2023 - Scholars is Back Online

[Xiahou Mao]

If you have a theory to try out, it hasn't worked yet as of today. But maybe it's yet to be implemented!

[James]

If you have a theory to try out, it hasn't worked yet as of today. But maybe it's yet to be implemented!

Okay! I have narrowed down a possible cause for the issue and implemented part of a fix. I have not reset all the caches and everything as, theoretically, it shouldn’t be necessary. But I’m curious if this resolves some of the strange issues we have been experiencing. If issues continue, I’ll proceed with all the corresponding refreshes (e.g. changing the cookie name).

I did have to sign in again after making this change.

Fingers crossed…

[Xiahou Mao]

So far so good for me, though I'm not sure if I've closed my browser since the last time I logged in here. I'll be rebooting my computer later today, so that'll be the real test.

[SunXia]

Okay! I have narrowed down a possible cause for the issue and implemented part of a fix. I have not reset all the caches and everything as, theoretically, it shouldn’t be necessary. But I’m curious if this resolves some of the strange issues we have been experiencing. If issues continue, I’ll proceed with all the corresponding refreshes (e.g. changing the cookie name).

I did have to sign in again after making this change.

Fingers crossed…

Sorry, me this time, although it was on the browser on the phone, whereas it's not happened on PC for me so far. I use multiple browsers and have been unable to repeat the problem since it happened on my phone, (Android, Chrome) where it logged me out and then again when I hit the Post button for a reply. For a moment, the second time, it almost didn't seem to recognise my username for a few seconds but then it eventually refreshed and let me login.

Could be issues chaching acrosss devices and services but might as well leave info here just in case its useful.

[Xiahou Mao]

Closed my browser, came back, same problem still. I look logged in at first, but once I click a page/thread/notification, I'm dropped.

[SunXia]

Got a repeat of the error "Submitted form is invalid" seems to be the message on phone, if the error code is any help. It will log out and then when I'm logging in, it can say that message (ish) upto twice, when logging back in. Will resolve itself though mostly, but can be an inconvenient log out sometimes, after a post.

[James]

Got a repeat of the error "Submitted form is invalid" seems to be the message on phone, if the error code is any help. It will log out and then when I'm logging in, it can say that message (ish) upto twice, when logging back in. Will resolve itself though mostly, but can be an inconvenient log out sometimes, after a post.

Thanks all. It is this “submitted form invalid” issue in particular that has me intrigued. Because I imagine whatever it is that is causing problems is probably what is tripping up to create this issue.

Maybe next I’ll just try re-uploading all of the forum software to the server.

No credible risk of anything breaking or going away, though.

[ZL181]

So far, you have made some progress! When my SoSZ tab reloads due to high memory usage on my mobile device, it no longer logs me out, which was a huge frustration before. I hope the problem can be fixed once and for all but it is great that your improvements have worked!

[Xiahou Mao]

I think it might be good for me now. I'll keep it posted if that changes.

[James]

So far, you have made some progress! When my SoSZ tab reloads due to high memory usage on my mobile device, it no longer logs me out, which was a huge frustration before. I hope the problem can be fixed once and for all but it is great that your improvements have worked!

I think it might be good for me now. I'll keep it posted if that changes.

This is promising, at least. I’m still tinkering and making small optimizations that may somehow be involved. Since the major adjustment mentioned before, I haven’t had a strange signed-out-while-active sort of event.

Today I just disabled Varnish, a server-side caching solution that was running for the forum. I’m not sure if that is related as SimRTK was not experiencing regular issues and should have had the same configuration enabled, but there’s not really a good reason for that caching level to be running on the server and it can create issues.