I have faced this question in my Interview as well. I do have many confusion with Session Scope & it management in java.
In web.xml we do have the entry :
<session-config>
<session-timeout>
30
</session-timeout>
</session-config>
What does it indicate actually ? Is it scope of whole project ?
Another point confusing me is how can we separate the session scope of multiple request in the same project? Means if I am logging in from a PC & at the same time I am logging in from another PC, does it differentiate it ?
Also, another confusing thing is the browser difference. Why does the different Gmails possible to open in different browsers ? And Gmail can prevent a session from Login to Logout. How is it maintained with our personal web ?
Session management is not something limited to Java and servlets. Here's roughly how it happens:
The HTTP protocol is stateless, so the server and the browser should have a way of storing the identity of the user through multiple requests
The browsers sends the first request to the server
The server checks whether the browser has identified with the session cookie (see below)
3.1. if the server doesn't 'know' the client:
the server creates a new unique identifier, and puts it in a Map (roughly), as a key, whose value is the newly created Session. It also sends a cookie response containing the unique identifier.
the browser stores the session cookie (with lifetime = the lifetime of the browser instance), containing the unique identifier, and uses it for each subsequent request to identify itself uniquely.
3.2. if the server already knows the client - the server obtains the Session corresponding to the passed unique identifier found in the session cookie
Now onto some the questions you have:
the session timeout is the time to live for each session map entry without being accessed. In other words, if a client does not send a request for 30 minutes (from your example), the session map will drop this entry, and even if the client identifies itself with the unique key in the session cookie, no data will be present on the server.
different gmails (and whatever site) can be opened in different browsers because the session cookie is per-browser. I.e. each browser identifies itself uniquely by either not sending the unique session id, or by sending one the server has generated for it.
logging from different PCs is the same actually - you don't share a session id
logging-out is actually removing the entry for the session id on the server.
Note: the unique session id can alternatively be stored:
in a cookie
in the URL (http://example.com/page;JSESSIONID=435342342)
2 or 3 other ways that I don't recall and aren't of interest
What does it indicate actually ?
The lifetime of a session. The session expires if there is no transaction between the client and the server for 30 minutes (per the code segment)
Is is scope of whole project ?
It has application scope. Defined for each web application
Another point confusing me is how can
we separate the session scope of
multiple request in the same project?
Means if I am logging in from a PC &
at the same time I am logging in from
another PC, does it differentiate it ?
Yes. The session ids (JSESSIONID for Apache Tomcat) will be different.
Also, another confusing thing is the
browser difference. Why does the
different Gmails possible to open in
different browsers ?
Each login by the same user from a different browser is a different session altogether. And the cookies set in one browser will not affect in another. So different Gmail instances are possible in different browsers.
And Gmail can prevent a session from
Login to Logout. How is it maintained
with our personal web ?
Persistent cookies
Servlets in Java have an HttpSession object which you can use to store state information for a user. The session is managed on the client by a cookie (JSESSIONID) or can be done using URL rewrites. The session timeout describes how long the server will wait after the last request before deleting the state information stored in a HttpSession.
The scope is per browser instance, so in the example you give logging in from two different pcs will result in two session objects.
if you open the same application in different window i mean multiple instance of a browser it will create different session for every instance.
I recommand Apache Shiro for session management,Authentication and authorization.
I take it back.
As #BalusC commeneted below, only servlet container is in charge of managing the http session. Shiro is just using that. It will hook to HttpSession via a filter you explicitly define.
we have 4 ways to manage a session.
1.Cookies
2.URL rewriting
3.Hidden form fields
4.HTTP session
the fourth one is powerful and mostly used now-a-days.
Related
I am working on a Java web application which involves allocating server resources on a per http session basis. So the http session ID would become part of the key of a database table, and its uniqueness in the table is a must. The description of the Java HttpSession.getID() method claimed that this ID is unique, so the server side should have someway to enforce such nature. But on the client browser, HTTP session is usually shared by multiple browsing windows unless explicitly request a new session to be to used. This suggests that the browser have active control over session usage as well. So which side is responsible for the generation of HTTP session ID? Or it is a combined effort of both sides? Is there any chance that the same session ID is generated accidentally by browsers on different computers accessing the same website?
Session id is generated by server and is usually granted to a visitor on his/her first visit to a site.
A session ID is a unique number that a Web site's server assigns a
specific user for the duration of that user's visit (session). The
session ID can be stored as a cookie, form field, or URL (Uniform
Resource Locator). Some Web servers generate session IDs by simply
incrementing static numbers. However, most servers use algorithms that
involve more complex methods, such as factoring in the date and time
of the visit along with other variables defined by the server
administrator.
1 - Where session variables is stored in java web application ? on the client or the server side ?
2 - If i put a lot of objects and variables through the session it will slow down client's requests ?
P.S In my case i use spring mvc.
The "session" variable consists of two pieces, a very small session identifier which is stored on the client usually named jSessionId and stored as a cookie. But, the sessionId may also be encoded into URLs.
The second piece of a session is the actual data, and it is stored on the server. Possibly in a server-side database if the your server is part of a multi-server cluster. Each session is identified by that sessionId and the client sends it with every request. That is why it is designed to be very small.
Simple answer is : your session data are stored on the server side.
Web browser will get only an string id to identify it's session.
In fact, spring security takes more care of session information, because if users even don't login, session may not exist at all.
When I use spring mvc only, I don't use session to store important data, because session is stored in memory only. It is designed to save data temporarily.
When I use spring security, I have to save many important things in memory, such as account data which could not be transmitted on internet, and I won't load them from database every time. So I have to choose session.
So when the server which stored login session is down, all users have logged in on this server would have to relogin to retrieve another session id.
Session is not always the best choice, because when we have many servers that use session data, I have to share the data among all servers, anyway, net IO is expensive for servers.
1.it stored on server
2.Session stored on server ,so the Object you set in it may also stored on server.Request only send a SessionId to server to indentify this users Session to other users Session.
I am having some issues with my web application while doing a performance test with Jmeter. My question is not around Jmeter instead, it's around a simple Servlet session management behavior.
So we have a web application, where in when you request a login page, it passes back a "Session Id" in response headers and that is used for subsequent request made by browser. Session Id is passed along with username and password and if authenticated a new session id is returned and session is maintained with that session id going ahead.
This is using cookies.
Now in Jmeter we have a thread based approach for load testing. When I run threads parallely what is essentially happening is that each thread request a login page and somehow only the last thread to request login page is authenticated as I feel that subsequent login page that comes with a new session id in cookie, invalidates the old or other session ids.
This is inspite of the fact that each thread is a different session and has it's own cookie manager. It's quite wierd.
However my questions are:
Does it make sense to have session id coming with login page; I see that maybe session is created as soon as application is accessed, but is it that, what sets a new cookie with session id? This application was already written so I am just wondering.
If each thread's session id is being overriden in jmeter does that mean, that i am not able to allocate a seperate cookie manager properly? Also even if threads are different sessions is there a possibility, old session id or cookie would be discarded?
How would Server know to invalidate the session id/cookie for subsequent requests? I am sure, not basis of IP address of requestor, as different browsers would still let me open parallel multiple sessions.
Any ideas, clarifications and light on the issue would be much appreciated.
Server will just timeout the session it definitely sounds like you are over writing cookies here.
Have you tried seeing how it manages by including session ids in the url (simulate cookies being disabled in the browser) Does this work?
Has this been solved at all?
I would say that having a session id in the login page is quite unusual. Session are meant for storing data server side and associate that data to connected users. Of course, prior to authenticate the user, there shouldn't be any data server side.
Best way managing session in Java. I heard that cookies are not reliable option for this as they gets stored into browser and can be accessed later on? Is this correct? If possible please come up with the answers with the coding example.
Which is the best among:
URL Rewriting: Server will add an additional parameter at the end of URL link
Hidden parameter in Form: server will add an additional parameter at every form in HTML
cookie: Server will ask browser to maintain a cookie.
The session management (client identification, cookie handling, saving session scoped data and so on) is basically already done by the appserver itself. You don't need to worry about it at all. You can just set/get Java objects in the session by HttpSession#setAttribute() and #getAttribute(). Only thing what you really need to take care of is the URL rewriting for the case that the client doesn't support cookies. It will then append a jsessionid identifier to the URL. In the JSP you can use the JSTL's c:url for this. In the Servlet you can use HttpServletResponse#encodeURL() for this. This way the server can identify the client by reading the new request URL.
Your new question shall probably be "But how are cookies related to this? How does the server do it all?". Well, the answer is this: if the server receives a request from a client and the server side code (your code) is trying to get the HttpSession by HttpServletRequest#getSession() while there's no one created yet (first request in a fresh session), the server will create a new one itself. The server will generate a long, unique and hard-to-guess ID (the one which you can get by HttpSession#getId()) and set this ID as a value of the cookie with the name jsessionid. Under the hood the server uses HttpServletResponse#addCookie() for this. Finally the server will store all sessions in some kind of Map with the session ID as key and the HttpSession as value.
According to the HTTP cookie spec the client is required to send the same cookies back in the headers of the subsequent request. Under the hood the server will search for the jsessionid cookie by HttpServletRequest#getCookies() and determine its value. This way the server is able to obtain the associated HttpSession and give it back by every call on HttpServletRequest#getSession().
To the point: the only thing which is stored in the client side is the session ID (in flavor of a cookie) and the HttpSession object (including all of its attributes) is stored in the server side (in Java's memory). You don't need to worry about session management youself and you also don't need to worry about the security.
See also:
Authenticating the username, password by using filters in Java (contacting with database)
How to redirect to Login page when Session is expired in Java web application?
How to implement "Stay Logged In" when user login in to the web application
All Java web frameworks support cookies or URL-encoded session IDs. They will chose the correct approach automatically, so there is nothing you need to do. Just request the session object from your container and it will handle the details.
[EDIT] There are two options: Cookies and a special URL. There are problems with both approaches. For example, if you encode the session in an URL, people can try to pass the session on (by putting the URL into a mail, for example). If you want to understand this, read a couple of articles about security and build app servers. Otherwise: Your Java app server will do the right thing for you. Don't think about it.
The cookie just stores the session ID, this ID is useless once the session has expired.
Servlet specification defines the API for accessing/setting session data in standard J2EE application. Also it defines that session data is stored on the server-side and nothing is transferred to the client except the session identifier. There are 2 mechanisms how session id is transferred:
1) request URL e.g. jessionid=....
2) cookie
Mechanism is determined automatically based on client capabilities.
EDIT. There is no best option, there is servlet specification that defines the way.
Http is a stateless, client-side pull only protocol.
To implement a stateful conversation over it, Java EE Web Server need to hide some information (which is sessionid) in client-side and the mechanism it can use should follow HTTP and HTML spec.
There are three ways to accomplish this goal:
URL Rewriting: Server will add an additional parameter at the end of URL link.
Hidden parameter in Form: server will add an additional parameter at every form in HTML.
cookie: Server will ask browser to maintain a cookie.
Basically, modern web server will have a "filter" to choose which way to use automatically.
So if Server detected that browser already turn off cookie support, it will switch to other ways.
2 important questions:
Which web technology are you using? JSF, Struts, SpringMVC or just plain servlets/JSPs.
Servlets/JSPs already give you the session support you need. JSP Example: Hello, <%= session.getAttribute( "theName" ) %>
I really don't think you have something to worry about cookies, since the data is stored safely in the server and handeling the cookie is done automaticlly.
Is your application installed on a single server?
If YES than you have no problem, use the servlet session option.
if NO than you gotta find another way to do this. Like using a sticky session, or maybe parse the entire session object in the requests/responses as a field. This option indeed requires you to take security measures.
I work on a task that involves moving/traversing from one application to another. The applications are in separate JVMs.
While traversing to the other application, I keep track of the session ID. However, as I traverse back and forth, a new session gets created. Is there any way for me to get back the same session, using the sessionId that I retain, when I navigate back into my parent application from a child application?
Environment: J2EE with WebSphere.
As mentioned by Mork0075, the sessionID is tied to the cookie name and the server domain. If you're using the same server domain for two apps on separate JVMs, I see two options to maintain the session when switching between applications:
The long shot:
1) If you're using a database for session replication purposes, you can use the same database for both applications, and the sessionID will be available for both apps. The one problem I see here is that the objects in the session may not be available on both sides, since the code would be different etc. They'd probably clobber the other side's session objects unless you maintained the code and such on both sides so the objects were available.
The likely possibility:
2) Use different cookie names for the session on one of the two apps. By default, sessions use JSESSIONID as the cookie, and when you switch to the second app, it tries to look up a session based on that cookiename and can't find it. So, it creates a new sessionID and sends it back to the browser, thus causing your sessionID to change and not be available when you switch back to the original app. However, if you change the second app's sessionID to something else (say, JSESSIONID2) your browser would end up with two valid sessionIDs that would each be valid on their correct application. You can change the name via the administration console under the application server's Session Management->Enable cookies page.
I'am not sure if this helps, but in a ONE application scenario, you would submit a sessionID with every reponse, save it in the URL, a cookie or as a hidden field. By submitting a new request to the server, the sessionID is also submitted, to resolve it at server side. In my understand switching from one application to another means, that you have to give the sessionID with the user, across the applications. If you save the sessionID in a cookie, this perhaps is not possible, because the cookie is restricted to a certain server domain. So ensure that the session is still valid and the sessionID is present after returning to the application started.
You shouldn't have to do this manually. Most app servers support Single Sign On (SSO) so that you can log in to one application and have access to all the applications in the same SSO domain. The app server will keep track of session ids and link them to an HTTPSession object specific to the web app.
See http://publib.boulder.ibm.com/infocenter/wasinfo/v6r1/topic/com.ibm.websphere.base.doc/info/aes/ae/csec_sso.html