Login session Puns

When ever I login to this one user, SDDM freezes and takes 10 minutes to bring me to a plasma session and while doing so I can still move my mouse, I only have this problem with one user. I couldn't find any sort of solution from searching online as those that had a similar problem didn't have a workable solution.

I'm using openrc for my init system and the relevant information from my sddm.log file is here. I can provide any additional information if needed.

I was without a computer for a few months, and things changed with Firefox (for the better I guess). However I have some problems which I cannot find the fix for. The temporary containers went missing, and I have to login to every website every-time I close/open the browser. I created a separate container for reddit, and still have to login every time. I have Decentraleyes, CookieAutoDelete, and FireFox Multi Account Containers active. Any help or guidance is appreciated.

I think this is somewhat privacy related, so I decided to ask. I'm paranoid about XSS software attacks, so, I decided "Why not just delete my cookies?". I don't think I thought that through fully and didn't check if that also completely gets rid of them.

So, does deleting cookies stop XSS attacks or invalidate your login sessions?

For example, say I delete all my cookies from chrome or firefox. Are they gone permanently (no one can use them to login if they stole them) or?

P.S, does anyone know if this link is fine?

When a webmail session times out, I think it would make more sense to direct you back to the new login page instead of the homepage. As it is now, an extra step is required to login again.

I have an SPA frontend that connects to the DRF api using token auth (using header authentication). The SPA is served from the static assets in the django app, so they are on the same domain. I have a requirement that we be able to link to the Browsable API docs which DRF supplies. I CAN do this, but it requires re-logging in to the API server using username/password and not the user token. How do I enable token login via the Browsable API login form, and also set the session cookie in my SPA so users don't have to login twice?

When I make a request from my frontend to my backend (Flask), to a route FOO, which has the proper CORS headers attached to it (by using Flask-CORS), my browser sets 2 cookies when 1 request is made to that endpoint. As a result, I end up having 2 sessions stored in my Redis instance (created by using Flask-Session).

From what I can tell, a cookie is set after my browser makes the preflight request (an OPTIONS request) as part of the CORS request. After the OPTIONS request succeeds, my browser makes a GET request as part of the CORS request, which again sets a cookie.

I think this is happening because Flask-Login is treating each of these 2 separate HTTP requests as an anonymous user, and each anonymous user is stored in the form of a session (which leads to the response from the route telling the browser to set cookies for each of these 2 requests).

I don't want the OPTIONS request in this context to be treated as a "user" by Flask-Login; it doesn't make sense to store a session for it. I'm not sure how to fix this issue and just get 1 session created for 1 CORS request at a time.

My Flask code configuring Flask-CORS:

    # create_app() follows the app factory pattern
    app = create_app()
    CORS(app, origins=["http://localhost:3000"])
    app.run()

My route:

@authentication.route('/dummy')
@cross_origin(supports_credentials=True)
def dummy():
    return jsonify({'data': 'generic data'})

What my frontend CORS request looks like:

const frontendRequest = async (response) => {
    await axios
      .get("http://localhost:5000/authentication/dummy", {
        headers: {
          Authorization: "dummyInfo",
        },
      })
      .then((responseAuth) => {
        console.log("testing success response", responseAuth);
      })
      .catch((errAuth) => {
        console.log("testing error", errAuth);
      });
  };

The requests in Firefox console:

What the requests look like in Firefox console (hosted on imgur)

A few months ago we deployed AVD, backed by Windows 10 Enterprise Multi-Session session hosts.

Our biggest complaint has been login times. Usually it's about 90 seconds at the Windows login screen (applying policies, etc), and then another 30-60 seconds of sluggishness post-login. So a good 2-3 minute login. This is what it was like back in the Win2K3 days...

Given these are always-on servers, plus FSLogix, I'm surprised at how slow this is. At a prior company we had Server 2016 session hosts w/ FSLogix, and logins were like... 10 seconds. Maybe another 10 seconds before all your Windows profile stuff and anti-virus fully loaded. But ultimately a fully-loaded desktop ready to go in about 15-25 seconds - what you would expect of a modern Windows login experience on reasonably fast disk.

Is this normal? All of our infrastructure is in azure, and some 3rd party consultants who architected WVD/AVD keep saying 'well this is normal'. Am I crazy?

FWIW - The Win10 Session hosts are 16 core, 64GB, with premium SSD (3500 IOPS, 150MB/s).

We DON'T have a lot of users/logins. Maybe 10-15 per VM.

Is this normal? I'm new to AVD/WVD.

I would like to implement a "keep me logged in" feature.

My question is now that is the best way to store session data, e.g. the shopping basket, which should persist throughout different sessions.

I guess I could...

1. Read and write these data to/from a database directly without using $_SESSION at all.

2. Serialize the $_SESSION variable to a database once it has been finalized, and deserialize when the session is unset, but the login and security-token cookies are valid.

    session_write_close();
    $array = base64_encode(serialize($_SESSION));
    pg_update(..., $array),
    ...
    $_SESSION = unserialize(base64_decode($row['session_vars'])); // coming from the database
   

3. Or I could perhaps write a custom IO handler for $_SESSION such that the data is stored in a database and somehow use that to restore old sessions.

I have tons of money in crypto on this account, and I had bought it all when it was blockfolio, and I can see my portfolio in the background of the popup, but I cant access it because all I can click is the option to sign out. When I sign out, and sign back in, the same pop-up comes up again and I still cant access my money. Help please.

Not really sure what to do but i can no longer access Bitwarden through Unraid. When a user tries it says login session expired. When I try to login to the admin page, it says my token is wrong. I check the docker config and the file itself. Tried those passwords and that didnt work either.

Any suggestions?

SOLVED! Deleted the RSA files, started the docker and it works again!

So just to preface this, this is my absolutely first ever time dealing with containers, and I haven't used plain Debian much and the last time was a few years ago, this is also on a Gentoo host.

In order to setup an LXC container to be managed by libvirt I have been performing this tutorial (subbing in ubuntu for debian): https://www.gahome.org/virt-manager-libvirt-lxc-and-os-containers/

Everything in the tutorial went just fine, until I went to log into root after starting the container. After trying to login I am greeted by the login motd message. Also listing the last login date and time.

However it right after prints out the error in the title and goes right back to the login prompt.

There are also errors further up in the boot text while the container is starting up that might be relevant:

Failed to create symlink /sys/fs/cgroup/net_cls: file exists
Failed to create symlink /sys/fs/cgroup/cpuacct: file exists
Failed to create symlink /sys/fs/cgroup/cpu: file exists

I belive these may be coming from libvirt as they are right before the welcome to debian message.

Then right after setting the hostname the following errors show up:

Coulden't move remaining userspace processes, ignoring: input/output error
-.slice: Failed to migrate controller cgroups from , ignoring: Input/Output error

I have no idea whats causing the input/output errors, smartctl says the disk should be performing just fine (although it is a very old one).

Any ideas on what I should try? I can chroot in and edit stuff in the container as needed.

Just had some odd occurrences of sessions timing out and couldn't find the session data in LocalStorage where I swear it was previously.

Logging out and logging back in seemed to do the trick and auth session data is in IndexedDB.

Can't seem to find any info suggesting this was recent and I've been on js SDK V8 for ages now.

Basically, I've installed dropbox, copied a minecraft folder that was around 500 MB in the dropbox folder on my machine and found my disc full even after deleting and purging dropbox from my system (cache problems, probably). I've then encoutered the dreaded login loop caused by 100% full "/" partition, and acted upon it via recovery mode to free up some space, using various instances of autoclean, clean, autoremove ecc, plus manual deleting. Now I find that my Minecraft installation already has its file in place (nothing has changed) and yet the actual program does not see it, as if it has resetted itself. Would a fresh reinstall of the program be beneficial?

Also, why there's a huge discrepancy (few gigs) between baobab and the df command ??

Hello. I downloaded neovim appimage as per the github page.

I ran the following commands:

curl -LO https://github.com/neovim/neovim/releases/latest/download/nvim.appimage
chmod u+x nvim.appimage

When I run the appimage with

./nvim.appimage

I get this message:

Usage: login [-p] [name]
       login [-p] [-h host] [-f name]
       login [-p] -r host

If I run sudo ./nvim.appimage , nvim opens up correctly.

The github page says that if the ./nvim.appimage command does not work, I should run

./nvim.appimage --appimage-extract
./squashfs-root/AppRun --version

# Optional: exposing nvim globally
mv squashfs-root / && ln -s /squashfs-root/AppRun /usr/bin/nvim
nvim

Running ./squasfs-root/AppRun --version yields the same message as before

When I run the optional part, I get permission denied. So I try with sudo, and it works and the link is created.

Still, when I run the nvim command, the same message appears, just like before.

If I do sudo nvim, nvim opens up correctly.

My workaround is to add

alias nvim="sudo ~/nvim.appimage"

to .zshrc. I am unhappy doing this, as I want to find the real cause behind this issue and learn from it.

In short, my question is: Why does this message appear when I run nvim without sudo? I tried changing permissions using chmod, but to no avail.

Thank you for your time

Hello as the title says I get that same error when trying to connect with protonvpn-cli connect, when I try to login it says Successful login but I still get the same error. I'm on Arch, without any desktop environment and from what I've seen in this thread proton needs a desktop environment to work? Any ways to resolve this issue? Thanks!

If you are seeing a message "Your session has expired, please login again" when using Joplin Cloud, simply close your app and open it again to fix it. This issue will be addressed in the next desktop and mobile versions.

Can't believe this is not a thing but Fidelity really needs to redirect to the login page after the browser session expires in the web version. As it stands you can leave the window open and it will stay on your account summary page indefinitely until you start clicking around on the interface, or refresh the page. Your account numbers and balances stay up on the screen!

Life happens sometimes you forget to log out. This should be really easy to implement and not require a user to manually refresh or log out if the user forgets to log out of an expired session. Think this should go without saying but this is exposing customers by leaving financial information up on the screen once a user has walked away from their workstation.

Its happening when i try to sign in, it gets me to account but it stops after that and i cannot do anything. Is anyone else having this happen?

A coworker asked me to see if NetSuite kept record of session information so that he can include his time worked after hours in his next salary negotiation. The login audit is already more detailed and helpful than he expected, but I’d be interested to know if I can get more out of it. Or if there’s some other kind of search I need to be starting with to get this. If it’s not possible that’s alright — as I said, the login times are pretty useful on their own. Thanks!

So not sure what I did, but I can't login. I get to login screen and get error, failed to load session. My wallpaper was also reset.

Got into terminal and my username/password works. Not sure what to do from here.

I must have broken something with budgie. Any ideas on what to try?

I did sudo eopkg up and updated, but still can't login.

I have not been able to resolve this error, spanning multiple hours/PC restarts/clean reinstallation of the client/trying any and all suggestions found on various forums. This is why everyone understands this region to be such a joke. BTW, other accounts on my PC work without any issue, and this account works on other PC....

https://preview.redd.it/uh3d3eo32qm71.png?width=1916&format=png&auto=webp&s=ef04832bb09eb9485180024e470cea2c16796e9f

I heard This is actually only supported by entreprise edition, which is a shame