Suddenly, multiple installed and clean copoes of 0.4.6 failing

[Erthome]

I went to use a 0.4.6 client (Running Windows 7, 4 GB Ram, Intel Processor) I've been using for months this morning and got the following error:


Problem signature:
Problem Event Name: APPCRASH
Application Name: minetest.exe
Application Version: 0.0.0.0
Application Timestamp: 515eb950
Fault Module Name: mingwm10.dll
Fault Module Version: 0.0.0.0
Fault Module Timestamp: 41e6d69f
Exception Code: c0000005
Exception Offset: 00001870
OS Version: 6.1.7601.2.1.0.768.3
Locale ID: 1033
Additional Information 1: 4c0d
Additional Information 2: 4c0d4d78887f76d971d5d00f1f20a433

At first I assumed that the problem was due to having moved files and improperly adjusted PATH strings in the .conf file...after an hour of piddling around with that presumed problem, I tried three other un-moved installs, THEN resorted to a clean install from the original zip. Same error everywhere. This error occurs before the client screen comes up so the problem is surely local to my machine...

I can't help but wonder, did a .dll file get placed in the windows directories that is circumventing the local copy from executing or something like that...

Meanwhile all my versions of 0.4.7 are still working as usual.

[Krock]

Then update to 0.4.8, your version is outdated.

[Erthome]

Then update to 0.4.8, your version is outdated.

There's a job waiting for you at Microsoft.

[Erthome]

Aw shucks I just can't leave it at the artful retort...

Truth is, while imbecilic opinions really grate me, I figure the bell curve assures us we have to have some, and I try to be tolerant, but when someone attempts to ram the idiotic opinions up my backside it's time to take a dump.

The 'upgrade' panacea answer is asinine. Using it, especially repeatedly, serves little apart from proving that the speaker is unknowledgable of the real problem, and strongly suggests that they are not the least bit concerned with being helpful or civil.

I'm pretty sure that my question was perfectly valid, and if the only answer one can offer is an unprovoked jab then they should simply excuse themself from the invited conversation.

The entire premise of upgrading to a new system to fix a problem that wasn't in a previously working application is irrelevant.

Further, this oh-so-popular notion of continually upgrading without clear justifications, as if upgrading itself IS a justification, is ludicrous.

The only reason this dismal thought fallacy has come to such rampant voice is because in and about 1999, during the dot com boom, a whole generation of techies that never had a real mentor in their life, started filling overpaid positions wearing vice president titles in the void spawned by the tens of thousands of dot.com boom vaporware bust scams (that then belly-upped and whipped debutante would-be entrepreneurs into submission to the ruling order of worldbank shadow government vampire dung heaps that rule the world via the so-called economy, as planned), and this same army of self-important new blood that knew so little they didn't even know what they didn't know, antithetically applied MBA misinterpretations of the object oriented model of code development, which in turn allowed for the mediocre 'memorizers' instead of the expert 'building block' engineers to dominate SDLC and market interface culture of the software solutions world, by dictating policy, practice, standards, criteria, precedent and norms through brand hype and the art of duping the funding systems.

SOOO, if anyone wants to preach upgrades to me, and they don't have to use a dictionary or go read 35 years of history in order to know what I'm talking about, by all means do, I'm all ears. Otherwise, it would be far more appropriate to create one's own post, where a bunch of people who don't know what they are talking about can go agree with each other and wallow in substandard product...but I don't appreciate it when people project their condescending patronizing attitude and push the brainless fluff cliche's at me as if they are a reasonable answer to a legitimate question--and make no mistake, I see this done a thousand times for every time I break down and call the behavior out--people have brains, and they need to take responsibility for what they do and don't do with them.

[VanessaE]

0.4.6 and 0.4.7 are both old, obsolete, and unsupported in any way. 0.4.8 is the current release, and even that has already seen some bug fixes and new features since its release. Unlike a paid company, the Minetest community really does not have the resources to keep supporting old versions of the software indefinitely (then again, our primary commercial competitor doesn't exactly do that either).

Let me be perfectly clear: 0.4.6 and 0.4.7 are both End-of-Life. Get 0.4.8.

Now, that out of the way, as you started to imply, it is entirely possible that some routine update happened in your OS that broke your ability to run regular 0.4.6 builds, in which case there is little that we can do to help you.

At this point, you have three choices as far as I can see:

1) Update to a newer version of Minetest. Since you said you can run 0.4.7 successfully, clearly there is no technological issue in doing that. *
2) Recompile 0.4.6 from sources. On Windows, this is not really easy, but doable.
3) Reinstall Windows from known, good media and get it into a state that you know is identical to that which you had in place "months ago", then try re-installing your known-good Minetest 0.4.6 binary and see if it behaves, before you apply any pending OS updates. Basically, start from scratch and walk forward in your update history until you find what broke your system. Then you will at least know what update to avoid (if that's even possible in Windows).

Obviously, the first option is the most preferable. If your problem is some bug, regression, or deficiency in 0.4.7 (or at least, post-0.4.6) then try a nightly build - and I do not mean something where "oh, I did that X months ago". I mean today. Right now. As soon as you read this sentence, go download the newest nightly development build you can find on these forums from sfan5, PilzAdam, BlockMen, or Fess and see if it works as it's supposed to.

Why? Because then the developers (and perhaps some of the rest of us) will have the exact sources to the version you're running, and therefore, they'll perhaps have a fighting chance of finding the root cause of your problem and maybe fixing it, or at least showing you a workaround. A developer can't fix that which no longer sufficiently matches what is currently used for development.

It would be like asking someone to fix a bug in a Linux 2.4-series kernel using only the current 3.2-series sources. Probably not possible to do.

Furthermore, if you expect any kind of help from this community, you must first start by losing the attitude. "Then upgrade to the latest and see if the problem persists." is a perfectly legitimate answer when the question boils down to "My old, obsolete, unsupported version of $FOO no longer works."

* If your problem with needing to have an 0.4.6 client around is that odd "one node is always unknown" glitch that happens when connecting to a pre-0.4.7 server, get after those servers' admins to update their code. There is simply no reason for those server operators to stick to old, buggy versions.

EDIT: clarified a few things here and there.

[Inocudom]

It would be easy for Erthome to upgrade if there was a way to make Minetest builds for Macintosh. He could give Linux a try, but I don't know if he would be willing to. At least he is a dedicated builder (look at what he is up to on redcrab's main server.)

There are some programs where keeping up-to-date could be an issue, but Minetest usually isn't one of those programs (some people stick with stable builds.)

Of course, community kindness is important too. I will tell everyone reading this right now that this community better not end up like Minecrafts.

[Exio4]

SOOO, if anyone wants to preach upgrades to me, and they don't have to use a dictionary or go read 35 years of history in order to know what I'm talking about, by all means do, I'm all ears.

Hey, you know everything don't you?

Otherwise, it would be far more appropriate to create one's own post, where a bunch of people who don't know what they are talking about can go agree with each other and wallow in substandard product

"upgrade to a release that works" => "i'm a dumbass that doesn't know anything"

I translated the phrase as you would do!

...but I don't appreciate it when people project their condescending patronizing attitude and push the brainless fluff cliche's at me as if they are a reasonable answer to a legitimate question

It is a proper answer because the bugs that are in 0.4.6, get fixed in 0.4.7, and so on, there aren't 0.4.6.1 releases with only bugfixes here.

--and make no mistake, I see this done a thousand times for every time I break down and call the behavior out--people have brains, and they need to take responsibility for what they do and don't do with them.

You don't make mistakes, do you? because we all know you are perfect!

bullshit.

[rubenwardy]

Looking at the report you are giving, I suggest you check your mingwm10.dll version, and maybe update or redownload.

Quick search for mingwm10.dll Exception Code: c0000005 and similar:
See this: http://niftools.sourceforge.net/forum/viewtopic.php?f=31&t=2151
(and related: http://forums.steampowered.com/forums/archive/index.php/t-2320797.html)

MinGW is a compiler, so recompiling it may solve the problem (maybe try the MSC++ build, if there is one?)

-------------

Suggesting you update is not stupid; it's common sense.

The bug you are talking about may be fixed by 0.4.8

If you want to use an old version, don't complain about bugs.

The entire premise of upgrading to a new system to fix a problem that wasn't in a previously working application is irrelevant.

It is not a new system, and you can not fix bugs unless you update.

Further, this oh-so-popular notion of continually upgrading without clear justifications, as if upgrading itself IS a justification, is ludicrous.

After reading your entire post, I can only conclude that
A) You are trolling
B) You are currently
---> drunk
---> stoned
---> confused
C) #rant

[RealBadAngel]

were aslo consulting that error report with uncle Google.
looks like this is MinGw issue...

and the sidenote: bitching at us, who are doin it all for free, wont help you in any case

[Inocudom]

were aslo consulting that error report with uncle Google.
looks like this is MinGw issue...

and the sidenote: bitching at us, who are doin it all for free, wont help you in any case

I feel that Erthome might be more interested in 0.4.8 if he knows about the shaders you did for it. They add improved bump mapping, parallax occlusion mapping, a breezy plant effect, and a wavy liquids effect.

Some of the recent bug fixes in the current version of Minetest do involve compiling errors.

[stormchaser3000]

were aslo consulting that error report with uncle Google.
looks like this is MinGw issue...

and the sidenote: bitching at us, who are doin it all for free, wont help you in any case

I feel that Erthome might be more interested in 0.4.8 if he knows about the shaders you did for it. They add improved bump mapping, parallax occlusion mapping, a breezy plant effect, and a wavy liquids effect.

Some of the recent bug fixes in the current version of Minetest do involve compiling errors.

yeah switch to 0.4.8 but only use the old versions to go on servers like redcrab.suret.net. and yes the shaders are awesome (i can't use them because my computer is complete crud (my good computer broke))

oh and you can compile minetest yourself to use if the 0.4.6 builds aren't working (sometimes the download messes up to my knolage)

[Erthome]

First, let me be clear--I didn't ask anyone to "solve" or "fix" MY problem, I voiced a question in hopes that someone was already familiar with, or on the lookout for the particular problem I encountered, or that someone who was already quite familiar with the code would go "oops...I overlooked that...thanks for catching the fluke because it just might have carried forward into the new version as well."

For the record, I use the 0.4.6 along side of as many versions that appear to be in use as I lay hands on because I am TESTING and backward compatibility is important to me. I am guessing that the installed base of 0.4.6 is considerably larger than either 0.4.7 or 0.4.8, so I don't want to dismiss all the previously loyal users from the equation just yet.

RBA I didn't BITCH at anyone with the possible exception of the person who wasn't qualified to address my question and who thus resorted to an insulting echo of "upgrade"...and even there I tried very hard to address the logical problem rather than redress the person... (unlike almost everyone who weighed in felt inspired to do to me--that's okay, seems I asked for it, just saying, ye ole double standard seems to be in play once more.)

I did, however, bitch about a rampant ill-conceived misconception that the world revolves around continual upgrades, and I offered some toes in the water backup for what I allude to... Sadly, people today take disagreeing with their erroneous or overstated "positions" and "assumptions" as a personal assault and feel that, in turn, justifies personal attacks on those who disagree with them...no matter how the individual's relative qualifications or substance of assertions stack up.

In my letter I clearly identified that I assumed the problem was with my machine since multiple unaltered instances of the code that had been working for months were suddenly failing... While I didn't say stupid, I did say imbecilic, alluding to the matter of fact insistence that I upgrade rather than ask a question about the older version--and, howsoever insensitive that might seem, it was and is accurate given the context. I stated in my original post that I was successfully using a number of newer versions. I've been watching both the forum and the IRC logs and it appeared to me that 0.4.8 was in the midst of having critical problems so I did not jump on that version immediately, but it's on the todo list and I have downloaded several gits and fired them up for preview sake.

When I have been the developer of software I most certainly wanted to know about this sort of 'environmental' behavior because these kinds of problems are the ones that fall between the cracks in planning, unit, stress and even regression testing and allow other vendors (eg. Microsoft, Apple, the Lords of ?nix, the Googly cloud giants, go go go Mozilla, et ad nauseum) to break and/or make YOUR software look bad, and it often reflects fatally on the product (particularly to new users) if installed apps suddenly fail...Moreover, to the extent that the new code uses the old, sleeper issues in older code can often offer very good (easier to isolate) clues to complex (particularly multiple-cause) problems in more robust descendent code. I made no blustery demands and had no heavy expectations--I just asked a question in the forum designated for such questions.

I have little doubt that the problem is indeed a version conflict with the identified DLL (maybe others), and the various auto update programs WindowsOS/Browser/etc, are all problematic no matter how good of job one does to try and turn and keep it all off...

I confess I did not expect to get the response folks have offered up, but with 20/20 hindsight, that was my bad--given the truth and pervasiveness of the industry ails about which I shared my criticism with some gusto, I should have clearly expected most everyone to be highly defensive of their investments and hard won comfort in those same prejudices under fire.

so @ Ruben, thank you for digging a bit...I'll have a look there... and, Yes, sorry, but definitely a bit of C) #rant...which, honestly, I believe I have earned the right to do once in a while having spent 50 of the past 35 years in developing software and wearing every hat in the industry for well over 200 companies from mom and pop to Fortune 10, and having patiently and politely listened to tens of thousands of iterations of people who don't know any better in the shadows of people who should know better spouting the same handful of Bernaysian marketing-bogged-to-insidious thought fallacy hype.

Nothing I said was meant to criticize the game, or to seem less than extremely grateful for the efforts expended by numerous people to keep it alive and well, and infused with a steady dose of shiny and new. And going further, I have no ill intent toward folks who worship the very ills of which I speak...we are what we have experienced and what we have done in life so far...but as I said, every once in a while when the button gets pushed at me directly I have my say.

@ Exio New User --hmm, well I guess thank you for letting it hang out and saying what other more civilized folks were feeling but had the good graces not to spew with your pointed venom. Apart from that I can't see where you said anything that you actually wanted an answer to, or that warrants one.

@ Vanessa -- I love you and I enthusiastically admire and support all that I see you doing...I hope you believe me when I say I welcome your reasoned and heart felt criticisms and ideas in general...but please bear in mind that 'my attitude' is a vital part of what separates me from all those who roll over and die in the workplace or marketplace or techno space, becoming slaves to the Marketeers and Strumpet master shadow governments, long before they reach their fifties...deep down you really wouldn't want me to stop being who I am. That said, I may refrain from occupying the pulpit with quite so exuberant of a retaliatory air--hopefully that will help the overall tone :-)

@ Inocudom -- bless you you wonderful feature-hungry peace maker

[Erthome]

As a minor followup, for the benefit of others who might stumble into the same problem...while I have not completely isolated the problem, I can say that rebooting the OS then NOT opening Firefox on Facebook or the Skyrim game caused all of the previously installed 0.4.6 clients to resume functioning as usual.

I will check further to see if I can isolate the cause, and if I'm successful I will report back here.

[Inocudom]

So you were trying out and previewing 0.4.8, Erthome. With your experiences in the field of software development, you would make for a good core developer. You must have more experience in programming than most others in this community.

The topic below might interest you, Erthome:
https://forum.minetest.net/viewtopic.php?id=8031

[Erthome]

I coded hard-core full-time for about ten years, mostly assembly language, micro-coding, and an untold number of dialects of "everybody's BASIC" pre-1988...at that point migrated to UNIX scripting, SQL, and started into ANSI C at which point I decided I wasn't going to learn any more 6000+ word languages to accomplish what could already be better done with a couple dozen key words and logical, consistently working syntax in machine code. SO I started doing systems and solutions analysis, design, documentation, testing, QC, tech-support (that one sucked...poor people) and then development and integration team and project management. Eventually I moved into systems and process auditing with a focus on Infosec.

Obviously, software development in the business, industry, government, defense and technical research realms is about much more than the coding itself. My coding, in C++, JAVA, Perl, Python, etc... particularly when it comes to supporting other author's code, is such that I wouldn't hire me to do it LOL.

I confess I have pushed myself, largely against my will and out of necessity, to learn a fair amount of php, because the LAMP platform, where the other open source products I've put to work on recent years seem to live, rules the roost (Wordpress, Drupal, ELGG, Moodle, SMF, Joomla, Zencart/OS Commerce, MediaWiki, being the main ones I've worked with and customized for my own use.)