Avast WEBforum
Consumer Products => Avast Free Antivirus / Premium Security (legacy Pro Antivirus, Internet Security, Premier) => Topic started by: MarshallO on April 02, 2008, 12:59:41 AM
-
I've spent all day trying to get Avast! to work properly again since updating to 4.8.1169 (the prior version worked just fine!), but it seems to be no use! To test it out, I did two "Quick Scans", which each took over 23 minutes to accomplish (which was announced at the end of each of the scans), yet, when returning to the Avast! main screen, Avast! claimed that they had each only taken about 14 minutes to accomplish!
Something is hideously wrong with this latest Update! The scans are vastly slower than they've ever been for me, and the after-the-fact reporting of scan times don't represent the actual times taken!
Unless someone knows a way to do something about these problems, I'm switching anti-virus programs.
Thanks!
-
I don't know what avast main screen you are talking about (a screenshot of both might help us to help you). I only see one screen in use for scans (SUI) and only one reported time ?
There is only one screen, the simple user interface and the scan gets done there once you start the scan, the progress bar is there, the number of files and when concluded the time taken along with the total size of the files scanned.
What do those other figures say ?
-
There is definitely something amiss.
I just conducted the following tests on the same system.
Scanned C:\Program Files\ and all subfolders Type of scan: Quick (no archive scanning)
VPS: 080401-0
avast 4.8.1169 1.4Gb reported scanned Folders/files 1215/4537 Time 2:40
avast 4.7.1098 1.4Gb reported scanned Folders/files 1212/4540 Time 0:58
The only difference between the two tests was the version of avast installed in C:\Program Files\
The tests were repeated to check the consistency of the timings. Both tests were conducted under the same conditions (which included Standard Shield being terminated).
The times recorded are my stopwatch times.
-
Well I have noticeable increase in the Standard no archive scan, but I didn't put a stopwatch on it.
With 4.8 avast is doing a little more on the on-demand scan, the improved anti-spyware and anti-rootkit are integrated into the scan.
I see an increase in the duration of the scan from a little over 11 minutes to 15 and that is for a Standard scan without Archives. So if that is the kind of increase for me when only scanning 8GB of files, I can imagine what it might be for thorough and a lot of data being scanned (?), none of which to mention. My no speed demon, but it isn't a slouch, though some might argue ;D.
The timings are however from the SUI.
-
DavidR,
It took me 23 minutes for a "Quick Scan", but I don't know HOW to determine how many GB of files that I have. All I know is that when I click on "Computer", the C Drive is noted as having "115 GB free of 138GB"; is the difference the amount of files?
I don't know what "names" that Avast! gives to its screens, but I can explain what I mean (I don't know how to provide screenshots!):
When you first log-on to avast! there is a screen (the "main screen"?) that notes (among other things): Date of last scan: 4/1/2008 14:56 (which is what mine currently says), which is noting the date and the supposed amount of time that the last scan took
However, when I actually finished the scan that this particular note is referring to (using what might be called the "scanning screen", or some such), at the end of said scan it indicated that the scan actually took 23:36 minutes to complete!
Thus, the main screen is lying about the amount of time that the scan actually took! In addition, I have absolutely now idea what "SUI" means!
Too, like alanrf, all my test scans have taken about three times as long to do in 4.8.1169 as they did in 4.7.1098!
"There is definitely something amiss" as alanrf says! I don't know about him, but I have Windows Vista with its new Service Pack 1, and I feel that perhaps the Alwil team is not yet up to date with the new Windows, or some such.
In any event, I won't tolerate such slow scan times, or "lying" about them! Taking about three times longer to scan than it did in the last version of avast! is unacceptable, and "lying" about it is deceptive and reprehensible!
Thank you for any help you might be able to give me in this matter (none of my above anger was directed towards you, of course)!
In the meantime, I'm waiting for the Vista SP1 version of AntiVir to come-out in the next few days, unless avast! gets fixed soon. . . .
-
When the scan completes the scan progress is replaced by the total size of the files scanned, that is where I get ming from.
14:56 is the time the scan was run and not the duration, see image.
SUI = Simple User Interface, the on-demand scanning control (Home version).
-
A little perspective and forbearance goes a long way. If this is the first time a product upgrade has given you a few problems then you are very young or have been well sheltered.
We here are doing our best to investigate and confirm the problems you have reported.
We know that the avast team are a dedicated bunch of professionals who want their product to be the best.
No I am not using Vista but XP. When it comes to scanning disk drives there is precious little difference between Vista and XP.
I am glad that I have been able to confirm your problem - I think that the avast team will be taking a long look at it.
-
Thanks, guys!
David: AHA! The time the scan occurred! I never thought of that! Sorry. Thanks for the definition of SUI, and the tip regarding total file size.
Alan: You note that you are just an "Avast user" under your name, but your response seems to indicate that you are actually a member of the avast! team! I am definitely neither "young" nor "very well sheltered", and I certainly have had my fair share of software problems. My main problem is that I have an "anxiety disorder" that I've suffered with my entire life, which makes it very hard for me to deal with problems I don't fully understand (as you can tell, I am only "semi"-computer literate)!
I'll give Alwil a chance to see if they can fix things, and am sorry if I got a bit "crazed" over all of this (that's all part of my disorder)!
Thanks again, and I hope you understand. . . .
-
We are well used to the frustrations that folks can feel at times like this. It is also a pretty hectic time for those of us who try to help out here.
If you read some of my other comments in recent posts I think it will be clear that I am not a member of the avast team and I am too sometimes quite open in my comments about the product.
I am one used to being a senior IT manager in charge of very large developments and worldwide systems. That background and my time in this forum has brought me to a high level of respect for the avast team.
I hope that the team will quickly find resolutions to the problems you are encountering.
-
The majority of those helping on these forums are 'just' avast users trying to help other avast users get the best out of avast. Though the forums are well supported by avast developers.
The avast evangelist reflects the total posts/contribution you have made to the forums and not that you work for avast, it is actually Alwil software (the Alwil team) who are the developers of avast, so that is one of the true indications of who works for Alwil software. The other indication is that most are also the Moderators, etc.
-
Do I take it, then, that the Alwil Team reads these Forums, and that way they find-out about the problems we're having? Or, does someone actually have to post a "bug" message to them (like I have to Firefox via Bugzilla)?
The Anxious One.
-
This is the best forum I know for direct participation by the developers/managers of a product.
And is very unlike the regular support forums of say Firefox and Thunderbird where the developers almost never participate.
-
Good! I'd hate to find the problems going unheeded by their developers, which is certainly the case in many other Forums I've been in!
Now, it's just "wait and see" (I say, while dissolving into a puddle of exposed nerve endings). . . .
Thanks once again!
-
OK, so what we have here...
You're right that the duration indicator in the Simple user interface can actually "lie" in this release. This is caused by the fact that if you do a local disk scan (Normal or Thorough sensitivity), it actually does two things now
1. a rootkit scan
2. a local disk (AV) scan
Now, only the local disk (AV) scan is actually counted into the duration value. This is an unintentional omission (or 'bug' if you want) and will be fixed. The inclusion of the rootkit scan would also explain the increased duration you're now seeing.
However, the rootkit scan should not take place if you're performing a "Quick" scan.
Also, the rootkit scan does not take place if you run the scan via ashQuick.exe (i.e. the avast Explorer context menu extension).
So if you're saying that these scans now take much longer than they used to in v4.7, there's something else going on here. It may be the new unpackers, but frankly, I doubt that as there were not too many new ones in this release.
A question: If you watch the scan progress (the currently scanned path), does it halt for some time on some specific items? Or is it just generally slower (on all files)?
Thanks
Vlk
-
See my post here: http://forum.avast.com/index.php?topic=34307.msg287614#msg287614
-
Well currently the rootkit scan fails for me on boot under the FAT32 format and I see corresponding errors in the event viewer (other topic).
No I can only assume that it would also fail in the on-demand scan too and having checked I have two corresponding errors for both of these on-demand scans.
So would there be any delay (hang, etc.) failure or would it just get on with the local disk scan ?
If it just gets on with the next task, the on-demand scan, then my scan duration has indeed increased greatly, approaching 40% allowing a little time for the failed rootkit scan attempt. See my detailed post, http://forum.avast.com/index.php?topic=34307.msg287608#msg287608 (http://forum.avast.com/index.php?topic=34307.msg287608#msg287608).
I generally don't hang around to watch I fire and forget and go get a coffee and my scan is only 16min 11sec much less than the figures quoted in the other topic.
-
I haven't noticed which files it does it on, yet, but particularly the "Quick" scan holds-up for minutes at a time (staying at say, 15% or 26%, or whatever) for 4-5 minutes or so, before the percentage starts to rise again! It happens as well on the other scans, but not as noticeably, since they're "supposed" to be slower.
I'm not a computer "geek", but all I know is that something is definitely wrong somewhere in this new update! In 4.7, my scans were much, much faster! I hope that the Alwil Team can find and fix the problem soon.
-
The percentage is not what you should be looking at - it's not possible to do perfectly smooth. The main question is - does the number of scanned files increase or not?
If not, then I'd suggest to go into program settings, enable creation of the Report file and let everything, even "OK files", be included in the scan. Then, run another quick scan - and when the program "stops", interrupt the scan. The end of the report file should indicate where the scan was when you stopped it... i.e. the "troublesome" area (if there's any).
-
That is not exactly helpful advice Igor.
I have set the reporting in the way you recommend.
Here is a scan of one of my dirves - that I have already reported above.
All it shows it that it took the ashsimple scanner 1:48 for a Quick scan to do what ashquick.exe does in 3 seconds.
* avast! Report
* This file is generated automatically
*
* Task 'Simple user interface' used
* Started on Thursday, April 03, 2008 14:53:46
* VPS: 080402-0, 04/02/2008
*
Infected files: 0
Total files: 0
Total folders: 48
Total size: 0.0 B
*
* Task stopped: Thursday, April 03, 2008 14:55:34
* Run-time was 1 minute(s), 48 second(s)
How much more evidence of a problem do you want? I feel like we users are doing the work that you folks missed in regression testing.
-
The number of files scanned does continue to increase, but at a much, much slower rate than in 4.7!
I'm not "computer savvy" enough to know how to do the "Report file" things you mention, or even why I should be doing it! Any further explanation would be appreciated!
Not to "beat a dead horse", but I have made no "behind the scenes" changes from 4.7 (because I don't know how!), yet 4.8 is multiple times slower in doing all the scans!
Please forgive my lack of computer knowledge! All I know is that everything worked find with my avast! up until this new upgrade, and now it's "freaking me out" with these hideously slow scans!
-
alan, well, what can I say... I don't see any such effects here. Quick scan - 20s, Standard scan - 40s.
However, in my post I was mostly refering to that percentage staying for 5 minutes at the same value - it might mean some problem with a particular file, if the files are not incrementing... or maybe the scanner is inside of a big archive (if the archives were enabled)...
-
Igor:
As I am not a computer "geek", it is very upsetting to see that you claim you are getting "no such effects here"! That would seem to indicate to me that you feel that nothing is wrong with the new avast! update, and that the Team therefore intends to not do anything about what seems to be plaguing a goodly number of us!
If that's the case, and you actually intend to do nothing to try to fix this problem, please let me know, and I'll switch to another virus-scanning program ASAP.
I don't want to wait several times as long for my scans to complete as they did in 4.7. It's unfair to me (and all the rest of us suffering this problem), and a waste of computer time.
Thanks!
-
If that's the case, and you actually intend to do nothing to try to fix this problem, please let me know, and I'll switch to another virus-scanning program ASAP.
Igor never said he would do nothing. Alwil always try to make avast better, so do us, users.
MarshallO, look, the forum atmosphere is not for blaming. Igor is not running away, just saying they need time to implement *any* deep change into avast 4.8.
In my opinion, on-demand scanning speed is in a very low priority. The on-access resources usage is the central point of an antivirus.
-
No need to get emotional here.
All Igor is saying is that the problem is not as obvious as you may think. That is, it does not replicate on the test machines he's tried it on. That doesn't mean it's not there, it's just that it's going to be a bit harder to track down (and fix) the problem as it doesn't take place on all machines.
Thanks
Vlk
-
Well, would you prefer me to lie and say that I am able to reproduce it here?
I am not saying anything right now - but I can't just do some magic to fix the problem if I don't know where the problem is. So, I'm trying to get some clues, maybe find out differences in some settings on your computer and ours... and if we'll be able to simulate the problem, and if the problem is indeed in avast!, then we'd certainly fix it.
-
Tech and Vik:
You're both probably right; I just hope that there are no problems behind the scenes with the "on access" scanning, too.
As I've mentioned before, I suffer from an anxiety disorder, which makes things always seem worse to me!
I hang my head in shame. . . .
I was not trying to "blame" anybody!
MARSHALL
-
I Ran the quick scan again and sat and watched it (yawn) and their were distinctly noticable pauses in folders.
c:\Windows\INF
c:\Windows\system32\drivers
c:\Windows\system32\dllcache
c:\Windows\system32 big pause, big folder.
c:\Windows\servicepackfiles\i386
Slow progress through, c:\Program Files\ 675MB, 4308 files 410 folders.
I have some of my programs not on c: but on d:\Program Files\ 582MB 5989 files, 456 folders and there doesn't seem to be anywhere near the same delay as c:\Program Files\.
I can watch it pause slightly on virtually every folder, many enough to quickly write it down, in the past these folders/files would wizz through without pausing and it was hard to record what would be being scanned.
One that surprised me was d:\program files\winfaxpro there was quite a pause and that is only 47.6MB 959 files in 6 folders.
Could there be any impact on the scan duration because of the scan progress ?
Quick Scan, no archives - confirm no aswar scan took place, no error coinciding with the scan start time.
Total Files/Folders scanned 39,657/3871
Scan duration 16min 03sec
Total scanned file size 7.3GB
My exclusions:
G:\drive-images\* my back-up hard disk images 2GB file sizes, etc.
D:\avast-excludes\* samples and tools that would otherwise be detected by avast.
?:\pagefile.sys I have my page file split over two HDDs, 128MB fixed on c: and 896MB min to 1536MB max on f:
So there you have it, personally like Tech I'm not to concerned about on-demand speed, I'm not normally sat waiting, but I would like to get to the bottom of for others reporting this as a problem to them. Previously program updates resulted in a slight speed increase or neutral/no impact at all. Mine are 40% slower than previously and that is on the Quick scan which has no additional rootkit scan (even though mine would fail {fat32})
-
Just out of my curiosity, could you try actually removing the exclusions for a while?
Alan also said that he's got some exclusions defined.
Maybe the problem is related to exclusion mask matching after all?
BTW MarshallO, do you also have some exclusions defined?
Thanks
Vlk
-
OK I will give it a try after I log off.
-
I think that I have been cooperative in testing before.
What we need here is a cooperative effort with avast that goes beyond "we cannot reproduce" to a "how can we diagnose the problem. And you know that I will be here all hours.
Let's get to it.
-
I assume that you can also take a large folder full of thousands of say jpg files, exclude jpg and then compare an ashquick scan with a ashsimple quick scan.
What differences?
-
Update, messed up went to get a drink at about 50% complete, still on c: drive and forgot about my avast-excludes folder with tools and when I came back up the alert pop-up was waiting for me, so that will screw with the scan duration.
The strange thing is after dealing with the alerts, the scan percentage seemed to have reset as when I next noticed it it was at 15%
I didn't notice any real difference in the speed the scan progressed through the folders previously mentioned or any of the comments I made.
Total Files/Folders scanned 39,692/3874 an increase of 35 files in 3 folders
Scan duration reported in SUI 19:53 my total 25 minutes.
Total scanned file size 29.3GB an increase of 20.9 of drive images, 1GB pagefile.sys and 1.4MB
It didn't seem to take that long over the .v21 drive image files considering they average close to 2GB each and 20.9 GB worth. So I don't think that removing the exclusions had any positive impact. But because of the additional large increase in scanned size and my getting a drink, the result isn't quantifiable. Perhaps tomorrow.
-
OK, Ran it again directly after boot no exclusions, no archives:
files and folders 39683/3872
Run Time 19:30
Total scanned 28.3GB
Ran again with the following exclusions:
G:\drive-images\* my back-up hard disk images 2GB file sizes, etc.
D:\avast-excludes\* samples and tools that would otherwise be detected by avast.
This time I left out this mask ?:\pagefile.sys, even so it does seem that they were excluded by default.
files and folders 39647/3866
Run Time 15:48
Total scanned 7.3GB
So given the fact that the large drive image files weren't scanned the scan run time wasn't much different a little under 4 minutes, so I don't think the masks made that much difference.
Last night I did a context ashquick scan of my Drive Image folder 13 files, 20.9GB and after 10 minutes it had only done 4 of the files, so I gave up and went to bed. So what gives between the quick scan taking less than 4 minutes to scan all those exclusions when ashquick was still going after 10 minutes for 4 files and still 9 files left to go. Was my assumption previously that the Quick scan skirts over these .v21 files ?
I think I have more than done my part to try and find out what is going on. Now it really is time for Alwil to look at what changes were made and why it might have an impact. Even if this means running 4.7 and 4.8 scans against the same data as Alan did, which clearly shows there is a big difference in scan duration.
I haven't had to do anything to replicate the problem, it was just there.
-
Tech and Vik:
You're both probably right; I just hope that there are no problems behind the scenes with the "on access" scanning, too.
As I've mentioned before, I suffer from an anxiety disorder, which makes things always seem worse to me!
I hang my head in shame. . . .
I was not trying to "blame" anybody!
MARSHALL
I would recommend that you just stop blaming your anxiety disorder and start acting a bit more professionally. If it is that much of a problem, write your posts and then read through them again before you post them. That will give you some time to edit them before you post. The users and the Avast! employees on this forum are trying to help you, but you come across as being impatient and antagonistic. The people here are wonderful. The support you'll receive from the people at Alwil (Avast) is better than anything you'll find elsewhere. Just relax and let them help you solve this problem. I'm sure you'll be pleased with the support you receive. You're using a user support forum and the software developers are participating in helping you to solve this problem. That is not very common and we should all be grateful that Alwil has such a strong commitment to quality customer support.
-
While I agree with all the comments about the great support afforded by this forum let's no go overboard on MarshallO.
We've seen far worse frustration vented here from time to time - probably even by me.
Sometimes we forget that the avast folks are trying to juggle a number of issues at once at a time like this and it can appear that the problem we care about is not getting the attention we feel it deserves.
In this case I know that the developers are working the problem - but as Vlk mentioned earlier - it is a matter of perspective. A scan taking a few minutes longer is not really as important as the problems of those folks who are experiencing more serious issues like an inability to start their system/hangs/crashes etc.
-
I just ran a quick scan with no archives and it seemed to take longer than I would have expected. It took 40 minutes to scan 100 GB (132,000 files). I would have thought that a quick scan with no archives would have been much faster than that. I know this isn't diagnostic because I didn't ever time scans with 4.7. This actually doesn't seem to be a huge problem. Honestly, how often do people do full scans and sit there and wait for them to finish? This doesn't really seem to affect on-access scanning, webshield, or any of the other "immediate" scanners.
-
If however you find that you've gone away to have dinner and last time when you came back the scan was finished and this time it still tuns for another half hour you might be a bit put out - and when a new release comes out people do tend to look a bit more closely.
The time taken by avast scans comes up frequently enough in this forum so it would be incorrect to say "who cares how long it takes". Apart from anything else the avast team ought to care - and they do.
Just look through the forum this week and count up the number of .... if this isn't fixed in the next 30 seconds I'm going to another antivirus adn I will tell all my friends to stop using avast .... posts.
-
Could somebody please confirm (or deny) that the slowdown of the quick scan occurs only when "All local disks" are selected to be scanned - while if you go to folder selection, and select all your local disks from there, it doesn't occur?
-
Well, I didn't mean it quite that way. If scans are taking much longer than they used to then there is a problem. I'm just saying that we shouldn't overstate the problem. It's not as if this problem is breaking any functionality. Hopefully, it's a minor fixable issue. As long as all of the other on-access stuff is working properly, your PC is protected. It is an annoyance to have slow scans, but it's temporary and fixable...at least I hope that's the case. :)
-
The slowdown is very evident whether it is all local disks or folder selection.
-
What we need here is a cooperative effort with avast that goes beyond "we cannot reproduce" to a "how can we diagnose the problem. And you know that I will be here all hours. Let's get to it.
I feel uncomfortable trying to help when I'm not experiencing trouble.
In my tests, for some reason, Pro (Task) scanning seems to be faster than Home (Folder selection), indeed seems slow... but I wish a benchmark test could be defined: do this, this way, check the time... etc., etc.
The slowdown is very evident whether it is all local disks or folder selection.
What do you mean? Folder selection seems slower in my case...
-
Could somebody please confirm (or deny) that the slowdown of the quick scan occurs only when "All local disks" are selected to be scanned - while if you go to folder selection, and select all your local disks from there, it doesn't occur?
I'll do another quick scan by manually selecting the same disks as before and I'll see if the scan time is any different.
-
Igor,
the very slow Filemon timings scan I sent to Vlk was on a folder selection.
-
Okay, this is very interesting. I just ran another quick scan with no archives but I manually selected my drives, the same two drives that were scanned before. The first time I did the scan, over 132,000 files were scanned and it took over 40 minutes. This time, only 31,000 files were scanned and it took 20 minutes.
Hmm....
-
Could somebody please confirm (or deny) that the slowdown of the quick scan occurs only when "All local disks" are selected to be scanned - while if you go to folder selection, and select all your local disks from there, it doesn't occur?
Well this raises more questions than answers.
Previous Quick scan, no archives, local disks,
files and folders 39647/3866
Run Time 15:48
Total scanned 7.3GB
Folder selection, C, D, F, G (what local disks would scan), sensitivity = Quick , no archives.
files and folders 18503/3867
Run Time 11:12
Total scanned 4.6GB
Why the huge disparity in files scanned, when the two scans have the same sensitivity, no archives and drives selected ?
So unfortunately this can't be used to compare, but perhaps it can be used to try and find why the local disk Quick and Standard scans are almost identical. Are they perhaps the same processing when they shouldn't be.
Interestingly the run time of 11:12 is almost identical for 4.7 Standard scan, no archives, local disks when the number of files scanned was 39,000+
-
Could it be that the first quick scan does a more complete scan and then subsequent quick scans are true quick scans? Maybe the method of folder selection really has nothing to do with it.
-
There should be a degree of consistency, if you run two quick scans one after the other they should return the same file/folder results (IMHO).
The same should be true of folder selection if they cover the same drives/partitions and you are using the Quick level of sensitivity and archives not selected. The files scanned should really revolve around the sensitivity and if archives are selected or not.
Otherwise I wouldn't see why Igor would suggest the test if the file/folder results were going to be so different.
Could somebody please confirm (or deny) that the slowdown of the quick scan occurs only when "All local disks" are selected to be scanned - while if you go to folder selection, and select all your local disks from there, it doesn't occur?
-
I've selected C:\Program files and C:\Windows to be scanned thoroughly.
It's taking an eternity to finish with Simple User Interface (4 hours for 50% of the files...).
-
I haven't got the patience for a thorough scan and I have much less data on my system I believe.
-
Well, I found a possible problem related to the progress display, where a "quick scan" scans more thing than it used to previously. It occurs only when scanning through "Local disks" area (simply speaking).
I have also been able to simulate a slowdown on one old virtual machine... but I have no idea what's the problem yet. It's gonna be kinda tricky, I'm afraid.
-
I've selected C:\Program files and C:\Windows to be scanned thoroughly.
It's taking an eternity to finish with Simple User Interface (4 hours for 50% of the files...).
5 hours and 56%...
-
I just ran a standard scan with no archives and it was exactly the same amount of time (and total file size) as when I ran the quick scan for the first time. I think the program is doing a standard scan the first time you do it even if you select quick scan. Subsequent quick scans are actual quick scans.
-
To be honest I haven't noticed any slow downs - but I have Avast set to scan when the screen saver is on, otherwise I just use ashquick with my download manager and that seems fast enough. Every fourth sunday I do a boot scan when I read the paper - that is due this week so I will check it then :D smile people
-
I've selected C:\Program files and C:\Windows to be scanned thoroughly.
It's taking an eternity to finish with Simple User Interface (4 hours for 50% of the files...).
5 hours and 56%...
10 hours and 40 minutes to finish... it's really slow! :P
-
Tech,
earlier you asked for a baseline - well there is one.
I know that the following is not news to you but I want any other readers to understand.
Remember that we have two programs in avast that drive on demand scanning.
1) ashsimpl.exe - this manages the classic scanning with options for Quick, Standard and Thorough scans and with archive scanning optional in each of those scan levels. Scanning can be performed on all local drives, selected drives and/or selected folders.
2) ashquick.exe - this manages ad-hoc scanning and its options are fixed at Thoroughl scan with archive scanning. This scanner is the one most often used to scan newly downloaded files and is the program used for the shell extension scanning when you right click on a drive, folder or file.
While both of these functions have been modified in avast 4.8 to include scan progress information to the best I can tell there has been no degradation of the performance of ashquick.exe. It appears to me that there is significant degradation of the performance of ashsimpl.exe compared with its performance in avast 4.7.
So back to that baseline.
In a system with avast 4.7 installed if you perform an asquick scan of say C:\Program Files then it will scan that folder and all the folders within it. It will run for a given time and report on the files/folders scanned the total disk space scanned and the number of infected files found. If you then use the ashsimpl.exe interface and run a Thorough scan with archive scanning of the same folder it will run for a very roughly similar time to ashquick and give statistics very close to those of ashquick.exe for scanning the same.
In a system with avast 4.8 installed, as far as I can tell, an ashquick.exe scan of say c:\Program Files will run for about the same time as it took in the 4.7 version. This is the baseline - this is what avast's scanning code can achieve in avast 4.8.
If you then run an equivalent ashsimpl.exe scan of C:\Program Files and select Thorough scan with archive scanning (to match what ashquick.exe does) then you will see the considerable degradation of the performance of this scan in avast 4.8 compared with avast 4.7. I believe that all levels of the scanning by ashsimpl.exe are considerably less efficient than they were in avast 4.7 (irrespective of whether all local drives or selected folders are scanned).
Now my guesses ...
I do not think that there is any problem with the scanning of individual files being slower. I do not think that this is a rootkit scanning overhead problem (look how fast the standalone rootkit scanner is). It does appear from traces that something is taking up time between scanning of files - that may be something to do with the way the scan progress indicator has been implemented in ashsimpl.exe - the scan progress indicator in ashquick.exe does not seem to involve any delays between the processing of files.
Of course my guesses are irrelevant. What matters is the team getting in any finding the real cause.
Earlier tonight my full system scan using ashquick (as I do every week) completed in 34 minutes, the same time as it took with avast 4.7. I then ran an ashsimpl scan at Thorough with archive scanning (just as ashquick had done). I cancelled it after it had been running more than 2.5 hours and it said that it was 39% complete.
Tech, you may want to run an ashquick scan of your system and see what the difference is from that 10:40:++
-
My standard scan takes twice as long. It used to be about 20 minutes or so and now its in the 40s.
-
I believe that all levels of the scanning by ashsimpl.exe are considerably less efficient than they were in avast 4.7 (irrespective of whether all local drives or selected folders are scanned).
I'm experiencing the same. What I can add is that seems the Enhanced Interface scans quickly at the same level (or even deeper) than the Simple User Interface scanning.
Tech, you may want to run an ashquick scan of your system and see what the difference is from that 10:40:++
I'll test it with:
"C:\Program Files\Alwil Software\Avast4\ashQuick.exe" "C:\Program files\*" "C:\Windows\*"
-
I'll test it with:
"C:\Program Files\Alwil Software\Avast4\ashQuick.exe" "C:\Program files\*" "C:\Windows\*"
Exactly 1h and 1 minute! Something very different than 10h+
-
Did you use the skinned Simple UI?
Does it make any difference if you use the skinless with the same settings? (and folder selection if it's a quick scan, not "local disks")
-
I always use the no-skin version of Avast.
I just tried another quick scan by selecting Local Disks, no archives. It looks to me like if you select that option for a quick scan, Avast actually does a standard scan instead.
-
Hi Folks, I'm new to Avast (this morning) and started a thorough scan (with archives) at 9.30am. I've approx 180Mb on my PC and its currently at 63% at 4.20pm. Just thought I would report for info. My CPU is an old AMD XP2500 so its not the fastest machine out there.
-
I suspect you have somewhat more than 180MB of data on your system, when the minimum RAM is usually in the 512MB-1GB, so I suspect you HDD is somewhat larger than 180MB.
If you only have one PC and you have continued to work on it, like browsing and posting this, then the scan will take much longer, even more so than the current slow scans.
-
I just did a couple of tests. It definitely appears to be a UI problem. If you select Local Hard Disks for a quick scan, you will get a standard scan instead. For some reason, selecting that option forces Avast to do a standard scan. I tried selecting quick scan before I selected local hard disks, but it didn't make a difference.
-
Did you use the skinned Simple UI?
Yes, avist skin.
Does it make any difference if you use the skinless with the same settings? (and folder selection if it's a quick scan, not "local disks")
I've used folder selection before. I'll test the skinless at the same area.
-
I'll test the skinless at the same area.
1h 11 minutes... not a significant difference... could be anything running in background.
Igor, what I'm surprised is that you can't reproduce this in your labs ::)
-
So, it is faster than the skinned Simple UI scan, isn't it?
If I understand your previous posts correctly, you had 10 hours when using the Skinned SUI, 1 hour using ashQuick and 1 hour using skinless SUI. Or were they different areas or sensitivities?
-
I'm not computer-savvy enough to know what Vik meant by whether I have "some exclusions defined", so I couldn't answer the Question. Sorry.
In addition, I couldn't reply to it, as I haven't received a singleEmail notification regarding the last three pages of replies to this Post!
Too, neiby's comment, "I would recommend that you just stop blaming your anxiety disorder and start acting a bit more professionally," was very hurtful! I've struggled with my mental illness for 55 years, and you just don't "drop it on cue", as
neiby apparently seems to think you can.
Therefore, I will leave the solution of this problem to those of you who apparently know what you are doing--and are apparently free from any mental illnesses--and will just slink away quietly.
I apologize to all of you who feel that I've somehow offended you. That certainly was not my intention. All I wanted to do was to have my on-demand scan times returned to being as speedy as they were in the last release.
MARSHALL
-
Igor, does it appear to you that by selecting Local Hard Disks for a quick scan that it is forcing a standard scan, instead? That certainly seems to be the case in my testing.
-
halfmilefinal, don't hijack this thread with issues unrelated to Simple UI slowdown, please!
-
Marshall,
I apologize for the tone of my comment. It wasn't intended to be hurtful, but my phrasing was very poor. Let's all get back to working toward fixing this problem. I'm sure we're on the right track.
-
neiby, yes, something like that - but it's not the only issue, there's some more.
-
neiby: Thank you for the apology!
I still haven't really got anything to offer here, however, as I am just not that computer savvy.
I'll keep an eye on this thread, though--and if there's any way that I can help, I will!
Thanks again!
MARSHALL
-
So, it is faster than the skinned Simple UI scan, isn't it?
If I understand your previous posts correctly, you had 10 hours when using the Skinned SUI, 1 hour using ashQuick and 1 hour using skinless SUI. Or were they different areas or sensitivities?
You've understand it correctly. I've tried to use the thorough scanning (same sensitivity) in all scannings.
-
Igor - I'm sorry but every test (of the hundreds on multiple systems) I have performed shows that the selection of "local disks" versus "folder selection" does not appear to be a relevant factor in the observed slow down.
-
Well, I believe there are actually 2 problems here.
1. When using "Local disks" for a Quick scan, it performs more like a Standard scan. However, (unlike a Thorough scan) Standard scan should not be significantly slower than the Quick one with usual file structure / names.
2. Slowdown of Simple UI in general. Right now, I think that the problem is related to the skins (slow painting on them) - so the problem should not occur in skinless SUI, Enhanced UI, ashCmd, ashQuick, ...
-
I must keep up ... I missed a whole page while sleeping.
Igor,
for me scanning the same disk (the non-ashquick scans were Quick - no archive)
ashquick:___________ 00:00:02
Simple UI - no skin ___ 00:00:23
Using avist skin ______ 00:01:44
-
I just turned off the "transparency" effect in settings and repeated the same tests:
ashquick:___________ 00:00:02
Simple UI - no skin ___ 00:00:21
Using avist skin ______ 00:00:30
-
How many / what files are there that ashQuick is able to scan it in 2 seconds?
-
I don't know if this will help anything, but I just found out about the SUI (I only thought that all we had were "skins"!), and got the following results on my latest "Quick" Scan using SUI, and only asking for a Scan of the Hard Drive:
Number of scanned files/folders: 96,822/19,877 (NOTE: I didn't check the box for any "folders" to be scanned.)
Runtime of last scan: 00:21:38 (Hardly a "Quick" scan to my way of thinking.)
Number of infected files: 0
Total size of scanned files: 18.8 GB
I don't if this information helps anybody with anything, but I thought I'd post it, since the scan time is approximately the same as it took me to do a "Quick" scan with a skin, and it would therefore seem to me that skins therefore probably have little or nothing to do with the problem.
Remember, I'm not that computer-literate, and perhaps I've contributed nothing!
Good luck finding out what's wrong!
-
Igor,
I simply used this to remove the physical file scans from the issue.
This is the same disk I have described to you before ... all of the files on it (structure provided above in this thread) are excluded.
However I do note that all those new failed registry RecentItemList searches are taking about 200 milliseconds per directory entry. When you have 20,000 directory entries that's some very expensive registry processing.
-
I might be missing some info here :)
What RecentItemList searches are you talking about? ;)
So, you excludeed all the files? What form does the exclusion have? *.* - or something else?
Thanks!
-
*.xxx
Just as in the 4.7 system running next to this one.
-
What RecentItemList searches are you talking about?
See attached screenshots of RecentItemList registry lookups by ashSimp2.exe for scanning the same folder in 4.7 versus 4.8
-
Hmm, what the hell are they doing there? Certainly not intentional... seems like a third, separate issue.
Thanks for that.
-
Hi again igor,
I have been a little puzzled by the discrepancies in the admittedly somewhat subjective reports on scan times. Some folks seem to see a large effect and some smaller and you saw practically none initially. So I began to wonder if it might not be related to variables in our setups and specifically the size of our exclusion lists. I thought to experiment a bit.
I do not know if this observation is any use to you ....
I just uninstalled 4.8 and did a fresh install and made no changes to the default settings (specifically the on demand exclusion list is empty).
If I run the SUI and scan the folder I have been talking about then the scan for a Quick scan (no archive) is 0 seconds.
If I add just one entry to the on demand exclusion list of the type *.xxx that covers 0% of the files in folder structure then the Quick scan time remains 0 seconds.
If I add just one entry to the on demand exclusion list that covers 99% of the files in folder structure then the Quick scan time jumps to 31 seconds.
If I use a skinned interface (avist) with the excusion list empty then the Quick scan is also 0 seconds and with the single (99% cover) entry in the exclusions list then the Quick scan goes to 38 seconds.
If this hypothesis holds any water then those of us that have built an on demand scan exclusion list that includes large areas of our disk surfaces might expect (currently at least) to see much longer scan times than those with no or small (or ineffective) exclusion lists and those longer scan times would be amplified if using a skinned interface.
Just for your consideration ...
Time to go restore the system again.
-
DavidR wrote
I suspect you have somewhat more than 180MB of data on your system, when the minimum RAM is usually in the 512MB-1GB, so I suspect you HDD is somewhat larger than 180MB.
If you only have one PC and you have continued to work on it, like browsing and posting this, then the scan will take much longer, even more so than the current slow scans.
Sorry should have been clearer in my post. My machine is AMD XP2500, 1gb Ram, 1 80gb and 1 250gb HD. I left the scan to run without any others process going until I became concerned and looked up this forum and posted.
Cheers
-
That explains a little more, the Thorough scan is by its nature very thorough and will scan every file on your system if you have also elected to scan archives. So a lot depends on how much data is on those drives. Even in 4.7 the thorough scan took a considerable time if you had lots of data and elected to scan archives.
Personally I don't do thorough scans (did one when avast first installed) with archives but do a Standard Scan, which brings to to a more reasonable duration even with this increase on duration many are seeing.
-
Just to add some stats to the overall response. On my system (see signature), I'm now taking about 17 minutes to do a standard scan of about 4.5 gigs, without archives, with 4.8, as compared with about 12 minutes using 4.7.