Forums Bug Reports $com WMI memory leak

There is a memory leak when connecting to WMI it seems.

To verify this, follow these steps, but note that you need atleast mIRC 6.16 and Windows 2k pro or higher:

1. Download $memuse
2. Load it in your remote: alt+r -> tab "Remote" together with the following code:
   
   
   Code:

   alias memleak {
     if (!$isalias(memuse)) return
     .timermemuse 5 2 echo -a $!memuse(mirc.exe,1,m).suf ** $!memuse(mirc.exe,2,m).suf
   }

   
   
   Make sure that the alias is on 4 lines as it is shown in this post.
3. If you only have 1 mIRC open, then type: //run $mircexe
4. In the original mIRC (not the one you just opened) type /memleak

Each time the identifier $memuse is called, the memory usage for the calling mIRC is increased. As far as I can tell, the code in $memuse does what it's supposed to do to prevent memory leakage, leaving me to believe that this is an mIRC issue.

Confirmed. WinXP Home + SP2, mIRC 6.16.

I guess mIRC doesn't close the objects properly, or something.

Confirmed.

XP Pro, mIRC 6.16.

Too bad this program is dead.

Let's hope it gets revived somewhere in the near future. I've been more active than I used to be lately, posting bug reports/feature suggestions, and I hope that we'll get to see a new version, or it will all have been a waste.

It would seem that way. Many seem to think that Khaled is holding out for a christmas release, but I personally hope he isn't. I honestly don't think I've ever had to wait several months for an exploit fix before.

I seem to have missed something here. How did we progress from a bug/memory leak to an "exploit"?

In addition (to earlier posts), I can't see that "mIRC is dead - it seems to be alive and well - has there been a pressing need for an upgraded version recently?

Granted that a few "minor" bug fixes wouldn't go astray - not that I have come across any myself.

Cheers,

DK

He's talking about another well known exploit, not related to this bug report. He was simply expressing his frustration that even the existance of this exploit hasn't led to a new mIRC version, while one would expect the creators of software to immediately release a patch/new version to solve the matter as soon as possible.

As I've stated before, there is little doubt in my mind that Khaled is still working on mIRC, but with equally a small amount of doubt, not as much as in the past.

Usually I would agree with the exploit thing, but this exploit is extraordinarily minor (due to the specific situation it would have to be used) and in my opinion does not warrant a hasty release. If it were something like the old DCC exploit then I would join in with the whining. I'm not saying exploits are acceptable, nor am I saying it shouldn't be fixed, but I don't see the need for the dramatics and constantly bringing it up at every opportunity.

I understand others have a differing point of view, and I'm not trying to convince others to agree with me if they don't already, so don't please start replying trying to convince me I'm wrong, it's pointless.

Now as this thread is about a memory leak, let's stay on topic

Regards,

That might be true, i,ve opened and closed Messenger.MsgrObject and tried to view its interfaces in oleview.exe and it says "server busy" whereas i was able to before i opened it in mIRC. Granted that im no COM expert so hence the "might" in that sentence.

With dead, they mean they think development has stopped.

Reasons:

* This is probably the longest period with no release, but worse, with no posts at all from Khaled on these forums. His last dates from july 2004. Since there is no communication whatsoever, some people assume that its development is dead.

*There used to be the occasional "This has been fixed for the next version" post by Khaled, which is a clear sign that development is still active, however it's been ages since such post.

Personally, I do believe mIRC's development is still going, however I don't think it's "alive and well", how could I if we haven't heard from its author in almost a year and a half. I know release dates aren't ever pre-announced, but it would be good to hear from Khaled himself for once on these forums how things are going with mIRC. I know Khaled has been busy the last year with a master, but hey, I graduated from a masters degree this year as well, and I'm doing a new one now, and I still find the time to post on these boards, so I'm sure Khaled could find a few minutes just to say hi to us and let us know how things are going.



What kind of an argument is this? As if there should only be upgrades if there is a pressing need for it. Needs are subject to user preferences, your needs will not equal mine, although it's possible of course.
I will assume you just meant a need as in "there is a serious exploit and it needs urgent fixing" but my argument remains, most releases of mIRC weren't because of a "pressing need".

Was there a pressing need for mIRC 6.16 compared with mIRC 6.15? No there wasn't, but it got released anyway. And I'm glad it did, as it introduced $comval, something I've used dozens of times, and has opened the path to a lot of COM snippets which couldn't have been made without it (atleast not without doing ugly hacks as running VBScript, and having the results written to a file = crappy/inefficient)



Sorry, but that means nothing to me. Just because you haven't come accross any bugs, doesn't mean there aren't many bugs still there, or many things that can be improved.

FO,
You must have been in too many "battles" lately.

was meant as a question, not an argument. If you read the rest of the post with that in mind it may seem clearer :-)

But I think you have answered the question anyway - there is no *pressing need* to upgrade - but obviously there are a lot of users out there who would like to see some changes.


Cheers,

DK

I suppose you're right, I had been in some kind of little battle, mainly involving someone ripping code from other users, I definitely have a different angle on it now, sorry if I came on too strong