into a war file and deployed it onto my Application Server. When the servlet initialized, the "n" second schedule was set/initialized. From then on, the servlet kept populating the RSS Feed file. I just ensured that my "Job" code keeps only 30 latest events within it, so that the feed file is small and under control. [a few kbs]   Next I opened up the feed xml on my browser - It requested a subscription - and Here I was - watching new deployments/life cycle events all popping up on my browser toolbar every 5 (actually n)  minutes!   Well, you could do it on a browser/reader of your choice - or perhaps read them like you read an email on your thunderbird!.      

    Read the article

  • Windows Azure Recipe: Software as a Service (SaaS)

    - by Clint Edmonson
    The cloud was tailor built for aspiring companies to create innovative internet based applications and solutions. Whether you’re a garage startup with very little capital or a Fortune 1000 company, the ability to quickly setup, deliver, and iterate on new products is key to capturing market and mind share. And if you can capture that share and go viral, having resiliency and infinite scale at your finger tips is great peace of mind. Drivers Cost avoidance Time to market Scalability Solution Here’s a sketch of how a basic Software as a Service solution might be built out: Ingredients Web Role – this hosts the core web application. Each web role will host an instance of the software and as the user base grows, additional roles can be spun up to meet demand. Access Control – this service is essential to managing user identity. It’s backed by a full blown implementation of Active Directory and allows the definition and management of users, groups, and roles. A pre-built ASP.NET membership provider is included in the training kit to leverage this capability but it’s also flexible enough to be combined with external Identity providers including Windows LiveID, Google, Yahoo!, and Facebook. The provider model provides extensibility to hook into other industry specific identity providers as well. Databases – nearly every modern SaaS application is backed by a relational database for its core operational data. If the solution is sold to organizations, there’s a good chance multi-tenancy will be needed. An emerging best practice for SaaS applications is to stand up separate SQL Azure database instances for each tenant’s proprietary data to ensure isolation from other tenants. Worker Role – this is the best place to handle autonomous background processing such as data aggregation, billing through external services, and other specialized tasks that can be performed asynchronously. Placing these tasks in a worker role frees the web roles to focus completely on user interaction and data input and provides finer grained control over the system’s scalability and throughput. Caching (optional) – as a web site traffic grows caching can be leveraged to keep frequently used read-only, user specific, and application resource data in a high-speed distributed in-memory for faster response times and ultimately higher scalability without spinning up more web and worker roles. It includes a token based security model that works alongside the Access Control service. Blobs (optional) – depending on the nature of the software, users may be creating or uploading large volumes of heterogeneous data such as documents or rich media. Blob storage provides a scalable, resilient way to store terabytes of user data. The storage facilities can also integrate with the Access Control service to ensure users’ data is delivered securely. Training & Examples These links point to online Windows Azure training labs and examples where you can learn more about the individual ingredients described above. (Note: The entire Windows Azure Training Kit can also be downloaded for offline use.) Windows Azure (16 labs) Windows Azure is an internet-scale cloud computing and services platform hosted in Microsoft data centers, which provides an operating system and a set of developer services which can be used individually or together. It gives developers the choice to build web applications; applications running on connected devices, PCs, or servers; or hybrid solutions offering the best of both worlds. New or enhanced applications can be built using existing skills with the Visual Studio development environment and the .NET Framework. With its standards-based and interoperable approach, the services platform supports multiple internet protocols, including HTTP, REST, SOAP, and plain XML SQL Azure (7 labs) Microsoft SQL Azure delivers on the Microsoft Data Platform vision of extending the SQL Server capabilities to the cloud as web-based services, enabling you to store structured, semi-structured, and unstructured data. Windows Azure Services (9 labs) As applications collaborate across organizational boundaries, ensuring secure transactions across disparate security domains is crucial but difficult to implement. Windows Azure Services provides hosted authentication and access control using powerful, secure, standards-based infrastructure. Developing Applications for the Cloud, 2nd Edition (eBook) This book demonstrates how you can create from scratch a multi-tenant, Software as a Service (SaaS) application to run in the cloud using the latest versions of the Windows Azure Platform and tools. The book is intended for any architect, developer, or information technology (IT) professional who designs, builds, or operates applications and services that run on or interact with the cloud. Fabrikam Shipping (SaaS reference application) This is a full end to end sample scenario which demonstrates how to use the Windows Azure platform for exposing an application as a service. We developed this demo just as you would: we had an existing on-premises sample, Fabrikam Shipping, and we wanted to see what it would take to transform it in a full subscription based solution. The demo you find here is the result of that investigation See my Windows Azure Resource Guide for more guidance on how to get started, including more links web portals, training kits, samples, and blogs related to Windows Azure.

    Read the article

  • How to get paid and figure out if I want to keep this client [migrated]

    - by Heiner Fawkes
    I have a client who is not paying on time, but it looks like the specifics don't match similar questions on this SE site. I got a call from a client I did website work for years ago. I had not done this kind of work for many years and frankly I'm not sure I want to now, but nevertheless about a month ago I agreed to bring his website, SEO, social media, and overall marketing for his small business up to speed. Why? He has told me many times how I'm the most honest, most well-informed contractor he's had experience with. And I personally kind of like him too. So I started working on an hourly basis. I sent one very small invoice and got paid. Then we talked a whole lot about all sorts of feature he would like me to implement. I started that work, and sent a second invoice on the first of the month (one of my two stated billing days). I didn't get paid. On every invoice it states that I charge a whopping ten percent per week late. I sent many voicemails and emails asking to please let me know what's going on with payment, and didn't get replies. Then the 15th of the month rolled around (which I stated initially as one of my invoicing dates). Since I hadn't been paid for the last invoice, I simply didn't send him an invoice at that time but emailed him and said that I will combine it with the next scheduled invoice for this reason (probably a bad idea I realize). Eventually he sent a portion of the invoice payment. I emailed back to let him know that he's three weeks late and what the remaining balance is. Finally we got in touch via phone. He basically told me that he thought I hadn't done all of the work I said I did. He looked at the page source code and it didn't look complete to him. I explained why his perception would be different and what work I had done as specified. He accepted this and said that part of the reason he didn't pay in full is that he's been swamped with personal family stuff, and part of the reason is that he didn't think I did all the work. That struck me as pretty weird. He also expressed concern that he has no idea now how much all the changes he has asked for are going to cost. And once again, he told me how honest and high-quality my services are compared to others he has dealt with. He also said he would pay me more (but not all) of the now three weeks overdue invoice that day. I didn't receive any payment. Basically this is how the client relationship strikes me: He's not good at communication. He's very busy and English isn't his first language. He almost never replies to emails but phone calls are fine. He's asked me to avoid emails for communication and I've asked him to please use email. He might not have enough money to afford all the things he has asked for. But so far I have been working for an hourly fee (which is quite high). He also has started paying monthly for hosting and social media services from me. What seems very abnormal is for a client to be so overdue on payments and to actually withhold payment of an invoice without any communication because he didn't think the work was done. I told him that I will send dollar estimates of each module of remaining work so that we can decide which ones are the highest priority if he cannot afford them all. I also reiterated that in the future if he has doubts about the work or an inability to pay, he must contact me immediately to say so. I basically plan to state the following to him: I would like to work for him and help his business. I also have sympathy for his recent family difficulties. I am happy to figure out payment plans that would work better for him, but first I need to be paid in full for all outstanding invoices, especially given that I skipped one of them just to be nice. The most crucial thing I need is communication about any problems with my work or his ability to pay. Once again, he heeds to pay in full immediately before we negotiate anything else. Does the above seem like an appropriate communication? Is anything missing from it? Is anything I'm doing here really abnormal?

    Read the article

  • Your Next IT Job

    - by BuckWoody
    Some data professionals have worked (and plan to work) in the same place for a long time. In organizations large and small, the turnover rate just isn’t that high. This has not been my experience. About every 3-5 years I’ve changed either roles or companies. That might be due to the IT environment or my personality (or a mix of the two), but the point is that I’ve had many roles and worked for many companies large and small throughout my 27+ years in IT. At one point this might have been a detriment – a prospective employer looks at the resume and says “it seems you’ve moved around quite a bit.” But I haven’t found that to be the case all the time –in fact, in some cases the variety of jobs I’ve held has been an asset because I’ve seen what works (and doesn’t) in other environments, which can save time and money. So if you’re in the first camp – great! Stay where you are, and continue doing the work you love. but if you’re in the second, then this post might be useful. If you are planning on making a change, or perhaps you’ve hit a wall at your current location, you might start looking around for a better paying job – and there’s nothing wrong with that. We all try to make our lives better, and for some that involves more money. Money, however, isn’t always the primary motivator. I’ve gone to another job that doesn’t have as many benefits or has the same salary as the current job I’m working to gain more experience, get a better work/life balance and so on. It’s a mix of factors that only you know about. So I thought I would lay out a few advantages and disadvantages in the shops I’ve worked at. This post isn’t aimed at a single employer, but represents a mix of what I’ve experienced, and of course the opinions here are my own. You will most certainly have a different take – if so, please post a response! I also won’t mention a specific industry – I’ve worked everywhere from medical firms, legal offices, retail, billing centers, manufacturing, government, even to NASA. I’m focusing here mostly on size and composition. And I’m making some very broad generalizations here – I am fully aware that a small company might have great benefits and a large company might allow a lot of role flexibility.  your mileage may vary – and again, post those comments! Small Company To me a “small company” means around 100 people or less – sometimes a lot less. These can be really fun, frustrating places to to work. Advantages: a great deal of flexibility, a wide range of roles (often at the same time), a large degree of responsibility, immediate feedback, close relationships with co-workers, work directly with your customer. Disadvantages: Too much responsibility, little work/life balance, immature political structure, few (if any) benefits. If the business is family-owned, they can easily violate work/life boundaries. Medium Size company In my experience the next size company I would work for involves from a few hundred people to around five thousand. Advantages: Good mobility – fairly easy to get promoted, acceptable benefits, more defined responsibilities, better work/life balance, balanced load for expertise, but still the organizational structure is fairly simple to understand. Disadvantages: Pay is not always highest, rapid changes in structure as the organization grows, transient workforce. You may not be given the opportunity to work with another technology if someone already “owns” it. Politics are painful at this level as people try to learn how to do it. Large Company When you get into the tens of thousands of folks employed around the world, you’re in a large company. Advantages: Lots of room to move around – sometimes you can work (as I have) multiple jobs through the years and yet stay at the same company, building time for benefits, very defined roles, trained managers (yes, I know some of them are still awful – trust me – I DO know that), higher-end benefits, long careers possible, discounts at retailers and other “soft” benefits, prestige. For some, a higher level of politics (done professionally) is a good thing. Disadvantages: You could become another faceless name in the crowd, might not allow a great deal of flexibility,  large organizational changes might take away any control you have of your career. I’ve also seen large layoffs happen, and good people get let go while “dead weight” is retained. For some, a higher level of politics is distasteful. So what are your experiences? Share with the group! Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Maximize Performance and Availability with Oracle Data Integration

    - by Tanu Sood
    Normal 0 false false false EN-US X-NONE X-NONE /* 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-top:0in; mso-para-margin-right:0in; mso-para-margin-bottom:10.0pt; mso-para-margin-left:0in; line-height:115%; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Calibri","sans-serif"; mso-fareast-font-family:Calibri; mso-bidi-font-family:"Times New Roman";} Alert: Oracle is hosting the 12c Launch Webcast for Oracle Data Integration and Oracle Golden Gate on Tuesday, November 12 (tomorrow) to discuss the new capabilities in detail and share customer perspectives. Hear directly from customer experts and executives from SolarWorld Industries America, British Telecom and Rittman Mead and get your questions answered live by product experts. Register for this complimentary webcast today and join in the discussion tomorrow. Author: Irem Radzik, Senior Principal Product Director, Oracle Organizations that want to use IT as a strategic point of differentiation prefer Oracle’s complete application offering to drive better business performance and optimize their IT investments. These enterprise applications are in the center of business operations and they contain critical data that needs to be accessed continuously, as well as analyzed and acted upon in a timely manner. These systems also need to operate with high-performance and availability, which means analytical functions should not degrade applications performance, and even system maintenance and upgrades should not interrupt availability. Oracle’s data integration products, Oracle Data Integrator, Oracle GoldenGate, and Oracle Enterprise Data Quality, provide the core foundation for bringing data from various business-critical systems to gain a broader, unified view. As a more advance offering to 3rd party products, Oracle’s data integration products facilitate real-time reporting for Oracle Applications without impacting application performance, and provide ability to upgrade and maintain the system without taking downtime. Oracle GoldenGate is certified for Oracle Applications, including E-Business Suite, Siebel CRM, PeopleSoft, and JD Edwards, for moving transactional data in real-time to a dedicated operational reporting environment. This solution allows the app users to offload the resource-heavy queries to the reporting instance(s), reducing CPU utilization, improving OLTP performance, and extending the lifetime of existing IT assets. In addition, having a dedicated reporting instance with up-to-the-second transactional data allows optimizing the reporting environment and even decreasing costs as GoldenGate can move only the required data from expensive mainframe environments to cost-efficient open system platforms.  With real-time data replication capabilities GoldenGate is also certified to enable application upgrades and database/hardware/OS migration without impacting business operations. GoldenGate is certified for Siebel CRM, Communications Billing and Revenue Management and JD Edwards for supporting zero downtime upgrades to the latest app version. GoldenGate synchronizes a parallel, upgraded system with the old version in real time, thus enables continuous operations during the process. Oracle GoldenGate is also certified for minimal downtime database migrations for Oracle E-Business Suite and other key applications. GoldenGate’s solution also minimizes the risk by offering a failback option after the switchover to the new environment. Furthermore, Oracle GoldenGate’s bidirectional active-active data replication is certified for Oracle ATG Web Commerce to enable geographically load balancing and high availability for ATG customers. For enabling better business insight, Oracle Data Integration products power Oracle BI Applications with high performance bulk and real-time data integration. Oracle Data Integrator (ODI) is embedded in Oracle BI Applications version 11.1.1.7.1 and helps to integrate data end-to-end across the full BI Applications architecture, supporting capabilities such as data-lineage, which helps business users identify report-to-source capabilities. ODI is integrated with Oracle GoldenGate and provides Oracle BI Applications customers the option to use real-time transactional data in analytics, and do so non-intrusively. By using Oracle GoldenGate with the latest release of Oracle BI Applications, organizations not only leverage fresh data in analytics, but also eliminate the need for an ETL batch window and minimize the impact on OLTP systems. You can learn more about Oracle Data Integration products latest 12c version in our upcoming launch webcast and access the app-specific free resources in the new Data Integration for Oracle Applications Resource Center.

    Read the article

  • Real Time BI in the Real World

    - by tobin.gilman(at)oracle.com
    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:10.0pt; font-family:"Times New Roman","serif";} 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:10.0pt; font-family:"Times New Roman","serif"; mso-fareast-font-family:"Times New Roman";} One of my favorite BI offerings from Oracle is a solution called Oracle Real Time Decisions.  Whenever I mention this product in customer meetings, eyes light up.  There are some fascinating examples of customers using it to up-sell, cross-sell, increase customer retention, and reduce risk in real time, with off the charts return on investment. I plan to share some of those stories in a future blog.  In this post however, I want to share some far more common real time analytics use case scenarios that are being addressed with widely deployed Oracle BI and data integration technologies Not all real time BI applications require continuous learning, predictive modeling, and data mining.  Many simply require the ability to integrate, aggregate, and access information that is current (typically within in few minutes or a few seconds).  The use cases are infinite.  A few I've seen: ·         Purchasing agents need to match demand against available inventory ·         Manufacturing planners need to monitor current parts and material against scheduled build plans ·         Airline agents need to match ticket demand against flight schedules, ·         Human resources managers need to track the status of global hiring requisitions against current headcount authorizations...you get the idea. One way of doing this is to run reports or federated queries directly against transactional systems.  That approach can be viable if you only need to access simple data sets on rare occasions.  High volume and complex queries can quickly bog down performance of mission critical transactional systems.  There is an architecturally simple way of solving the problem, and it's being applied by real companies around the world to solve real needs in real time.    Cbeyond is an Atlanta, GA based  provider of voice, data and mobile business applications delivers.  They deliver real time information to its call center agents  as they are interacting with their customers. The data they need resides in production CRM and other transactional systems, but  instead or reporting directly off the those systems, data is first moved to an operational data store (ODS).  Rather than running data intensive, time consuming, and performance degrading batch ETL routines to populate the ODS, Cbeyond uses Oracle Golden Gate software to incrementally capture and move only the changed records from log files of the transactional systems every few minutes.  There is no impact on transactional system performance, and the information needed by call center representatives is up to date.  Oracle Business Intelligence software presents the information to services reps in a rich, visual, and highly interactive format. Avea is similar to Cbeyond.  They are a telecommunications company who integrates billing and customer information in an ODS that is accessed by their call center agents in real time using Oracle Golden Gate and Oracle Business Intelligence.  They've taken it a step further by using the ODS to feed a data warehouse.  The operational data store provides the current information needed by call center agents during "in flight" customer interactions.  The data warehouse is used for more sophisticated analysis of historical data.  For maximum performance, both the ODS and data warehouse run on the Oracle Exadata Database Machine. These are practical illustrations of companies addressing real time reporting and analysis needs using established business intelligence/data warehousing methodologies and tools common to many IT departments.  If real time BI could benefit your organization, you may be already be closer than you thought to having the pieces in place to solving the problem.    Give us a shout if you are interested in learning more or if you have an interesting use or approach to real-time BI.

    Read the article

  • PayPal - CreateRecurringPaymentsProfile - is this request valid?

    - by PHP thinker
    I sending this request to make a recurring paypent (SOAP request to SandBox), but in response I get error message about missing token and other fields invalid (Missing Token or payment source). What could be wrong with this CreateRecurringPaymentsProfile request? <?xml version="1.0" encoding="UTF-8"?> <SOAP-ENV:Envelope xmlns:xsi="http://www.w3.org/1999/XMLSchema-instance" xmlns:SOAP-ENC="http://schemas.xmlsoap.org/soap/encoding/" xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/" xmlns:xsd="http://www.w3.org/1999/XMLSchema" SOAP-ENV:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/"> <SOAP-ENV:Header> <RequesterCredentials xmlns="urn:ebay:api:PayPalAPI" SOAP-ENV:mustUnderstand="1"> <Credentials xmlns="urn:ebay:apis:eBLBaseComponents"> <Username>xxxxx_biz_api1.gmail.com</Username> <Password>xxxxxxx</Password> <Subject/> </Credentials> </RequesterCredentials> </SOAP-ENV:Header> <SOAP-ENV:Body> <CreateRecurringPaymentsProfileReq xmlns="urn:ebay:api:PayPalAPI"> <CreateRecurringPaymentsProfileRequest> <Version xmlns="urn:ebay:apis:eBLBaseComponents" xsi:type="xsd:string">58.0</Version> <CreateRecurringPaymentsProfileRequestDetails> <RecurringPaymentsProfileDetails xmlns="urn:ebay:apis:eBLBaseComponents"> <BillingStartDate></BillingStartDate> </RecurringPaymentsProfileDetails> <ScheduleDetails> <Description>Must match</Description> <PaymentPeriod> <BillingPeriod>Day</BillingPeriod> <BillingFrequency>1</BillingFrequency> <Amount>39.95</Amount> </PaymentPeriod> </ScheduleDetails> <Token>EC-480620864W522011V</Token> </CreateRecurringPaymentsProfileRequestDetails> </CreateRecurringPaymentsProfileRequest> </CreateRecurringPaymentsProfileReq> </SOAP-ENV:Body> P.S. I am sending this request the correct way, after "Doexpresscheckout" command.

    Read the article

  • Good Email Notification Sending Service

    - by Philibert Perusse
    I need to send a few but important email notifications to individual users. For instance, when they register their software I send them a confirmation email. Right now, I am using 'sendmail' from my Perl CGI script to do the job. Most of my automated email are lost or marked as junk. Unfortunately, I am using shared hosting services and not a very good control over the SPF and SenderID DNS records. Even more bad, some other user of that shared server has been infected with some kind of SPAM-BOT and the IP is now blacklisted until further notice! Anyway I just don't want to deal with this kind of headache. I am looking for an online service that I will be able to subscribe to and pay something like 0.10$ per email I send with no monthly fees. I just need and API to be able to send the email from PHP or Perl code I will have to write. I have been looking around at all those "Email Sending Services" and they are all wrapped around creating campains and managing lists for bulk email marketing distribution and newsletters. But remember, I want to send an email notification to a "single" recipient. So far, I have look at MailChimp, SocketLabs, iContact, ConstantContact, StreamSend and so many others to no avail. I have seen one comment at Hackers News saying that MailChimp have an API for transactional e-mails (i.e. ad-hoc ones to welcome a user for example). So you're not just restricted to using them for bulk emails But I cannot find this in the API documentation supplied, maybe this was removed. Any suggestions out there. Here is a summary of my requirements: Allows ad hoc sending of email to a single recipient. Throughput may well be throttle I don't care, i am sending like 2-5 emails a day. API available in PHP or Perl to connect to that web service. Ideally I can send HTML formatted emails, otherwise I will live with text only. Solution not too expensive, between 0.01$ and 0.25$ per email would be acceptable. No recurring monthly fees.

    Read the article

  • Keeping track of business rules within IT department?

    - by evaldas-alexander
    I am looking for the best way to keep track of the business rules for both developers and everybody else (support staff / management) in a startup enviroment. The challenge is that our business model requires quite a lot of different business rules, which are created pretty much on the fly and evolving organically after that. After running this project for 3+ years, we have so many of such rules that often the only way to be sure about what the application is supposed to do in a certain situation is to go find the module responsible for that process and analyze its code and comments. That is all fine as long as you have one single developer who created the entire application from the scratch, but every new developer needs to go over pretty much entire codebase in order to understand how the application works. Even bigger problem is that non technical employees don't even have that option and therefore are forced to ask me pretty much every day how some certain case would be handled by the application. Quick example - we only start charging for our customer campaigns once they have been active for at least 72 hours, but at the same time we stop creating invoices for campaigns that belong to insolvent accounts and close such accounts within a month of the first failed charge. That does not apply to accounts that are set to "non-chargeable" which most commonly belongs to us since we are using the service ourselves. The invoices are created on the 1st of each month and include charges from the previous month + any current balance that the account might have. However, some customers are charged only 4 days after their invoice has been generated due to issues with their billing department. In addition to that, invoices are also created when customer deactivates his campaign, but that can only be done once the campaign is not longer under mandatory 6 month contract, unless account manager approves early deactivation. I know, that's quite a lot of rules that need to be taken into account when answering a question "when do we bill our customers", but actually I could still append an asterisk at the end of each sentence in order to disclose some rare exceptions. Of course, it would be easiest just to keep the business rules to the minimum, but we need to adapt to changing marketplace - i.e. less than a year ago we had no contracts whatsoever. One idea that I had so far was a simplistic wiki with categories corresponding to areas such as "Account activation", "Invoicing", "Collection procedures" and so on. Another idea would be to have giant interactive flowchart showing the entire customer "life cycle" from prospecting to account deactivation. What are your experiences / suggestions?

    Read the article

  • Optimizing tasks to reduce CPU in a trading application

    - by Joel
    Hello, I have designed a trading application that handles customers stocks investment portfolio. I am using two datastore kinds: Stocks - Contains unique stock name and its daily percent change. UserTransactions - Contains information regarding a specific purchase of a stock made by a user : the value of the purchase along with a reference to Stock for the current purchase. db.Model python modules: class Stocks (db.Model): stockname = db.StringProperty(multiline=True) dailyPercentChange=db.FloatProperty(default=1.0) class UserTransactions (db.Model): buyer = db.UserProperty() value=db.FloatProperty() stockref = db.ReferenceProperty(Stocks) Once an hour I need to update the database: update the daily percent change in Stocks and then update the value of all entities in UserTransactions that refer to that stock. The following python module iterates over all the stocks, update the dailyPercentChange property, and invoke a task to go over all UserTransactions entities which refer to the stock and update their value: Stocks.py # Iterate over all stocks in datastore for stock in Stocks.all(): # update daily percent change in datastore db.run_in_transaction(updateStockTxn, stock.key()) # create a task to update all user transactions entities referring to this stock taskqueue.add(url='/task', params={'stock_key': str(stock.key(), 'value' : self.request.get ('some_val_for_stock') }) def updateStockTxn(stock_key): #fetch the stock again - necessary to avoid concurrency updates stock = db.get(stock_key) stock.dailyPercentChange= data.get('some_val_for_stock') # I get this value from outside ... some more calculations here ... stock.put() Task.py (/task) # Amount of transaction per task amountPerCall=10 stock=db.get(self.request.get("stock_key")) # Get all user transactions which point to current stock user_transaction_query=stock.usertransactions_set cursor=self.request.get("cursor") if cursor: user_transaction_query.with_cursor(cursor) # Spawn another task if more than 10 transactions are in datastore transactions = user_transaction_query.fetch(amountPerCall) if len(transactions)==amountPerCall: taskqueue.add(url='/task', params={'stock_key': str(stock.key(), 'value' : self.request.get ('some_val_for_stock'), 'cursor': user_transaction_query.cursor() }) # Iterate over all transaction pointing to stock and update their value for transaction in transactions: db.run_in_transaction(updateUserTransactionTxn, transaction.key()) def updateUserTransactionTxn(transaction_key): #fetch the transaction again - necessary to avoid concurrency updates transaction = db.get(transaction_key) transaction.value= transaction.value* self.request.get ('some_val_for_stock') db.put(transaction) The problem: Currently the system works great, but the problem is that it is not scaling well… I have around 100 Stocks with 300 User Transactions, and I run the update every hour. In the dashboard, I see that the task.py takes around 65% of the CPU (Stock.py takes around 20%-30%) and I am using almost all of the 6.5 free CPU hours given to me by app engine. I have no problem to enable billing and pay for additional CPU, but the problem is the scaling of the system… Using 6.5 CPU hours for 100 stocks is very poor. I was wondering, given the requirements of the system as mentioned above, if there is a better and more efficient implementation (or just a small change that can help with the current implemntation) than the one presented here. Thanks!! Joel

    Read the article

  • sharepoint: editing webpart caml

    - by Jack
    I added this code to my sharepoint content query web part, which is looking at an events list from my calender, .webpart file in order to only show recurring events within the next month and regular events within the next month. However, I can't get the web part imported and working. Also is there any way to replace <Month /> with a range like <Today:Today OffsetDays="30"/> except with valid code? Here is the code: <property name="QueryOverride" type="string"> <Where> <Or> <And> <Neq> <FieldRef Name="FRecurrence"/> <Value Type="Recurrance">1</Value> </Neq> <And> <Lt> <FieldRef Name="EventDate" Type="DateTime"/> <Value Type="DateTime"><Today OffsetDays="30"/></Value> </Lt> <Gt> <FieldRef Name="EventDate" Type="DateTime"/> <Value Type="DateTime"><Today /></Value> </Gt> </And> </And> <DataRangesOverlap> <FieldRef Name="EventDate" /> <FieldRef Name="EndDate" /> <FieldRef Name="RecirrenceId" /> <Value Type="DateTime"><Month /></Value> </DataRangesOverlap> </Or> </Where> </property> When I upload this I get "Unable to add selected web part(s). The file format is not valid" and when I add <![CDATA[ and ]]> I can import it but the query doesn't return anything. How can I get this to work?

    Read the article

  • How can I iterate over a collection of objects returned by a LINQ-to-XML query?

    - by billmaya
    I've got this XML: <BillingLog> <BillingItem> <date-and-time>2003-11-04</date-and-time> <application-name>Billing Service</application-name> <severity>Warning</severity> <process-id>123</process-id> <description>Timed out on a connection</description> <detail>Timed out after three retries.</detail> </BillingItem> <BillingItem> <date-and-time>2010-05-15</date-and-time> <application-name>Callback Service</application-name> <severity>Error</severity> <process-id>456</process-id> <description>Unable to process callback</description> <detail>Reconciliation timed out after two retries.</detail> </BillingItem> </BillingLog> That I want to project using LINQ-to-XML into a collection of BillingItem objects contained in a single BillingLog object. public class BillingLog { public IEnumerable<BillingItem> items { get; set; } } public class BillingItem { public string Date { get; set; } public string ApplicationName { get; set; } public string Severity { get; set; } public int ProcessId { get; set; } public string Description { get; set; } public string Detail { get; set;} } This is the LINQ query that I'm using to project the XML (which is contained in the string variable source). XDocument xdoc = XDocument.Parse(source); var log = from i in xdoc.Elements("BillingLog") select new BillingLog { items = from j in i.Descendants("BillingItem") select new BillingItem { Date = (string)j.Element("date-and-time"), ApplicationName = (string)j.Element("application-name"), Severity = (string)j.Element("severity"), ProcessId = (int)j.Element("process-id"), Description = (string)j.Element("description"), Detail = (string)j.Element("detail") } }; When I try and iterate over the objects in log using foreach. foreach (BillingItem item in log) { Console.WriteLine ("{0} | {1} | {2} | {3} | {4} | {5}", item.Date, item.ApplicationName, item.Severity, item.ProcessId.ToString(), item.Description, item.Detail); } I get the following error message from LINQPad. Cannot convert type 'UserQuery.BillingLog' to 'UserQuery.BillingItem' Thanks in advance.

    Read the article

  • c++ class member functions selected by traits

    - by Jive Dadson
    I am reluctant to say I can't figure this out, but I can't figure this out. I've googled and searched stackoverflow, and come up empty. The abstract, and possibly overly vague form of the question is, how can I use the traits-pattern to instantiate non-virtual member functions? The question came up while modernizing a set of multivariate function optimizers that I wrote more than 10 years ago. The optimizers all operate by selecting a straight-line path through the parameter space away from the current best point (the "update"), then finding a better point on that line (the "line search"), then testing for the "done" condition, and if not done, iterating. There are different methods for doing the update, the line-search, and conceivably for the done test, and other things. Mix and match. Different update formulae require different state-variable data. For example, the LMQN update requires a vector, and the BFGS update requires a matrix. If evaluating gradients is cheap, the line-search should do so. If not, it should use function evaluations only. Some methods require more accurate line-searches than others. Those are just some examples. The original version instatiates several of the combinations by means of virtual functions. Some traits are selected by setting mode bits. Yuck. It would be trivial to define the traits with #define's and the member functions with #ifdef's and macros. But that's so twenty years ago. It bugs me that I cannot figure out a whiz-bang modern way. If there were only one trait that varied, I could use the curiously recurring template pattern. But I see no way to extend that to arbitrary combinations of traits. I tried doing it using boost::enable_if, etc.. The specialized state info was easy. I managed to get the functions done, but only by resorting to non-friend external functions that have the this-pointer as a parameter. I never even figured out how to make the functions friends, much less member functions. Perhaps tag-dispatch is the key. I haven't gotten very deeply into that. Surely it's possible, right? If so, what is best practice?

    Read the article

  • CSS Collapsing/Hiding divs with no data in <span>

    - by Chance
    I am trying to display an address which includes the following information: Title, division, address1, address2, town/state/zip, and country (5 seperate lines worth of data). The problem is sometimes the company may only have the title, address1, and town/state/zip yet other times it may be all but address2. This is determined upon a db record request server side. Therefore how can I make my output look proper when some of my labels will be blank? I would like div's that contain an empty span to be essentially collapsed/removed. My only idea of how was to use jquery and a selector to find all divs with blank spans (since thats all an asp.net label really is) and then remove those divs however this seems like such bad form. Is there any way to do this with css? Possible Code would be something like: $('span:empty:only-child').parent('div').remove(); Picture Examples (Ignore spacing/indentation issues which I will fix) Missing Division, Address2, and Country All Possible Fields The Html <asp:Label runat="server" ID="lblBillingAddressHeader" CssClass="lblBillingAddressHeader" Text="Billing Address:" /> <div style="position:relative; top:150px; left: 113px;"> <div class="test"> <asp:Label runat="server" ID="lblBillingDivision" CssClass="lblBillingShippingDivisionFont" /> </div> <div class="test"> <asp:Label runat="server" ID="lblBillingAddress" CssClass="lblBillingShippingFont" /> </div> <div class="test"> <asp:Label runat="server" ID="lblBillingAddress2" CssClass="lblBillingShippingFont" /> </div> <div class="test"> <asp:Label runat="server" ID="lblBillingAddress3" CssClass="lblBillingShippingFont" /> </div> <div class="test"> <asp:Label runat="server" ID="lblBillingAddress4" CssClass="lblBillingShippingFont" /> </div> </div> The CSS .test { position: relative; top: 0px; left: 0px; height: 12px; width: 300px; } .lblBillingShippingDivisionFont { font-size: small; font-weight: bold; } .lblBillingShippingFont { font-size: 10.6px; }

    Read the article

  • Thin & Sinatra not taking port

    - by NekoNova
    I'm having problems settig up my application using Thin and Sinatra. I have created a development-config.ru file that contains the following settings: # This is a rack configuration file to fire up the Sinatra application. # This allows better control and configuration as we are using the modular # approach here for controlling our application. # # Extend the Ruby load path with the root of the API and the lib folder # so that we can automatically include all our own custom classes. This makes # the requiring of files a bit cleaner and easier to maintain. # This is basically what rails does as well. # We also store the root of the API in the ENV settings to ensure we have # always access to the root of the API when building paths. ENV['API_ROOT'] = File.dirname(__FILE__) $:.unshift ENV['API_ROOT'] $:.unshift File.expand_path(File.join(ENV['API_ROOT'], 'lib')) $:.unshift File.expand_path(File.join(ENV['API_ROOT'], 'db')) # Now we can require all the gems used for the entire API by simpling requiring # them here. We can also include the classes that we have defined inside the lib # folder. require 'rubygems' require 'bundler' # Run Bundler to setup our gems properly. This will install all the missing gems on # the system and ensure that the deployment environment is ready to run. Bundler.require # To make the loading easier for the application, we will now automatically load all # models that have been defined inside the lib folder. This ensures that we do not need # to load them anymore anywhere else in our application, as the models will be known to # ruby everywhere. Dir.glob(File.join(ENV['API_ROOT'], 'lib', '**', '*.rb')).each{|file| require file} # Now we will configure the Sinatra application so that we can fire up the entire API. # This requires some detailed settings like whether logging is allowed, the port to be # used and some folder locations. require 'sinatra' require 'app' set :logging, true set :dump_errors, true set :port, 3001 set :views, "#{ENV['API_ROOT']}/views" set :public_folder, "#{ENV['API_ROOT']}/public" set :environment, :test # Start up the Sinatra application with all the settings that we have defined. run App.new This is based upon the information I found on the Sinatra website. However, the problem is that I cannot get the application running on port 3001. If I use thin start -R development-config.ru it runs on port 3000. If I use rackup config-development.ru it runs on port 9696. However I never see Sinatra kick in or run over port 3000. My application looks like this: # Author : Arne De Herdt # Email : # This is the actuall application that will be running under Sinatra # to serve the requests for the billing middleware API. # We use the modular approach here to allow control when deploying # the application using Capistrano. require 'sinatra/base' require 'logger' require 'savon' require 'billcrux' class App < Sinatra::Base # This action responds to POST requests on the URI '/billcrux/register' # and is responsible for handeling registration requests with the # BillCrux payment system. # The post "/billcrux/register" do # do stuff end end Can someone tell me what I am doing wrong?

    Read the article

  • how to get $form_state outside of FAPI's functions?

    - by logii
    I'm writing a custom module and I'd like to use $form_state of the current form in another non-form api function - custom_facet_view_build(). any help is appreciated :) <?php /** * Implementation of hook_perm(). */ function custom_facet_perm() { return array( 'access foo content', 'access baz content', ); } /** * Implementation of hook_menu(). */ function custom_facet_menu() { $items['faceted-search'] = array( 'title' => 'Faceted Search', 'page callback' => 'drupal_get_form', 'access arguments' => array(), ); $items['facet-search-test'] = array( 'page callback' => 'drupal_get_form', 'page arguments' => array('custom_facet_form'), 'access callback' => TRUE, 'type' => MENU_CALLBACK, ); return $items; } /** * Form definition; ahah_helper_demo form. */ function custom_facet_form($form_state) { $form = array(); ahah_helper_register($form, $form_state); if (isset($form_state['storage']['categories'])) { $categories_default_value = $form_state['storage']['categories']["#value"]; } $form['facet_search_form'] = array( '#type' => 'fieldset', '#title' => t('Faceted Search'), '#prefix' => '<div id="billing-info-wrapper">', // This is our wrapper div. '#suffix' => '</div>', '#tree' => TRUE, // Don't forget to set #tree! ); $form['facet_search_form']['categories'] = array( '#type' => 'select', '#title' => t('Category'), '#options' => _custom_facet_taxonomy_query(1), '#multiple' => TRUE, '#default_value' => $categories_default_value, ); $form['save'] = array( '#type' => 'submit', '#value' => t('Save'), ); return $form; } /** * Validate callback for the form. */ function custom_facet_form_validate($form, &$form_state) { } /** * Submit callback for the form. */ function custom_facet_form_submit($form, &$form_state) { drupal_set_message('nothing done'); $form_state['storage']['categories'] = $form['facet_search_form']['categories']; // dpm($form_state); // There's a value returned in form_state['storage] within this function } /** * Implementation of hook_views_api(). */ function custom_facet_views_api() { return array( 'api' => 2, ); } function custom_facet_view_build(&$view) { dpm($form_state); // form_state['storage] remains NULL even though there's a value on previous submission }

    Read the article

  • c++ class member functions instatiated by traits

    - by Jive Dadson
    I am reluctant to say I can't figure this out, but I can't figure this out. I've googled and searched stackoverflow, and come up empty. The abstract, and possibly overly vague form of the question is, how can I use the traits-pattern to instantiate non-virtual member functions? The question came up while modernizing a set of multivariate function optimizers that I wrote more than 10 years ago. The optimizers all operate by selecting a straight-line path through the parameter space away from the current best point (the "update"), then finding a better point on that line (the "line search"), then testing for the "done" condition, and if not done, iterating. There are different methods for doing the update, the line-search, and conceivably for the done test, and other things. Mix and match. Different update formulae require different state-variable data. For example, the LMQN update requires a vector, and the BFGS update requires a matrix. If evaluating gradients is cheap, the line-search should do so. If not, it should use function evaluations only. Some methods require more accurate line-searches than others. Those are just some examples. The original version instantiates several of the combinations by means of virtual functions. Some traits are selected by setting mode bits that are tested at runtime. Yuck. It would be trivial to define the traits with #define's and the member functions with #ifdef's and macros. But that's so twenty years ago. It bugs me that I cannot figure out a whiz-bang modern way. If there were only one trait that varied, I could use the curiously recurring template pattern. But I see no way to extend that to arbitrary combinations of traits. I tried doing it using boost::enable_if, etc.. The specialized state info was easy. I managed to get the functions done, but only by resorting to non-friend external functions that have the this-pointer as a parameter. I never even figured out how to make the functions friends, much less member functions. The compiler (vc++ 2008) always complained that things didn't match. I would yell, "SFINAE, you moron!" but the moron is probably me. Perhaps tag-dispatch is the key. I haven't gotten very deeply into that. Surely it's possible, right? If so, what is best practice?

    Read the article

  • I'm searching for a messaging platform (like XMPP) that allows tight integration with a web applicat

    - by loxs
    Hi, At the company I work for, we are building a cluster of web applications for collaboration. Things like accounting, billing, CRM etc. We are using a RESTfull technique: For database we use CouchDB Different applications communicate with one another and with the database via http. Besides, we have a single sign on solution, so that when you login in one application, you are automatically logged to the other. For all apps we use Python (Pylons). Now we need to add instant messaging to the stack. We need to support both web and desktop clients. But just being able to chat is not enough. We need to be able to achieve all of the following (and more similar things). When somebody gets assigned to a task, they must receive a message. I guess this is possible with some system daemon. There must be an option to automatically group people in groups by lots of different properties. For example, there must be groups divided both by geographical location, by company division, by job type (all the programers from different cities and different company divisions must form a group), so that one can send mass messages to a group of choice. Rooms should be automatically created and destroyed. For example when several people visit the same invoice, a room for them must be automatically created (and they must auto-join). And when all leave the invoice, the room must be destroyed. Authentication and authorization from our applications. I can implement this using custom solutions like hookbox http://hookbox.org/docs/intro.html but then I'll have lots of problems in supporting desktop clients. I have no former experience with instant messaging. I've been reading about this lately. I've been looking mostly at things like ejabberd. But it has been a hard time and I can't find whether what I want is possible at all. So I'd be happy if people with experience in this field could help me with some advice, articles, tales of what is possible etc.

    Read the article

  • How to setup Lucene/Solr for a B2B web app?

    - by Bill Paetzke
    Given: 1 database per client (business customer) 5000 clients Clients have between 2 to 2000 users (avg is ~100 users/client) 100k to 10 million records per database Users need to search those records often (it's the best way to navigate their data) Possibly relevant info: Several new clients each week (any time during business hours) Multiple web servers and database servers (users can login via any web server) Let's stay agnostic of language or sql brand, since Lucene (and Solr) have a breadth of support For Example: Joel Spolsky said in Podcast #11 that his hosted web app product, FogBugz On-Demand, uses Lucene. He has thousands of on-demand clients. And each client gets their own database. They use an index per client and store it in the client's database. I'm not sure on the details. And I'm not sure if this is a serious mod to Lucene. The Question: How would you setup Lucene search so that each client can only search within its database? How would you setup the index(es)? Where do you store the index(es)? Would you need to add a filter to all search queries? If a client cancelled, how would you delete their (part of the) index? (this may be trivial--not sure yet) Possible Solutions: Make an index for each client (database) Pro: Search is faster (than one-index-for-all method). Indices are relative to the size of the client's data. Con: I'm not sure what this entails, nor do I know if this is beyond Lucene's scope. Have a single, gigantic index with a database_name field. Always include database_name as a filter. Pro: Not sure. Maybe good for tech support or billing dept to search all databases for info. Con: Search is slower (than index-per-client method). Flawed security if query filter removed. One last thing: I would also accept an answer that uses Solr (the extension of Lucene). Perhaps it's better suited for this problem. Not sure.

    Read the article

  • C++ class member functions instantiated by traits

    - by Jive Dadson
    I am reluctant to say I can't figure this out, but I can't figure this out. I've googled and searched Stack Overflow, and come up empty. The abstract, and possibly overly vague form of the question is, how can I use the traits-pattern to instantiate non-virtual member functions? The question came up while modernizing a set of multivariate function optimizers that I wrote more than 10 years ago. The optimizers all operate by selecting a straight-line path through the parameter space away from the current best point (the "update"), then finding a better point on that line (the "line search"), then testing for the "done" condition, and if not done, iterating. There are different methods for doing the update, the line-search, and conceivably for the done test, and other things. Mix and match. Different update formulae require different state-variable data. For example, the LMQN update requires a vector, and the BFGS update requires a matrix. If evaluating gradients is cheap, the line-search should do so. If not, it should use function evaluations only. Some methods require more accurate line-searches than others. Those are just some examples. The original version instantiates several of the combinations by means of virtual functions. Some traits are selected by setting mode bits that are tested at runtime. Yuck. It would be trivial to define the traits with #define's and the member functions with #ifdef's and macros. But that's so twenty years ago. It bugs me that I cannot figure out a whiz-bang modern way. If there were only one trait that varied, I could use the curiously recurring template pattern. But I see no way to extend that to arbitrary combinations of traits. I tried doing it using boost::enable_if, etc.. The specialized state information was easy. I managed to get the functions done, but only by resorting to non-friend external functions that have the this-pointer as a parameter. I never even figured out how to make the functions friends, much less member functions. The compiler (VC++ 2008) always complained that things didn't match. I would yell, "SFINAE, you moron!" but the moron is probably me. Perhaps tag-dispatch is the key. I haven't gotten very deeply into that. Surely it's possible, right? If so, what is best practice?

    Read the article

  • SQL Server Issue: Could not allocate space for object ... primary filegroup is full

    - by Luke
    Trying to figure out a problem at an office that has SQL Server 2005 installed on Windows SBS Server 2008. Here's the setup: It's an office, and the person who set this all up is nowhere to be found. I'm the best hope they have... One of the programs they use on a workstation gives them an error of "Could not allocate space for object 'Billing' in database "MyDatabase" because primary filegroup is full" when trying to save an entry in their software. I searched around for hours, looking for possible solutions. One was to check for available disk space, and another was to defrag. I checked the hard drives on the server, and there is plenty of space free. I also defragged, which may have helped the problem somewhat. It's hard to say, because it seems like with the nature of the error, if you try over and over you might get it to actually save. My next step was to try to see if autogrowth was enabled on the database. This would seem to be a likely / possible solution, but I can't access the database! If I run the SQL Management Studio, I can log in as my Windows user and view the list of databases. However, if I try to do anything (actually view the database, view the properties, add or edit users), I get errors that I don't have permission. For what it's worth, I also tried runing Management Studio as Administrator, in case that would help. No difference, though. Now, what I'm guessing is going on -- from my limited knowledge of SQL and from reading online -- is that though I'm logged in as a Windows administrator, that account does NOT have SQL access. I do see a list of SQL users, including SA, but I again don't have permission to add one or to change the password on an existing one. And nobody at the office has any idea what the SQL passwords could be. So... here's my thinking thus far: 1 - The "Could not allocate" error likely points to a database that needs to be allowed to autogrow. Especially since I verified there is plenty of free space and the HD has been defragmented. 2 - Enabling autogrow would be very easy to do if I had the proper access within SQL Management Stuido. That leads me to this link: http://blogs.technet.com/b/sqlman/archive/2011/06/14/tips-amp-tricks-you-have-lost-access-to-sql-server-now-what.aspx It sounds like it's a step-by-step guide for giving me the access I need to SQL. I'm guessing that if I followed this guide, I would be able to then log in to the SQL server via Management Studio with the proper permissions, and would be able to enable autogrow (or simply view the status of the existing database), and hopefully solve the "Could not allocate space" problem! So I guess I have a few questions: 1 - Would you guys agree with my "diagnosis"? Think I'm barking up the right tree? 2 - Is there any risk at all in hurting / disabling / wrecking the current SQL database or setup with me going through the guide to regain SQL access? I understand that per the guide, I would have to temporarily shut down SQL, so obviously it wouldn't be accessible during that time. But it wouldn't be worth the risk if there's a chance I could mess anything up... Like I said, the workstations ARE currently accessing the database somehow, but nobody knows with what login info or anything. Basically, it's set up, it works (usually), but if they had to reload the software, nobody would know how. Any feedback would be appreciated!! The problem is such that it's not an emergency for them, but an annoyance. If I could fix it, it would be wonderful. But if not, I think they'll manage, especially as they are going to eventually stop using this software. Thank you so much for your time! Luke

    Read the article

  • How can I forward ALL traffic over a site-to-site VPN on Cisco ASA?

    - by Scott Clements
    Hi There, I currently have two Cisco ASA 5100 routers. They are at different physical sites and are configured with a site-to-site VPN which is active and working. I can communicate with the subnets on either site from the other and both are connected to the internet, however I need to ensure that all the traffic at my remote site goes through this VPN to my site here. I know that the web traffic is doing so as a "tracert" confirms this, but I need to ensure that all other network traffic is being directed over this VPN to my network here. Here is my config for the ASA router at my remote site: hostname ciscoasa domain-name xxxxx enable password 78rl4MkMED8xiJ3g encrypted names ! interface Ethernet0/0 nameif NIACEDC security-level 100 ip address x.x.x.x 255.255.255.0 ! interface Ethernet0/1 description External Janet Connection nameif JANET security-level 0 ip address x.x.x.x 255.255.255.248 ! interface Ethernet0/2 shutdown no nameif security-level 100 no ip address ! interface Ethernet0/3 shutdown no nameif security-level 100 ip address dhcp setroute ! interface Management0/0 nameif management security-level 100 ip address 192.168.100.1 255.255.255.0 management-only ! passwd 2KFQnbNIdI.2KYOU encrypted ftp mode passive clock timezone GMT/BST 0 clock summer-time GMT/BDT recurring last Sun Mar 1:00 last Sun Oct 2:00 dns domain-lookup NIACEDC dns server-group DefaultDNS name-server 154.32.105.18 name-server 154.32.107.18 domain-name XXXX same-security-traffic permit inter-interface same-security-traffic permit intra-interface access-list ren_access_in extended permit ip any any access-list ren_access_in extended permit tcp any any access-list ren_nat0_outbound extended permit ip 192.168.6.0 255.255.255.0 192.168.3.0 255.255.255.0 access-list NIACEDC_nat0_outbound extended permit ip 192.168.12.0 255.255.255.0 192.168.3.0 255.255.255.0 access-list JANET_20_cryptomap extended permit ip 192.168.12.0 255.255.255.0 192.168.3.0 255.255.255.0 access-list NIACEDC_access_in extended permit ip any any access-list NIACEDC_access_in extended permit tcp any any access-list JANET_access_out extended permit ip any any access-list NIACEDC_access_out extended permit ip any any pager lines 24 logging enable logging asdm informational mtu NIACEDC 1500 mtu JANET 1500 mtu management 1500 icmp unreachable rate-limit 1 burst-size 1 asdm image disk0:/asdm-522.bin no asdm history enable arp timeout 14400 nat-control global (NIACEDC) 1 interface global (JANET) 1 interface nat (NIACEDC) 0 access-list NIACEDC_nat0_outbound nat (NIACEDC) 1 192.168.12.0 255.255.255.0 access-group NIACEDC_access_in in interface NIACEDC access-group NIACEDC_access_out out interface NIACEDC access-group JANET_access_out out interface JANET route JANET 0.0.0.0 0.0.0.0 194.82.121.82 1 route JANET 0.0.0.0 0.0.0.0 192.168.3.248 tunneled timeout xlate 3:00:00 timeout conn 1:00:00 half-closed 0:10:00 udp 0:02:00 icmp 0:00:02 timeout sunrpc 0:10:00 h323 0:05:00 h225 1:00:00 mgcp 0:05:00 mgcp-pat 0:05:00 timeout sip 0:30:00 sip_media 0:02:00 sip-invite 0:03:00 sip-disconnect 0:02:00 timeout uauth 0:05:00 absolute http server enable http 192.168.12.0 255.255.255.0 NIACEDC http 192.168.100.0 255.255.255.0 management http 192.168.9.0 255.255.255.0 NIACEDC no snmp-server location no snmp-server contact snmp-server enable traps snmp authentication linkup linkdown coldstart crypto ipsec transform-set ESP-3DES-SHA esp-3des esp-sha-hmac crypto ipsec transform-set ESP-AES-256-SHA esp-aes-256 esp-sha-hmac crypto map JANET_map 20 match address JANET_20_cryptomap crypto map JANET_map 20 set pfs crypto map JANET_map 20 set peer X.X.X.X crypto map JANET_map 20 set transform-set ESP-AES-256-SHA crypto map JANET_map interface JANET crypto isakmp enable JANET crypto isakmp policy 10 authentication pre-share encryption aes-256 hash sha group 2 lifetime 86400 crypto isakmp policy 30 authentication pre-share encryption 3des hash sha group 2 lifetime 86400 crypto isakmp policy 50 authentication pre-share encryption aes-256 hash sha group 5 lifetime 86400 tunnel-group X.X.X.X type ipsec-l2l tunnel-group X.X.X.X ipsec-attributes pre-shared-key * telnet timeout 5 ssh timeout 5 console timeout 0 dhcpd address 192.168.100.2-192.168.100.254 management dhcpd enable management ! ! class-map inspection_default match default-inspection-traffic ! ! policy-map type inspect dns preset_dns_map parameters message-length maximum 512 policy-map global_policy class inspection_default inspect dns preset_dns_map inspect ftp inspect h323 h225 inspect h323 ras inspect rsh inspect rtsp inspect esmtp inspect sqlnet inspect skinny inspect sunrpc inspect xdmcp inspect sip inspect netbios inspect tftp inspect http ! service-policy global_policy global prompt hostname context no asdm history enable Thanks in advance, Scott

    Read the article

  • What would make a noise in a PC on graphics operations on a passively-cooled system?

    - by T.J. Crowder
    I have this system based on the Intel D510MO motherboard, which is basically an Atom D510 (dual-core HT Atom w/built-in GPU), an Intel NM10 chipset, and a Realtek Gigabit LAN controller. It's entirely passively cooled. I noticed almost immediately that there was a kind of very, very soft noise that corresponded with graphics operations, sort of the noise you'd get if you had a sheet of flat paper and slid something really light across it — but more electronic than that. I wrote it off as observation error and/or disk activity triggered by the graphics operation (although the latter seemed like a lot of unnecessary disk activity). It isn't. I got curious enough that I finally did a few controlled experiments, and here's what I've determined: It isn't the HDD. For one thing, the sounds the HDD makes (when seeking, when reading or writing, when just sitting there spinning) is different. For another, I used sudo hdparm -y /dev/sda (I'm using Ubuntu 10.04 LTS) to temporarily put the disk on standby while making sure that non-disk graphics op was happening in a loop. The disk spun down, but the other sound continued, corresponding perfectly with the timing of the graphics op. (Then the disk spun up again, but it takes long enough that I could rule out the HDD.) It isn't the monitor; I ensured the two were well physically-separated and the sound was definitely coming from the main box. It isn't something else in the room; the sound is coming from the box. It isn't cross-talk to an audio circuit coming out the speakers. (It doesn't have any speakers.) It isn't my mouse (e.g., when I'm trying to make graphics ops happen); the sound happens if I set up a recurring operation and don't use the mouse at all, or if I lift the mouse off the table slightly (but enough that the laser still registers movement). It isn't the voices in my head; they never whisper like that. Other observations: It doesn't seem to matter what the graphics operation is; anything that changes what's on the screen seems to do it. I get the sound when moving the mouse over the Chromium tab bar (which makes the tab backgrounds change); I get it when a web page has a counter on it that changes the text on the page: I get it when dragging window contents around. The sound is very, very slightly louder if the graphics op is larger, like scrolling a text area when writing a question on superuser.com, than for smaller operations like the tick counter on the web page. But it's very slight. It's fairly loud (and of good duration) when the op involves color changes to substantial surface areas. For instance, when asking a question here on superuser and you move the cursor between the question box and the tag box, and the help to the right fades out, changes, and fades back in. (Yet another example related to the web browser, so let me say: I hear it when operations completely unrelated to the web browser as well.) It doesn't sound like arcing or anything like that (I'd've shut off the machine Right Quick Like if it did). Moving windows does it. Scrolling windows (by and large) doesn't. I have the feeling I've heard this sort of thing before, when all system fans were on low and such, with other systems — but (again) written it off as observational error. For all the world it's like I'm hearing the CPU working (as opposed to the GPU; note the window scroll thing above) or data being transferred somewhere, but that just seems...unlikely. So what am I hearing? This may seem like a very localized question, but perhaps other silent PC enthusiasts may be interested as well...

    Read the article

  • Simple Cisco ASA 5505 config issue

    - by Ben Sebborn
    I have a Cisco ASA setup with two interfaces: inside: 192.168.2.254 / 255.255.255.0 SecLevel:100 outside: 192.168.3.250 / 255.255.255.0 SecLevel: 0 I have a static route setup to allow PCs on the inside network to access the internet via a gateway on the outside interface (3.254): outside 0.0.0.0 0.0.0.0 192.168.3.254 This all works fine. I now need to be able to access a PC on the outside interface (3.253) from a PC on the inside interface on port 35300. I understand I should be able to do this with no problems, as I'm going from a higher security level to a lower one. However I can't get any connection. Do I need to set up a seperate static route? Perhaps the route above is overriding what I need to be able to do (is it routing ALL traffic through the gateway?) Any advice on how to do this would be apprecaited. I am configuring this via ASDM but the config can be seen as below: Result of the command: "show running-config" : Saved : ASA Version 8.2(5) ! hostname ciscoasa domain-name xxx.internal names name 192.168.2.201 dev.xxx.internal description Internal Dev server name 192.168.2.200 Newserver ! interface Ethernet0/0 switchport access vlan 2 ! interface Ethernet0/1 ! interface Ethernet0/2 ! interface Ethernet0/3 shutdown ! interface Ethernet0/4 shutdown ! interface Ethernet0/5 shutdown ! interface Ethernet0/6 shutdown ! interface Ethernet0/7 shutdown ! interface Vlan1 nameif inside security-level 100 ip address 192.168.2.254 255.255.255.0 ! interface Vlan2 nameif outside security-level 0 ip address 192.168.3.250 255.255.255.0 ! ! time-range Workingtime periodic weekdays 9:00 to 18:00 ! ftp mode passive clock timezone GMT/BST 0 clock summer-time GMT/BDT recurring last Sun Mar 1:00 last Sun Oct 2:00 dns domain-lookup inside dns server-group DefaultDNS name-server Newserver domain-name xxx.internal same-security-traffic permit inter-interface object-group service Mysql tcp port-object eq 3306 object-group protocol TCPUDP protocol-object udp protocol-object tcp access-list inside_access_in extended permit ip any any access-list outside_access_in remark ENABLES OUTSDIE ACCESS TO DEV SERVER! access-list outside_access_in extended permit tcp any interface outside eq www time-range Workingtime inactive access-list outside_access_in extended permit tcp host www-1.xxx.com interface outside eq ssh access-list inside_access_in_1 extended permit tcp any any eq www access-list inside_access_in_1 extended permit tcp any any eq https access-list inside_access_in_1 remark Connect to SSH services access-list inside_access_in_1 extended permit tcp any any eq ssh access-list inside_access_in_1 remark Connect to mysql server access-list inside_access_in_1 extended permit tcp any host mysql.xxx.com object-group Mysql access-list inside_access_in_1 extended permit tcp any host mysql.xxx.com eq 3312 access-list inside_access_in_1 extended permit object-group TCPUDP host Newserver any eq domain access-list inside_access_in_1 extended permit icmp any any access-list inside_access_in_1 remark Draytek Admin access-list inside_access_in_1 extended permit tcp any 192.168.3.0 255.255.255.0 eq 4433 access-list inside_access_in_1 remark Phone System access-list inside_access_in_1 extended permit tcp any 192.168.3.0 255.255.255.0 eq 35300 log disable pager lines 24 logging enable logging asdm warnings logging from-address [email protected] logging recipient-address [email protected] level errors mtu inside 1500 mtu outside 1500 ip verify reverse-path interface inside ip verify reverse-path interface outside ipv6 access-list inside_access_ipv6_in permit tcp any any eq www ipv6 access-list inside_access_ipv6_in permit tcp any any eq https ipv6 access-list inside_access_ipv6_in permit tcp any any eq ssh ipv6 access-list inside_access_ipv6_in permit icmp6 any any icmp unreachable rate-limit 1 burst-size 1 icmp permit any outside no asdm history enable arp timeout 14400 global (outside) 1 interface nat (inside) 1 0.0.0.0 0.0.0.0 static (inside,outside) tcp interface www dev.xxx.internal www netmask 255.255.255.255 static (inside,outside) tcp interface ssh dev.xxx.internal ssh netmask 255.255.255.255 access-group inside_access_in in interface inside control-plane access-group inside_access_in_1 in interface inside access-group inside_access_ipv6_in in interface inside access-group outside_access_in in interface outside route outside 0.0.0.0 0.0.0.0 192.168.3.254 10 route outside 192.168.3.252 255.255.255.255 192.168.3.252 1 timeout xlate 3:00:00 timeout conn 1:00:00 half-closed 0:10:00 udp 0:02:00 icmp 0:00:02 timeout sunrpc 0:10:00 h323 0:05:00 h225 1:00:00 mgcp 0:05:00 mgcp-pat 0:05:00 timeout sip 0:30:00 sip_media 0:02:00 sip-invite 0:03:00 sip-disconnect 0:02:00 timeout sip-provisional-media 0:02:00 uauth 0:05:00 absolute timeout tcp-proxy-reassembly 0:01:00 timeout floating-conn 0:00:00 dynamic-access-policy-record DfltAccessPolicy aaa authentication telnet console LOCAL aaa authentication enable console LOCAL

    Read the article

  • The Partner Perspective from Oracle OpenWorld 2012 - IDC’s Darren Bibby report

    - by Richard Lefebvre
    Below is IDC’s Darren Bibby report on ‘The Partner Perspective from Oracle OpenWorld 2012’. If you missed the 2012 edition, I trust this will give you the willingness to attend next year one! October 26, 2012 I attended my fourth Oracle OpenWorld earlier in October. I always go in with the lens of, "What's in it for partners this year?" Although it's primarily thought of as a customer event - and yes, the bulk of the almost 50,000 attendees are customers - this year's conference was clearly the largest and most important partner event Oracle has ever run. Oracle PartnerNetwork (OPN) Exchange There were more partner attendees than ever, with Oracle citing somewhere around 5000. But the format for partners this year was different. And it was better. Traditionally, Oracle hosts a one-day only Partner Forum on the Sunday before the customer-focused conference begins. This year, the partner content still began on the Sunday, but the worldwide alliances and channels group created an exclusive track throughout the week, just for partners. It featured content specifically targeted towards partners, and was anchored at a nearby hotel. This was a great move for Oracle. The Oracle PartnerNetwork (OPN) team has been in a tricky position for years in that they have enough partners that they need a landmark event in the year, but perhaps not enough to justify a separate, worldwide, large, partner-only event. Coinciding a four day event with Oracle OpenWorld, where anybody who's anybody in the Oracle world attends anyway, is a good solution. The channels leadership team can build from this success for an even better conference next year. It's expected that they will follow a similar strategy. Cloud Announcements for Partners As for the content, it was primarily about the Cloud. For customers, for VARs, for ISVs, for everyone. There were five key Cloud related announcements for partners at the event: Cloud Builder Specialization. This is one of the first broader Specializations that isn't focused on one unique product. It is a designation for partners that offer design and implementation services for private cloud solutions. As such, it will surely be something that nearly every partner will consider, and many will pursue. New Specializations for Cloud Services. Unlike the broad, almost "strategy-level" Specialization above, there are a group of new product-based "merit badges" for many of the new Cloud offerings. Think about a Specialization for the Cloud version of HCM, for instance. Each of these particular specializations will also have Rapid Start implementation methodologies that allow a partner to offer a fixed scope and fixed price bid to customers. Based on the learnings from Oracle Consulting, this means a partner might be able to deliver Cloud HCM in six weeks for a fixed price. In the end, this means more consistent experiences for Oracle customers. Cloud Resale Program. For those partners who achieve one of these Cloud Specializations, it will mean they can actually resell the subscription-based Cloud product. This is important because it has been somewhat of a rarity in the emerging Cloud channel for partners to be able to "take the paper", take the revenue, do the billing, be first line of support etc. This is an important step for Oracle and one the partners will be happy to see. Cloud Referral Program. For those partners who are not as engaged with these specific Cloud products that the Specializations revolve around, there is a new referral program that provides an incentive to recommend Oracle Cloud products. This one-two punch of referral and resale programs is similar in many ways to other vendors who allow more committed partners to resell, while more casual partners can collect fees. It's the model that seems to work. The key to allow a company to resell a subscription product - something that is inherently delivered directly between the vendor and customer - is trust. Achieving a specialization is a good bar to have to meet. Platform as a Service for ISVs. Leveraging some of the overall announcements made by CEO Larry Ellison around a cloud version of its famous database, Oracle also outlined a new ability for ISVs to build cloud services on its new PaaS offering. Details were less available for this announcement, though it's an expected and fitting play for ISVs comfortable with Oracle technology who can now more easily build out cloud applications. There wasn't much talk of an app store to go along with this, but surely it's in the works. Specializations And "The Gap" Coming back to Specializations, Oracle PartnerNetwork (OPN) has 4600 partners worldwide that hold 20,000 Specializations. These are impressive numbers just three years into the new OPN framework. The actual number of Specializations has also grown significantly, up to 111 today and soon around 125 or so with the new Cloud designations. Oracle may need to look at grouping some of these and creating higher level, broader designations that partners could achieve by earning several Specializations in that group. At 125 and growing, this is a lot. On the top of the pyramid, Hitachi Ltd. successfully became the eleventh partner to make it to the highly prestigious Diamond level. Partner programs partially exist in order to recognize capable partners. And it's more than abundantly clear that the Diamond level does this. But I think Oracle has a gap. Specializations show capability in a very specific product area, and all sizes of partners can achieve these. The next level at which to show a level of expertise is the Advanced Specialization. However, this is a massive step up from the regular Specialization. The advanced level requires 50 people to have certification in that particular product area. Most other industry programs have similar higher level statuses, but none are even close to that number. Whereas a customer who sees an Oracle partner with an advanced specialization can be very sure of capability, there is a gap in that there are hundreds or even thousands of 20-50 person solution providers who are top notch in their area of expertise. They will never get to Advanced due to numbers alone. These boutique partners don't really have a way of showing off their talents in the current program. Advanced may not need to be so high to really show that a company has deep expertise. Overall it was a very successful Oracle OpenWorld for Oracle partners of all sizes. There was progress made on making it a bigger and more relevant event. And also on catching up and maybe even leading in some cases with cloud opportunities for partners.

    Read the article

< Previous Page | 17 18 19 20 21 22 23 24  | Next Page >