This is a rant about how so many apps on many different platforms (TVs, mobile devices, computers, etc…) have decided to not actually show detailed errors any more. Instead, we get something along the lines of:
Oops, somehting went wrong. Please try again later
… and then, well, we get to figure out what just happened and what in the world we need to do about it. And good luck with that, since you have no idea what just failed.
Why software developers?!? Why have you forsaken us?
EDIT 24 hours later: I feel like I need to clarify a few things:
I’ve worked for 8 software companies over 30+ years. I know why putting a DB error into the message users see is a bad idea. I know that makes me uncommon, but I still want more info from these messages.
You all are answering as if there are only two ways this can work: (a) what we have now (which is useless), and (b) a detailed error listing showing a full stack trace. I think the developers could meet me half-way.
What I want is either (a) “Something went wrong on the server, you can’t fix it, but we will” or (b) “Something on your end didn’t work. Check your network or restart the app or do something differently and then try the same thing again”. And if they’re blocking me because I’m using a VPN, fucking say so (but that’s a whole separate thing…)
Some apps do provide enough info so I have a clue what I should do next, and I appreciate the effort they put into helping me. I think what I am really ranting about is I want more developers to take the time to do this instead of reporting all errors with “Oops, try again”. (If the error is in their server, why should I try again?) Give me a hint as to the problem, so I have something to go on.
Cheers y’all. Still love you my techy brothers and sisters.
You must log in or register to comment.
Because 99% of the time these errors are caused by something on their end that the user is unable to fix, even on the off chance that they understand the problem in the first place. So there isn’t any need to give you more information than “something went wrong, please wait a minute and/or try again”.
OK but then inherent in what you’re saying is also the message, “… and don’t contact us about this, because we don’t want to deal with it” which is also mildly infuriating to me.
You’re assuming they aren’t already aware of the issue.
Sorry but how does that help me?
What I’m saying is that when you see one of these messages you should interpret it as “something is wrong on our end, nothing you can or need to do on your end, please hang tight as we’re aware of the issue and working on it”. They don’t give you more info than that because that average person is probably not a dev and doesn’t have any need for more details than that.
But it’s MY Internet, and I want it NOW!!
How does telling someone about a problem they’re already aware of help you?
When you’re in a restaurant and your server accidentally dumps the entire tray of food in your lap, would you prefer them to apologize or just stare at you because you already know what the problem is?
The “we don’t want to deal with it” part is something you’re attributing to them with no evidence. As a former SRE, I can guarantee you they are dealing with it.
Iit’s an internal error that is not handled properly. They don’t want to tell you the exact error message and detailed information around that, because it would expose the internal state of the backend and that would be a security issue. There is really nothing more that they can tell you, except that a developer needs to look at this (and possibly thousands to tens or hundreds of thousands of similar logged errors) and they probably already are.
LMFAO. I probably have to truncate at least five error log files a week on various vps servers at my company because they fill the SSD and crash the OS. We rent servers we don’t dev them for our cx.
Largest error file I’ve seen so far is 32 GB
Site owners are normally clueless. Site developers normally can’t give a single fuck and systems administrators like me. Get to pick up the pieces and tell them to tell their Deb to fix it and then we pick it up again and tell them to tell their Dev to fix it let me know when you sense a pattern
Are you not rotating your logs with for example logrotate?
So literally just found somebody’s exim main log archive.gz that was 12 GB and this is an archive from 2023. I have no idea why it even exists on this mans server still and was not reclaimed with the rest of its older brothers
It’s almost never an internal error. The vast majority of the time it’s vpn blocking or some such bullshit.
Maybe then, the message could be, “An internal error has occurred and we’re going to work on fixing it but there’s nothing you can do to fix it yourself right now”. It’s the “Oops” that fries my grits.
I do agree, the whole “oops sowwy” with a sad Labrador vibe is a little irritating. But I guess they do it cause it’s a harmless and layman-friendly response.
If you’re tech-savvy enough to want detailed error messages, you should also be tech-savvy enough to understand the implied message you just typed out. The ‘Oops’ isn’t for you, it’s for the average user.
Error messages are a common way for hackers to gain information about a system. Useless error messages are recommended for security.
If you enter your username as Robert’'); DROP TABLE Students;-- giving the error “Oops, something went wrong” is better than “NoSuchTable: ‘Students’ Table doesn’t exist in the database” because now the hacker knows you’re using a database that interprets SQL commands and inputs aren’t being sanitized.
Hacking programs like Burp Suite have functions that spam sites with all kinds of garbage data and uses error messages and delays in response times to highlight potential vulnerabilities.
Security through obscurity is still wack tho
Even if you don’t buy into this logic, you still have to do it in quite a few places because the security auditors have a line in their checklist about being able to extract any internal information from error pages
So in short it’s used to mask the developers errors?
What are you planning to do with information about the error? It’s not like these places have customer support. Usually it’s something like a caching layer failing, and there’s literally nothing you can do about that.
Edit after reading your edit:
I still don’t see why you want more information here. Those kinds of errors are almost always server side errors, and reporting more detailed information won’t help you.
You asked “why should I try again?” Answering this would almost always be unhelpful to the vast majority of users. “Try again later, because one of our cache layer hosts was down, and by the time you try again, it’ll have been taken out of the load balancer rotation, so you likely won’t hit that host on your next try.”
It would also cause more confusion with a non-insignificant portion of your users. With about 10% of users, you’ve only got 8 words before they’ll start to misunderstand you. “Something went wrong. Try again later.” That’s understandable by 100% of people according to that study.
Even saying “there’s nothing you can do about it” will probably be taken negatively by a certain portion of users. Not saying it’s incorrect, just that when you write an error message (or any microcopy for that matter), you should avoid sounding negative to the user. “Something went wrong. Try again later” conveys all the information the user needs in a way that won’t be misinterpreted.
If it’s an error code I’ve worked around before, apply same troubleshooting.
If its a new errror code, search the error code to see how other people solved it.
If no one else has solved the error code, try analogous troubleshooting, post results online with the error code name, successful or not.
I agree with Nouveau_Brunswick here.
And to add: to @[email protected] , are you not also a user of software and do you not see room for improvement in many apps? That’s where I am rn: I just want them to try harder to communicate a tiny bit more info when things go so wrong that a message has to be displayed on my screen. Telling me “There’s nothing you can do to fix the problem” would be a big help, for instance. Make sense?
These kinds of error messages are almost exclusively used for transient errors. You aren’t going to work around a transient error. The best thing you can do (the only thing you can do, really) is to try again later, hence, the message. It’s not helpful to show you a message like “cache-1234.example.com failed to respond within 300 milliseconds”. What are you going to do about that? By the time you submit a support ticket, that host has already been brought back up automatically. So now you’ve just wasted your time and the support staff’s time. The engineers already have a log of that error and a log of whatever error brought down that host, so you’re not telling them anything new by making a support ticket.
I agree that it’s (weirdly) uncommon to be the one saying “please give me more info about the error!”
A simple error code can be endlessly helpful (bonus points if there’s a corresponding support article explaining common codes)
Even if some codes are only useful to internal support, it’s handy to be able to search an error code and see “oh I can just jump straight to submitting a ticket/calling their support” or “oh, this fix might work”
Exactly, it’s especially infuriating on newpipe. WHAT went wrong? It’s an error 500 from YouTube (rare, unfixable, try again) or Google changed something and need to wait for a client fix? Or simply Google blacklisted the IP address or put some captcha that prevents playing the video??
Ringtone for when work calls/texts:
Most common place I see those is when the site doesn’t want to talk to a known VPN endpoint. Like another mentioned site owners don’t want to given any more info to what they perceive as a possible threat than needed, so they just give a generic failure page.
As a developer of many years I hate to tell you sometimes that it’s all the information we have when something breaks also. Most code is a god awful mess. Thankfully I love a good mystery.
Please apply to Crowd Control Productions. You’ll never find a better mystery than current live code that has spaghetti that traces back to 2002 or 2001. The game went live in 2003. There’s one, kinda, server. Technically there are three, but most EvE players only have access to Tranquility, most don’t have access to Singularity, and apparently they have renamed the dev server from Multiplicity to Serinity. Some of us still have access to Serinity because we were able to play test Multiplicity, back in the day.
Because the error message would be meaningless for 99% of people. Expected errors are already handled correctly, but unexpected errors like these would say something obscure like “couldn’t read property ‘count’ of undefined”. Very useful
Users ignore error messages.
I have seen my users request support, proceed to demonstrate the issue they’re having, and click through error messages so fast there isn’t even enough time for me to say “WAIT!” Forget about being able to actually read even one word of the message before it’s dismissed from the screen.
They treat the error messages like they are just an annoying mosquito to be swatted away as quickly as possible. This despite the fact that the whole reason I’m standing behind them is so I can see what it’s going wrong and, you know, read the error messages.
So the solution is to remove the error messages? That makes no sense.
I wasn’t saying it was a solution, I was just offering a possible explanation. I understand why someone wouldn’t bother telling the user what’s happening. For my code, I log everything that happens, and tell the user to call me. When I get the call, I check the log and use that to figure out what went wrong.
Distressingly typical user communication:
User: There was an error message.
Me: What did the error say?
User: I don’t know. Something about the problem.
If you remove error codes, all you’ve done is guarantee every interaction is like that.
You’ve got it backwards. Once every interaction is like that, you stop showing error codes.
I have trouble with my colleagues in IT doing the same crap. They come to me to ask me to help them with a problem, and they don’t have the errors!
That’s Windows’ training bearing fruit.
Them:
“What? Oh that? That always happens”
Me:
Grrrrr
Yeah, you’re right, but I still would prefer to see something telling me whether something I did caused the problem or something went wrong in the software / on the server. From this thread, I’m getting that my wishes will not be heard.
Never show the customer a big scary error message. In the unlikely event something goes wrong, dispatch an error event so the engineers can track the issue. They’ll pull detailed logs, and know what to do. Meanwhile, guide the customer to return later with the error screen, and collect their userId if possible. Once the issue is fixed, send them an apology email to let them know they should try again
Most companies just follow the first half of that process, unfortunately
https://knowyourmeme.com/memes/oopsie-woopsie
Verbose errors were nice
Reminds me of the old “Oopsie woopsie we made a fucky wucky!!” post
What I hate even more are error codes
Precise, easily searchable codes that describe the problem? What a ball ache!
This person gets it. Thank you.
