Search Results

Search found 65392 results on 2616 pages for 'http service'.

Page 7/2616 | < Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >

  • Does the SPDY protocol eliminate the need for cookieless domains?

    - by Clint Pachl
    With plain HTTP, cookieless domains were an optimization to avoid unnecessarily sending cookie headers for page resources. However, the SPDY protocol compresses HTTP headers and in some cases eliminates unnecessary headers. My question then is, does SPDY make cookieless domains irrelevant? Furthermore, should the page source and all of its resources be hosted at the same domain in order to optimize a SPDY implementation?

    Read the article

  • Understanding HTTP Cookies in Indy 10 for Delphi XE2

    - by Jerry Dodge
    I have been working with Indy 10 HTTP Servers / Clients lately in Delphi XE2, and I need to make sure I'm understanding session management correctly. In the server, I have a "bucket" of sessions, which is a list of objects which each represent a unique session. I don't use username and password to authenticate users, but I rather use a unique API key which is issued to a client, and has an expiration. When a client wishes to connect to the server, it first logs in by calling the "login" command, which is a path like this: http://localhost:1234/login?APIKey=abcdefghij. The server checks this API Key against the database, and if it's valid, it creates a new session in the bucket, issues a new cookie (unique string), and sets the response cookies with Success=Y and Cookie=abcdefghij. This is where I have the question. Assuming the client end has its own method of cookie management, the client will receive this login response back from the server and automatically save the cookies as necessary. Any future request from the client to the server shall automatically send along these cookies, and the client side doesn't have to necessarily worry about setting these cookies when sending requests to the server. Right? PS - I'm asking this question here on programmers.stackexchange.com because I didn't see it fit to ask on stackoverflow.com. If anyone thinks this is appropriate enough for stackoverflow.com, please let me know.

    Read the article

  • https:// search results appearing on Google for purely http:// site

    - by hydrurga
    I started weeding through my site's search results from Google today, using a site: search, to determine if there are any links that cause 404s and thus need redirecting. To my amazement I noticed numerous https:// results relating to various pages. My site doesn't have a SSL certificate, doesn't serve such pages, doesn't internally link to https:// pages, doesn't include any such files in its sitemap.xml and, for all of these, never has. I decided to do a Google search for https://<my site> and found one site that incorrectly refers to the root of my site with a https:// prefix - I will try to contact them to get them to correct this. I'm not sure however how Googlebot managed to index the non-root files as https://. I can't find any external links to them and surely, without certification, Googlebot should have stalled at the first request? I've just added the following lines to the site's .htaccess (although the surfer still has to navigate through the browser's "This site is a security risk. Abandon hope all ye who enter here!" message(s) first to get there): RewriteEngine On RewriteCond %{HTTPS} on RewriteRule ^(.*)$ http://www.<my site>.org/$1 [R=301,L] replacing <my site> with my domain name. My big question is this though - I would like to use the Google Webmaster Tools Remove URLs feature to remove the https:// pages from the index. Can I be guaranteed that this will only remove the https:// versions of each relevant page and not the valid http:// versions? My thanks to anyone who can help me out with this particular question and the issue in general.

    Read the article

  • SQL Server Service Broker Service Disappearing (Automatically Deleted)?

    - by mwigdahl
    I've implemented a messaging system over SQL Server Service Broker. It is working great, with the sole exception that every once in a while (maybe once per week per server) my initiator service just vanishes without a trace. The corresponding queue is still there, but the service is missing. Obviously this causes problems in my system. It's a simple matter to recreate the service by hand, but I'm confused as to what might cause this behavior. I understand that automatic poison message handling causes queues to be disabled, but I don't see anything that indicates services can be disabled or deleted automatically. When this happens, I usually have a large backlog of messages in multiple application queues, but nothing extreme. Total message backlog is around 200,000. Does anyone know what might be happening here?

    Read the article

  • Android Design - Service vs Thread for Networking

    - by Nevyn
    I am writing an Android app, finally (yay me) and for this app I need persistant, but user closeable, network sockets (yes, more than one). I decided to try my hand at writing my own version of an IRC Client. My design issue however, is I'm not sure how to run the Socket connectivity itself. If I put the sockets at the Activity level, they keeps getting closed shortly after the Activity becomes non-visible (also a problem that needs solving...but I think i figured that one out)...but if I run a "connectivity service", I need to find out if I can have multiple instances of it running (the service, that is...one per server/socket). Either that or a I need a way to Thread the sockets themselves and have multiple threads running that I can still communicate with directly (ID system of some sort). Thus the question: Is it a 'better', or at least more "proper" design pattern, to put the Socket and networking in a service, and have the Activities consume said service...or should I tie the sockets directly to some Threaded Process owned by the UI Activity and not bother with the service implementation at all? I do know better than to put the networking directly on the UI thread, but that's as far as I've managed to get.

    Read the article

  • Prevent service from starting

    - by Evan Plaice
    So, I do Arduino development on my system to program arduinos using the FTDI USB programming interface (if you have no idea what this means don't worry). The issue arises because the FTDI interface uses tty to communicate and it conflicts with one of the default ubuntu services. The default service in question is called brltty (which enables braille accessibility for people with impaired vision). Considering that I don't have any particular use for this service and it's annoying to stop it manually (using 'service brltty stop') after I restart my system... Where would I configure this (or any) service to prevent it from loading on startup? Note: I also have this issue with qemu-kvm conflicting with virtualbox.

    Read the article

  • WCF and Service Registry

    - by TK Lee
    I am about to build some WCF Services. Those services need to communicate to each others too, in some scenarios. I've done some "Google-ing" about Service Registry but can't figure out how to implement service registry with WCF; is there any other alternate? Is there any Microsoft technology available for Service Registry? I'm new to SOA and I will really appreciate any help or guidance (what and where should I exactly look for registry services).

    Read the article

  • Zookeeper naming service [closed]

    - by kolchanov
    I need a recommendation for naming service implementation. We are ISV and we have a lot of applications (services) with very different protocols such as http (Rest), low level tcp, amqp, diameter, telco protocols Rx, Ry, Ud and many others. We want to simplify configuration, deployment and service discovery procees and it seems that It's time to create central configuration registry. So I have few questions: - is zookeeper suitable for this purpose? - does exists more suitable and more special solution? - best practice for service naming for discoverin. Any standards? - recommendation for service configuration data structure Also we are keeping in mind future tasks For dynamic application distribution in a private cloud. Could you share your real life experience?

    Read the article

  • Python service using Upstart on Ubuntu

    - by Soumya Simanta
    I want to create to deploy a heartbeat service (a python script) as a service using Upstart. My understanding is that I've to add a /etc/init/myheartbeatservice.conf with the following contents. # my heartbeat service description "Heartbeat monitor" start on startup stop on shutdown script exec /path/to/my/python/script.py end script My script starts another service process and the monitors the processes and sends heartbeat to an outside server regularly. Are startup and shutdown the correct events ? Also my script create a new thread. I'm assuming I also need to add fork daemon to my conf file? Thanks.

    Read the article

  • Free web "caching" services for a web service

    - by Jason Banico
    I have a web service on Google App engine whose data is updated on a daily basis. To minimize bandwidth utilization from mobile clients connecting to it, I'd like to instead have an intermediary site where the clients will be getting their data from, and minimizing hits to my service to once or twice a day only. Is there such a service I can use? I'd like to explore this "pull" option first, before considering "push" options such as publishing to a blog site or a free website host that doesn't have bandwidth caps.

    Read the article

  • How to indicate to a web server the language of a resource

    - by Nik M
    I'm writing an HTTP API to a publishing server, and I want resources with representations in multiple languages. A user whose client GETs a resource which has Korean, Japanese and Trad. Chinese representations, and sends Accept-Language: en, ja;q=0.7 should get the Japanese. One resource, identified by one URI, will therefore have a number of different language representations. This seems to me like a totally orthodox use of content negotiation and multiple resource representations. But when each translator comes to provide these alternate language representations to the server, what's the correct way to instruct the server which language to store the representation under? I'm having the translators PUT the representation in its entirety to the same URI, but I can't find out how to do this elegantly. Content-Language is a response header, and none of the request headers seem to fit the bill. It seems my options are Invent a new request header Supply additional metadata in a multipart/related document Provide language as a parameter to the Content-Type of the request, like Content-Type: text/html;language=en I don't want to get into the business of extending HTTP, and I don't feel great about bundling extra metadata into the representation. Neither approach seems friendly to HTTP caches either. So option 3 seems like the best way that I can think of, but even then it's decidedly non-standard to put my own specific parameters on a very well established content type. Is there any by-the-book way of achieving this?

    Read the article

  • Calling Web Services with HTTP Basic Authentication from BPEL 10.1.3.4

    - by Ramkumar Menon
    Are you using BPEL 10.1.3.4 and hunting for the property names in the partnerlinkBindings that will work for outbound HTTP Basic Authentication? Here's the answer. <partnerLinkBinding ...>  <property name="basicHeaders">credentials</property>  <property name="basicUsername">WhoAmI</property>  <property name="basicPassword">thatsASecret</property></partnerLinkBinding>The drop down options in JDeveloper dont seem to work.

    Read the article

  • google chrome : http authentication issue on iframe

    - by Daniel Dzussa
    I have an HTML file with 2 links http://versionplus.in/pass/new.html, both links are in an iframe to load the contents inside the iframe. I have two protected directories one on the same server and other on another server. If you click on either link it will popup the login box, same for both links in all browsers except google chorme. google chrome doesn't show login box for the protected folder on another server, how can I fix this ?

    Read the article

  • How can I set up a 404 error page when people access http://ftp.mydomain.com?

    - by Tim B.
    I am a freelance videographer/developer, and part of my job involves transferring large files over FTP to production houses/television stations. While the majority of people in my industry understand the difference between FTP and HTTP, I've experienced several interactions in the past couple months of people who still open Internet Explorer and try to access http://ftp.mydomain.com, receive an error page served by HostGator, and tell me that they cannot access my FTP server. Instead of spending time delivering instructions via e-mail, I'd much prefer to serve up a custom error page in this instance that instructs them how to download and use an FTP client. I tried setting up a sub-domain in Cpanel hoping I could simply drop in an .htaccess file with the error page, but I got this error: ftp.mydomain.com domainadmin-domainexistsglobal I also tried creating a custom error page in PHP which reads the site URL and serves up the custom content only when http://ftp.mydomain.com is accessed. Unfortunately, the error page works for every subdomain except that one. I'm not entirely sure this is even technically possible, which is why I bring it to the good people of StackOverflow to help. Thanks!

    Read the article

  • Parse an HTTP request Authorization header with Python

    - by Kris Walker
    I need to take a header like this: Authorization: Digest qop="chap", realm="[email protected]", username="Foobear", response="6629fae49393a05397450978507c4ef1", cnonce="5ccc069c403ebaf9f0171e9517f40e41" And parse it into this using Python: {'protocol':'Digest', 'qop':'chap', 'realm':'[email protected]', 'username':'Foobear', 'response':'6629fae49393a05397450978507c4ef1', 'cnonce':'5ccc069c403ebaf9f0171e9517f40e41'} Is there a library to do this, or something I could look at for inspiration? I'm doing this on Google App Engine, and I'm not sure if the Pyparsing library is available, but maybe I could include it with my app if it is the best solution. Currently I'm creating my own MyHeaderParser object and using it with reduce() on the header string. It's working, but very fragile. Brilliant solution by nadia below: import re reg = re.compile('(\w+)[=] ?"?(\w+)"?') s = """Digest realm="stackoverflow.com", username="kixx" """ print str(dict(reg.findall(s)))

    Read the article

  • Simple HTTP Framework and time issue

    - by Deniz Acay
    I'm using Simple Framework for my recent project, and there is a problem with the date headers generated by the system. I can get my local time from Date class correctly, but setDate method of Simple's Response class sets date in milliseconds and when i do that by calling System.currentTimeInMillis, dates in response headers are incorrect. is there any way to solve this problem with Simple Framework? If not, what can i do except calling set method with a date string? Thanks in advance... EDIT: Date in the headers are in wrong timezone. And i also tried Calendar.getInstance().getTimeInMilis() method.

    Read the article

  • Finding an HTTP proxy that will intercept static resource requests

    - by pkh
    Background I develop a web application that lives on an embedded device. In order to make dev times sane, frontend development is done using apache serving static documents, with PHP proxying out to the embedded device for specifically configured dynamic resources. This requires that we keep various server-simulation scripts hanging around in source control, and it requires updating those scripts whenever we add a new dynamic resource. Problem I'd like to invert the logic: if the requested document is available in the static documents directory, serve it; otherwise, proxy the request to the embedded device. Optimally, I want a software package that will do this for me (for Windows or buildable on cygwin). I can deal with forcing apache to do it with PHP, but I'm unsure how to configure it to make it happen. I've looked at squid and privoxy, but neither of them seem to do what I want. Any ideas? I'd rather not have to roll my own.

    Read the article

  • Ideal HTTP cache control headers for different types of resources

    - by chris_l
    I want to find a minimal set of headers, that work with "all" caches and browsers (also when using HTTPS!) On my (GWT-based) web site, I'll have three kinds of resources: 1. Forever cacheable (public / equal for all users) These files don't ever change, and they get a filename based on the MD5 of their contents (this is GWT's approach). They should get cached as much as possible, even when using HTTPS (so I assume, I should set Cache-Control: public, especially for Firefox?) 2. Changing for every new version of the site (public / equal for all users) These files can be cached, but probably need to be revalidated every time. 3. Individual for each request (private / user specific) These resources (e. g. JSON responses) should never be cached unencrypted to disk under no circumstances. (Maybe I'll have a few specific requests that could be cached.) I have a general idea on which headers I would probably use for each type, but there's always something I could be missing.

    Read the article

  • How should clients handle HTTP 401 with unknown authentication schemes?

    - by user113215
    What is the proper behavior for an HTTP client receiving a 401 Unauthorized response that specifies only unrecognized authentication schemes? My server supports Kerberos authentication using WWW-Authenticate: Negotiate. On the first request, the server sends a 401 Unauthorized response with a body containing an HTML document. The behavior that I expect is for clients that support Kerberos to perform that authentication and for other clients to simply display the HTML document (a login form). It seems that most of the "other clients" I've encountered do work this way, but a few do not. I haven't found anything that mandates any particular behavior in this situation. There's a brief mention in RFC 2617: HTTP Authentication: Basic and Digest Access Authentication, but is there anything more concrete? It is possible that a server may want to require Digest as its authentication method, even if the server does not know that the client supports it. A client is encouraged to fail gracefully if the server specifies only authentication schemes it cannot handle.

    Read the article

  • Finding the Right Solution to Source and Manage Your Contractors

    - by mark.rosenberg(at)oracle.com
    Many of our PeopleSoft Enterprise applications customers operate in service-based industries, and all of our customers have at least some internal service units, such as IT, marketing, and facilities. Employing the services of contractors, often referred to as "contingent labor," to deliver either or both internal and external services is common practice. As we've transitioned from an industrial age to a knowledge age, talent has become a primary competitive advantage for most organizations. Contingent labor offers talent on flexible terms; it offers the ability to scale up operations, close skill gaps, and manage risk in the process of delivering services. Talent comes from many sources and the rise in the contingent worker (contractor, consultant, temporary, part time) has increased significantly in the past decade and is expected to reach 40 percent in the next decade. Managing the total pool of talent in a seamless integrated fashion not only saves organizations money and increases efficiency, but creates a better place for workers of all kinds to work. Although the term "contingent labor" is frequently used to describe both contractors and employees who have flexible schedules and relationships with an organization, the remainder of this discussion focuses on contractors. The term "contingent labor" is used interchangeably with "contractor." Recognizing the importance of contingent labor, our PeopleSoft customers often ask our team, "What Oracle vendor management system (VMS) applications should I evaluate for managing contractors?" In response, I thought it would be useful to describe and compare the three most common Oracle-based options available to our customers. They are:   The enterprise licensed software model in which you implement and utilize the PeopleSoft Services Procurement (sPro) application and potentially other PeopleSoft applications;  The software-as-a-service model in which you gain access to a derivative of PeopleSoft sPro from an Oracle Business Process Outsourcing Partner; and  The managed service provider (MSP) model in which staffing industry professionals utilize either your enterprise licensed software or the software-as-a-service application to administer your contingent labor program. At this point, you may be asking yourself, "Why three options?" The answer is that since there is no "one size fits all" in terms of talent, there is also no "one size fits all" for effectively sourcing and managing contingent workers. Various factors influence how an organization thinks about and relates to its contractors, and each of the three Oracle-based options addresses an organization's needs and preferences differently. For the purposes of this discussion, I will describe the options with respect to (A) pricing and software provisioning models; (B) control and flexibility; (C) level of engagement with contractors; and (D) approach to sourcing, employment law, and financial settlement. Option 1:  Enterprise Licensed Software In this model, you purchase from Oracle the license and support for the applications you need. Typically, you license PeopleSoft sPro as your VMS tool for sourcing, monitoring, and paying your contract labor. In conjunction with sPro, you can also utilize PeopleSoft Human Capital Management (HCM) applications (if you do not already) to configure more advanced business processes for recruiting, training, and tracking your contractors. Many customers choose this enterprise license software model because of the functionality and natural integration of the PeopleSoft applications and because the cost for the PeopleSoft software is explicit. There is no fee per transaction to source each contractor under this model. Our customers that employ contractors to augment their permanent staff on billable client engagements often find this model appealing because there are no fees to affect their profit margins. With this model, you decide whether to have your own IT organization run the software or have the software hosted and managed by either Oracle or another application services provider. Your organization, perhaps with the assistance of consultants, configures, deploys, and operates the software for managing your contingent workforce. This model offers you the highest level of control and flexibility since your organization can configure the contractor process flow exactly to your business and security requirements and can extend the functionality with PeopleTools. This option has proven very valuable and applicable to our customers engaged in government contracting because their contingent labor management practices are subject to complex standards and regulations. Customers find a great deal of value in the application functionality and configurability the enterprise licensed software offers for managing contingent labor. Some examples of that functionality are... The ability to create a tiered network of preferred suppliers including competencies, pricing agreements, and elaborate candidate management capabilities. Configurable alerts and online collaboration for bid, resource requisition, timesheet, and deliverable entry, routing, and approval for both resource and deliverable-based services. The ability to manage contractors with the same PeopleSoft HCM and Projects applications that are used to manage the permanent workforce. Because it allows you to utilize much of the same PeopleSoft HCM and Projects application functionality for contractors that you use for permanent employees, the enterprise licensed software model supports the deepest level of engagement with the contingent workforce. For example, you can: fill job openings with contingent labor; guide contingent workers through essential safety and compliance training with PeopleSoft Enterprise Learning Management; and source contingent workers directly to project-based assignments in PeopleSoft Resource Management and PeopleSoft Program Management. This option enables contingent workers to collaborate closely with your permanent staff on complex, knowledge-based efforts - R&D projects, billable client contracts, architecture and engineering projects spanning multiple years, and so on. With the enterprise licensed software model, your organization maintains responsibility for the sourcing, onboarding (including adherence to employment laws), and financial settlement processes. This means your organization maintains on staff or hires the expertise in these domains to utilize the software and interact with suppliers and contractors. Option 2:  Software as a Service (SaaS) The effort involved in setting up and operating VMS software to handle a contingent workforce leads many organizations to seek a system that can be activated and configured within a few days and for which they can pay based on usage. Oracle's Business Process Outsourcing partner, Provade, Inc., provides exactly this option to our customers. Provade offers its vendor management software as a service over the Internet and usually charges your organization a fee that is a percentage of your total contingent labor spending processed through the Provade software. (Percentage of spend is the predominant fee model, although not the only one.) In addition to lower implementation costs, the effort of configuring and maintaining the software is largely upon Provade, not your organization. This can be very appealing to IT organizations that are thinly stretched supporting other important information technology initiatives. Built upon PeopleSoft sPro, the Provade solution is tailored for simple and quick deployment and administration. Provade has added capabilities to clone users rapidly and has simplified business documents, like work orders and change orders, to facilitate enterprise-wide, self-service adoption with little to no training. Provade also leverages Oracle Business Intelligence Enterprise Edition (OBIEE) to provide integrated spend analytics and dashboards. Although pure customization is more limited than with the enterprise licensed software model, Provade offers a very effective option for organizations that are regularly on-boarding and off-boarding high volumes of contingent staff hired to perform discrete support tasks (for example, order fulfillment during the holiday season, hourly clerical work, desktop technology repairs, and so on) or project tasks. The software is very configurable and at the same time very intuitive to even the most computer-phobic users. The level of contingent worker engagement your organization can achieve with the Provade option is generally the same as with the enterprise licensed software model since Provade can automatically establish contingent labor resources in your PeopleSoft applications. Provade has pre-built integrations to Oracle's PeopleSoft and the Oracle E-Business Suite procurement, projects, payables, and HCM applications, so that you can evaluate, train, assign, and track contingent workers like your permanent employees. Similar to the enterprise licensed software model, your organization is responsible for the contingent worker sourcing, administration, and financial settlement processes. This means your organization needs to maintain the staff expertise in these domains. Option 3:  Managed Services Provider (MSP) Whether you are using the enterprise licensed model or the SaaS model, you may want to engage the services of sourcing, employment, payroll, and financial settlement professionals to administer your contingent workforce program. Firms that offer this expertise are often referred to as "MSPs," and they are typically staffing companies that also offer permanent and temporary hiring services. (In fact, many of the major MSPs are Oracle applications customers themselves, and they utilize the PeopleSoft Solution for the Staffing Industry to run their own business operations.) Usually, MSPs place their staff on-site at your facilities, and they can utilize either your enterprise licensed PeopleSoft sPro application or the Provade VMS SaaS software to administer the network of suppliers providing contingent workers. When you utilize an MSP, there is a separate fee for the MSP's service that is typically funded by the participating suppliers of the contingent labor. Also in this model, the suppliers of the contingent labor (not the MSP) usually pay the contingent labor force. With an MSP, you are intentionally turning over business process control for the advantages associated with having someone else manage the processes. The software option you choose will to a certain extent affect your process flexibility; however, the MSPs are often able to adapt their processes to the unique demands of your business. When you engage an MSP, you will want to give some thought to the level of engagement and "partnering" you need with your contingent workforce. Because the MSP acts as an intermediary, it can be very valuable in handling high volume, routine contracting for which there is a relatively low need for "partnering" with the contingent workforce. However, if your organization (or part of your organization) engages contingent workers for high-profile client projects that require diplomacy, intensive amounts of interaction, and personal trust, introducing an MSP into the process may prove less effective than handling the process with your own staff. In fact, in many organizations, it is common to enlist an MSP to handle contractors working on internal projects and to have permanent employees handle the contractor relationships that affect the portion of the services portfolio focused on customer-facing, billable projects. One of the key advantages of enlisting an MSP is that you do not have to maintain the expertise required for orchestrating the sourcing, hiring, and paying of contingent workers.  These are the domain of the MSPs. If your own staff members are not prepared to manage the essential "overhead" processes associated with contingent labor, working with an MSP can make solid business sense. Proper administration of a contingent workforce can make the difference between project success and failure, operating profit and loss, and legal compliance and fines. Concluding Thoughts There is little doubt that thoughtfully and purposefully constructing a service delivery strategy that leverages the strengths of contingent workers can lead to better projects, deliverables, and business results. What requires a bit more thinking is determining the platform (or platforms) that will enable each part of your organization to best deliver on its mission.

    Read the article

  • REST and redirecting the response

    - by Duane Gran
    I'm developing a RESTful service. Here is a map of the current feature set: POST /api/document/file.jpg (creates the resource) GET /api/document/file.jpg (retrieves the resource) DELETE /api/document/file.jpg (removes the resource) So far, it does everything you might expect. I have a particular use case where I need to set up the browser to send a POST request using the multipart/form-data encoding for the document upload but when it is completed I want to redirect them back to the form. I know how to do a redirect, but I'm not certain about how the client and server should negotiate this behavior. Two approaches I'm considering: On the server check for the multipart/form-data encoding and, if present, redirect to the referrer when the request is complete. Add a service URI of /api/document/file.jpg/redirect to redirect to the referrer when the request is complete. I looked into setting an X header (X-myapp-redirect) but you can't tell the browser which headers to use like this. I manage the code for both the client and the server side so I'm flexible on solutions here. Is there a best practice to follow here?

    Read the article

  • http requests, using sprites and file sizes -

    - by crazy sarah
    Hi all I'm in the process of finding out all about sprites and how they can speed up your pages. So I've used spriteMe to create a overall sprite image which is 130kb, this is made up of 14 images with a combined total size of about 65kb So is it better to have one http request and a file size of 130kb or 14 requests for a total of 65kb? Also there is a detailed image which has been put into the spite which caused it's size to go up by about 60kb odd, this used to be a seperate jpg image which was only 30kb. Would I be better off having it seperate and suffering the additional request?

    Read the article

< Previous Page | 3 4 5 6 7 8 9 10 11 12 13 14  | Next Page >