BoClean High Memory Fingerprint [Merged Threads]

Hello fellow users,

I have been using Boclean for quite a while now, it is a good product, but i have a few questions to kevin that need to be answered*****

  1. Boclean Memory fingerprint version-4.25 18885kbs which i thought was quite reasonable, why has the fingerprint gone up quite significantly to 50000kbs on version 4.26, this increase is more than double what it was in previous version, i find this quite disturbing, i haven’t got any application using anywhere near that ammount of memory, although i have plenty of memory it is a huge increase in memory useage, i would of thought that with new versions you would want to decrease system resource useage, anyway if kevin would enlighten me on why there is so much of a increase in memory useage, that would be appreciated. Cheers BAZZA!!!

System Specs: Intel Core DUO 6300[ at ]2800ghz, 2ghz XMS Corsair 6400 Ram, Nividia 8600 gt, XP pro, 28 active Processes:
Security:Avast pro, returnil,sandboxie, comodo firewall 2.4, MVP Host, Process Gaurd 3.4 full, various other diagnostic Tools (:KWL) (:KWL) (:NRD) (:NRD) (:NRD)

Hi bazza3000

Strange that your memory fingerprint is so high :-\ Mine is :

BO426.EXE : 17.688 kB
BOCore.exe : 1.188 kB

Greetz, Red.

Hello Rednose,

Thanks for the reply, but your reply wasnt a fix, you do notice this happening quite a lot in this forum, a person logs his problem on the forum and people reply that they are not experiencing any symtoms, why say anything when you havent got the problem, useful feedback would be much better, it is probably beyond the average laymans expertise to comment anyway.
Kevin would be the one to reply, but i suppose hes busy with fixing other issues, i notice that he posted about a issue dated BOC425.XVU is corrupted [INFO]
« on: February 15, 2008, 02:09:52 AM »
This is part of the contents:

that I still don’t see any solution to other than deleting the offending “hoseware” which is causing it) … here, I’m talking about BOClean eating about 16 Meg of memory (or even more) if you use the ctrl-alt-del and reading that. As it turns out, it’s XP causing that memory assignment, not us! I redid the code QUADRUPLING the memory assignments within BOClean for experimentation, and it made no difference! Same for SHRINKING it by a factor of eight!

We once again need to expand the amount of space for BOClean in our next release owing to more and more nasties and the need to spot variants, and in redoing the code, I’ve discovered that whether we shrink or EXPAND memory allocations for what BOClean really needs in memory, it’s makes NO difference until we expand it more than 16 times what we might ever need that the numbers actually rise. I’ve had philosphical battles among our own COMODO “elite” as to these memory issues, and internal arguments over “leaks” and they are CONFIRMED not to be BOClean, but rather Windows memory management itself! Seems as though there’s a whacky design in XP and Vista where the system itself allocates memory which is NOT required based upon what it determines is “failures of hits” to “paging errors” and those come from our opening and closing of the file syste, and various internal Windows protections rather than anything in BOClean’s code … reason why I’ve been so tied up lately is trying to FIND the problems in the code I’ve written and there ARE no faults in my code after all … talk about “chamionship pud-pulling” trying to find where I went wrong and it wasn’t me … MOST frustrating indeed!

This may be the problem, but cant understand that its not consistent over all windows operating systems, not just mine, and one other thing is that 4.25 doesnt exhibit such behaviour and 4.26 does on my particular system
so kevin stated that the there isnt that much changed from 4.25 to 4.26, but there obviously is, anyway if kevin has a spare moment i would sure like a reply, i could reinstall it and have a look at the offending process, stack image and post, maybe kevin can make sense of it!! Cheers Bazza3000!! :THNK :THNK :THNK (:CLP) (:CLP)

Wish I had an answer for you, but in the end I do not have any that would make you happy. As you quoted, I’ve seen this too and on the one machine here which exhibited the behavior, it was running Symantec AV and has next to nothing in common with yours other than the NVidia card being the same … and that makes no sense at all as a potential culprit. I went back and forth with Microsoft about it and they kept suggesting clearing out the preloads, explaining that once a certain amount of memory is required, even for a microsecond, the “stash” will record that as “desired amount” reloaded every time it restarts. So ran with that football, and it made NO difference! :frowning:

Just for laughs and giggles though, shut down BOClean - restart it, note the memory it uses. It’ll be the normal “somewhere around 16 megs” that Windows allocates for it regardless of the fact that the actual memory it wants is about 4 megs for the database and another 2 megs for everything else. The only thing I noticed is that at some point after an update or two on that ONE machine, it will suddenly “double-up” but I’ve never seen it ever do any more than about 22-26 meg tops. Thereafter, it can run for months and never change. And it correlated to right after an update had occurred. Made me crazy and held up release for a couple of months! :frowning:

The ONLY thing that makes sense as to the cause is WININET, which is part of windows and completely beyond our control because you can free it and it will ignore you, refuse to take down a connection after a program has closed it and force memory to be reallocated, whereupon windows never frees the old memory even when you say “pretty please with Ballmer on top.” Heh.

Let’s see if your experience matches mine … turn off the autoupdating for a while and see if that doesn’t make the problem go away. As I said, have spent months, delayed the release for several, all in search of that pookah I only saw once on one machine. I never DID find the cause but that seems to have stopped it by never calling the WININET stuff inside of BOClean itself … but been over the code so many times and I’m doing everything absolutely to MS specifications … and believe it or not since I’ve seen this myself once, I won’t stop looking. But as to an answer, sorry … ain’t got one. :frowning:

I’m having the same “problem”.
After a few day’s it uses 56MB memory according to ProcessExplorer.
If i close it and start it again it starts with 51MB, same if i reboot.
If i do a manual update it goes up to 57MB and drops a few seconds later back to 51MB.

4.26
Vista SP1 (UAC).

bazza3000,
I am a complete newbie here but read this thread and want to offer a thought.
I only recently installed BOClean 4.26 for the first time only a couple of days ago.
Reading this thread I checked on memory usage of the two BO processes on my PC and it was very close to that reported by Rednose in post 2 above. (I am not taking sides BTW.)

You mention you have lots of RAM. Have you disabled the paging file by chance or set it to a small value? I have found that disabling this can cause windows memory allocation to be very quirky.

Ronny,
If you are having the same problem on Vista, then I don’t know(I’ve never tried Vista.)

Just a thought I had after reading this thread.

BOClean never was thrifty when it comes to used system resources, but the latest version definitely uses too much CPU time and RAM.

[attachment deleted by admin]

Well, according to your snapshot, it uses 0% CPU, that’s not what I call “too much CPU time”… :slight_smile: It uses ~16MiB RAM here on about 3 boxes I quickly checked, though I find these “wow it uses too much RAM” debates a bit silly - RAM usage is primarily a Windows problem due to stupid memory management, you can have 100% physical RAM used under Linux and the system is still perfectly responsive and not thrashing your HDD by constant totally pointless swapping when it’s not needed.

I didn’t mean % CPU, but CPU TIME. I marked it on the second screenshot. Don’t you think it uses too much CPU TIME comparing to other security software?

As for RAM usage I trimmed BOC27.EXE with Process Lasso and it dropped down to 14-16MB. I’ll see if it stays that way, maybe there is some memory leak in BOClean.

Despite you found it silly or not, debates about RAM usage are justified. BOClean is designed for Windows, and we all know what Windows do when it runs out of RAM.

My system’s got 2 GB of RAM, so RAM usage isn’t much problem for me, but I bet there are some BOClean users who have got 1GB or less of RAM in their computers and for them every MB is precious. :smiley:

[attachment deleted by admin]

Yeah, definitely… The horrible BOClean resource hog took precious 4 minutes of CPU time on your mostly idle system doing its job and scanning the running processes and protecting you in real time… Clearly it has a huge resource impact only in your imagination. Wrt the memory usage, there’s another thread about this somewhere, can’t reproduce the issue at all.

Hi fOrTy_7 :slight_smile:

I don’t see anything abnormal in your screenshots. If you still have questions I think you should contact Kevin :wink:

Greetz, Red.

I have to agree with the O.P. It does use more resources on my computer anyway but I would say that as my system only has 512mb of ram. It is not a show stopper but I believe it does anyway that’s my 2 pence worth.

On my box it uses about 22Mb for BOC427 and 1.3Mb for BOCore.
For what it does, I would agree that it uses too much ram.

10-14 Mb for BOC427 would seem more appropriate.

EDIT: Please see at end of post

I was about to post about this problem of RAM usage, but found this thread.

Boy! Please back off a bit, you guys! Just because you can’t see it, does not mean it’s not being seen by others.

I have also seen other posts about this exact problem.
https://forums.comodo.com/comodo_boclean_antimalware/boclean_high_memory_fingerprint-t22325.0.html
for one, and I remember other older ones.

I can reproduce exactly what the OP has described. I am using 4.26, but I see almost identical RAM usage (~52MB) and I am pretty sure that in the past I have seen it gradually climb over time.

I feel that a malware programme that uses more RAM than my browser is tto big.

EDIT: I have found something interesting that I cannot explain. I use a programme called FreeRam to monitor and manage RAM. I used Procexp.exp to kill BoClean.exe, which showed as using 52MB RAM. I then freed all RAM that I could. I then restarted BoClean and it reported 52MB RAM usage…BUT FreeRAm only reported about 20MB less RAM available. So for some reason it seems that BoClean shows up as using more RAM than it actually is!

Nick

  • 1 for me… I have also seen the RAM start low and climb.

Using avast, CFP, 512 RAM, P4 2.7GHz (old machine)

Weird one indeed … am going to review the code for the next day or two and see if there’s anything we’re doing, but have seen this before. Both the time AND the memory seems to indicate exactly three times what it should be. I suspect there’s something weird going on there, and has to do with the WMI/PSAPI stuff which is used by Windows to provide all that data. 4.27 was the result of a really odd bug in the PSAPI stuff where it was reporting an incorrect amount of memory used by the process/thread/dependency listings it generates and that was what caused BOClean to misbehave with Miro and other insanely “piggy” programs with hundreds of subthreads. It was lying to BOClean about how much memory it needed to get ALL the data (reporting smaller than actually needed) and of course, that caused 4.26 to crash in the presence of data which was longer than the amount required.

So I suspect that’s what’s going on here as well for reasons I don’t quite know … that what you’re seeing there isn’t the truth. Not much I can do to fix Windows, but will see if there’s something I wrote which is somehow tickling the tiger … but the internal “performance data” libraries are KNOWN to be … well … messed up. :frowning:

Will let “ya’s” know if I find something, but I don’t think I will. Seems as though you guys have something in common there which is “mirroring” reality three times for all of the calculated figures you’re looking at.

Just went over this on the other thread …

https://forums.comodo.com/comodo_boclean_antimalware/boclean_427_huge_resource_impact-t25505.0.html

Perhaps a “merge” of these two might help …

Thanks for the reply Kevin. I hope my little find helped.

I don’t envy you…admire, yes, envy no! :slight_smile:

Nick

Threads Merged. (:m*)

Hehe…so Kevin has somebody’s ear on the forums.

Sorry. I know how Web humour can fail. Kevin deserves that ear…at the very least.

Nick