asp.net 4.0 中自定义成员资格提供程序超时
我在我的项目中实现了自定义会员资格提供程序。我在数据库上保留了一个标志以使用户在线。我想在注销时重置此标志，如果用户正确注销，我可以执行此操作。即使浏览器直接关闭或在任何机器故障或正常超时时，我也必须重置数据库上的标志，会员提供商有任何技巧可以解决此问题。专家等待您的解答！！！！！！！
谢谢。
I have implemented custom membership provider in my project. i kept one flag on database to get the users online. i want to reset this flag on the log out, i can do this if the user log out properly . I have to reset the flag on database even if the the browser closed directly or on any machine failure or normal time out, any tricks on membership provider to solve this. experts am waiting for your answers!!!!!!!
Thanks.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
只需为 IS_LOGOUT_BUTTON_PRESSED 这样的用户添加一个字段 typeof(bool) default false
如果用户单击 LogOut，则将其设置为 TRUE。
如果用户关闭浏览器，它将保持错误状态。然后你就可以看到用户做了什么。
just put a field for users like IS_LOGOUT_BUTTON_PRESSED typeof(bool) default false
if the user click LogOut set it TRUE.
if the user closes browser, it will stay false. Then you can see what user did.
我担心你正在向风车倾斜。
您会注意到，sql 提供程序通过简单地更新大多数提供程序方法（以及随后的存储过程）中的最后一个活动来实现“在线用户”，然后使用预定义的“在线时间窗口”来确定用户是否可以 /em> 仍然在线。
要实现权威的“is_online”标志，需要您完全控制用户启动会话和结束会话的方式，但由于浏览器 ui 的性质和会话的方式，您没有也不能这样做。并且票证的有效期也会被推迟。
因此，您永远无法真正确定用户是否在线，从而使您为跟踪该指标所做的任何努力都变得不可靠。
不可靠的数据通常比没有数据的价值还要低。
我非常确定，在尝试将绝对性（如您的问题中所述）应用于会话和票证的无状态和自由形式时，您只会感到沮丧。
抱歉我没有更好的消息。
I am afraid that you are tilting against windmills.
You will notice that the sql providers implement the 'users online' by simply updating last activity in most all provider methods (and subsequently the stored procedures) and then using a predefined 'online time window' to determine if the user may still be online.
To implement an authoritative 'is_online' flag would require that you have complete control over the user's means of starting a session and ending a session, which you do not and can not do due to both the nature of the browser ui and the way that sessions and ticket expirations are bumped.
So you can never really be sure whether a user is online or not, rendering any efforts you have made to track this metric unreliable.
And unreliable data is often of less value than no data at all.
I am quite certain that you will find nothing but frustration in attempting to apply an absolute, as described in your question, to the stateless and freeform nature of sessions and tickets.
Sorry I have not better news.