16

I asked a question (Can't connect to remote centos 7 install from at least one host) that has for whatever reason resolved itself after almost a day of not using the affected computer. I was going to do more troubleshooting just now but discovered that the problem went away.

I would delete this question if not for the answer that is/was not relevant.

What should be done in this case?

3 Answers 3

Reset to default
13

Firstly, thank you for asking! It's nice when people care about not cluttering up the site.

Secondly, as someone's already found it useful, I'd definitely leave the question undeleted. Your edit at the top makes it clear that, for now, you will be unable to test further suggestions because the problem has cleared itself.

The only change to Yagmoth's suggestion I'd add is to accept the existing answer. If the problem recurs, and that's not it, you can un-accept the answer at that time, and if you solve it yourself, post about that, and accept it. If it doesn't, at least the question won't float around forever like a querulous albtaross who cannot be satisfied.

Edit: it's been a week, and I notice you still haven't accepted the answer to your original question. You might want to fix that.

5
  • 1
    I don't like the idea to accept an answer that is not proven to solve the problem. If the question owner will become inactive and people report the solution is not working you end up with an answer that misleads people, because it stays on the top. In my opinion the right way is to answer the question itself and try to explain what happened, even when it was magic, then it was magic.
    – Martin Braun
    Commented Mar 14, 2016 at 9:22
  • @modiX I agree with your caution, but it's axiomatic in sysadmin that it is very difficult to fix a problem you're not having. If this problem doesn't recur, the question will remain open forever because in its current form it cannot be answered (or, more likely, it'll get closed, for the same reason). You may not be aware that an acceptance isn't forever; the OP can come back as-and-when the problem recurs, un-accept the current answer, and update the question at that time.
    – MadHatter
    Commented Mar 14, 2016 at 9:29
  • @MadHatter Sort of by definition, acceptance implied the answer is acceptable. Another person may decide not to answer based on the fallacious (but perfectly justified) assumption that the other answer solved the problem. "in its current form it cannot be answered" Exactly. So accepting an answer makes no sense; it would be the Q&A equivalent of an argument from ignorance.
    – Parthian Shot
    Commented Mar 14, 2016 at 13:31
  • @ParthianShot I agree that it seems odd, but I'll look at this from another angle: what do you think should happen? Should we close the question? Should the OP delete it? It cannot be answered in its current form, and I, for one, don't want it hanging around for ever. Of the tools we have available to us, accepting an interim answer makes most sense to me; I'm perfectly happy to be told we should do something else, but not that it should be left open as-is on the offchance the problem returns.
    – MadHatter
    Commented Mar 14, 2016 at 13:38
  • @MadHatter I suppose accepting it while explicitly noting that the accepted answer was untested would probably be the most pragmatic, reasonable response given the options and assuming leaving a question open forever is not an acceptable solution.
    – Parthian Shot
    Commented Mar 14, 2016 at 13:58
4

Keep the question open, and when the trouble will be back you might be able to answer yourselft, as I read the answer provided, and the answer seem ok, as you have no clue what your problem was anyhow (and I just learn that firewalld is by default in CentOS 7 at the same time)

-4

You did the right thing for the moment by editing in that it self-resolved, to prevent people from making futile attempts at answering it.

Personally, I'd wait a while to make sure it stays resolved... it has been my experience that problems which spontaneously resolve are also prone to spontaneously recur. If it stays spontaneously self-resolved, put that as an answer and accept it.

If you'd like, you always have the option to flag your own post for a mod to close or delete, as well, but given that you got the question-ban warning back in June, you might not want to risk it. If you want to do that anyway, just hit the In need of moderator intervention option and explain why you want your question closed or deleted. No guarantees they will, but at least right now, it doesn't look like there's much value to the existing question and answer, so there's a decent chance they'll go along with such a request.

enter image description here

5
  • 6
    "It doesn't look like there's much value to the existing question and answer" I'd be more easily convinced by that if we didn't have a decently-rep'ped user testifying - less than a foot above this comment - that (s)he got value out of it .
    – MadHatter
    Commented Mar 1, 2016 at 10:07
  • @MadHatter There's incidental information in most every question we get, even the crap ones. That doesn't give them value. For example, you can check in VM disk images to Git. That information does not impart the question with value, however, and despite the ♦-reopen-and-answer that went on there, that question is worthless and should be nuked with extreme prejudice, even though it contains information that is new to all of us.
    – HopelessN00b
    Commented Mar 9, 2016 at 7:19
  • HN, you know I'd follow you into the trenches in pursuit of bad questions, but in this case, I do think the community is making a valid point (with respect to that other question, I agree that it's worthless, and have just piled on a close vote). That said, these are judgement calls; they can lawfully be disagreed over, without the other side's point of view being prima facie dumb.
    – MadHatter
    Commented Mar 9, 2016 at 8:24
  • @MadHatter Just saying that the presence of incidental information doesn't give a question value, IMO. Not calling your position dumb or anything like that.
    – HopelessN00b
    Commented Mar 9, 2016 at 9:13
  • 1
    No, and I agree with you. In the original case here, I think there is useful information in an answer that gives the question value as a whole. In the case of the VM image one you linked, I don't think there is (not least because the answer is, as it can really only be, "your question is bad and you should feel bad").
    – MadHatter
    Commented Mar 9, 2016 at 9:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .