Crashes with version 4.2.67.10

Some feedback. Since update to the recent signatures no freezes. I think the problem is solved (fingers crossed).

 

Since 5760 we've seen no issues on about 500 systems, so I'd say it's solved.

I would still prefer a more open communication between the devs and the users...

How about telling us in simple terms what the problem was... and how you've added testing for the problem so we won't see it again in future updates...

Also when you release an update and have a flood of issues, why don't you release a new version update that is just the previous version before the issues re-numbered until you can figure out what is wrong and release the update again...

It would be nice if there was an official blog/status page that the developers maintained so that IT professionals and interested end users could follow.

I would trust a company way more if they told me all about their problems up front and then told me what they did to fix them, rather than keeping everyone in the dark through private support channels.

 

Hi Wallaby,

About that error message from EAV on Outlook Express folders "DBX - internal error (5001)" :

I have got them since I upgraded from NOD32 version 2 to EAV version 4.
See:
DBX - internal error (5001)

Sometimes it happens on folder A (A is here an example), sometimes on folder B, sometimes it doesn't happen at all, sometimes it is again folder A, and so on. I have learned to completely ignore those error messages; I don't even know what those error message means: Eset was not able to tell me (or didn't want to ..... ). This is on XP-home SP3 Dutch IE6.
Anyway, I am now about moving home very soon, and hope to be able to buy a new computer soon afterwards.

 

24 hours and no crashes or lockukps. All Fixed. Let's try to avoid this problem in the future Eset. It took you 5 days to resolve. Thanks.

 

Pre-release updates re-enabled. I shall test drive these settings for as long as they work.

 

If possible, please create such a dbx file that you could provide to us perusal.

 

After update 5760 no hang ups for 24 hours (8 terminal servers)

 

No issues anymore after the 5760 update.

Are we going to get some form of compensation for this? 5 days of which 3 working days of strange problems incl. crashes, loss of internet and conflicts in programs resulting in weird faults took my team a good few hours in support, maintenance reboots and stress..

Also please tell me how you are planning to make sure this does not happen for a third time??

 

This fiasco is now being reported on BluesNews.com. Perhaps some bad press will help: http://www.bluesnews.com/s/117389/evening-safety-dance

By the way, I had to restore an Acronis image because of all the hard lockups and necessary hot resets which caused some file corruption.

 

I'm a little concerned with that myself. My laptop has been screwy since all this started, and I'm wondering if having it lock up and having to shut it off several times before I figured out what was causing it corrupted something. Good times.

 

I've been following this thread closely as I also had some issues with Eset locking up doing my everyday scan on 1/2. At that time I checked here I saw this thread but I did not know if it pertain to my issue.

But considering Eset has been extremely reliable my first thought was a a bad definition file. And sure enough the next day when I scanned again with the new definition files I've had not any more lock-ups.

 

What is written below is well stated. Eset clearly does not get it. Between this latest issue, along with too many false positives in the last couple years, extreamly pitiful performance of Nod4 on Win7 (slow surfing, etc), letting too many Antivirus scams and the like get thru, we are moving on to a new vendor. Nod2.7 was fantastic program...started going downhill after that. We are speculating internally that key developers have left the company...don't know if true, but we have had enough and have got to move on.

 

Hello,

The underlying error was due to memory heap corruption, which manifested during file scans. There were several different symptoms of this underlying error, none of which showed up in the ESET Kernel Service (filename: EKRN.EXE) crashes we analyzed. In addition to crashes, we also had some reports of computers freezing and experiencing network connectivity issues. These different symptoms were caused because the nature of the underlying error: The exact impact depended on a variety of factors, such as the structure of the processes loaded in memory, the types of files being scanned by ESET's engine and network traffic.

The randomness of these symptoms, coupled with their dependency on affected computers' environment made it difficult for this error to be discovered in the quality assurance tests that all ESET updates have to pass before being released.

A fix for the memory heap corruption was introduced in virus signature database update version 5759 and updates to later versions should be without any errors.

Additional tests have been added to our pre-release quality assurance testing of virus signature database updates for this error and engine code updated to prevent the particular error from occurring in the future.

If you do experiencing a problem with an update, please contact ESET’s technical support department immediately for assistance.

Regards,

Aryeh Goretsky

 

Does this mean another version will come out (e.g. 4.2.67.11) or is it something already updated (where?) in 4.2.67.10 in one of its modules (which?)

 

No, it have been added to the (internal) testing at ESET.
Every Signature update will go through this testing (at ESET) before it is released to the users via the update servers.

 

I am really critical of how the way ESET handled this problem..I experienced the exact same probs as most people in this thread. Before I found this forum I emailed ESET with a customer support issue. There response was the typical uninstall / reinstall the program because it was MY FAULT installing some third party software that corrupted the virus definitions..I was stunned by that response. But I did uninstall and reinstall and defintion 5760 or what ever did the trick.. I did wait a few days to see if my system is stable and it is. I do believe that ESET distributed a bad update that screwed up alot of peoples systems. And people who are not computer literate probably paid some $$$ to have their machines repaired because they did'nt understand their machines were hosed by ESET.. Last but not least, I pointed out to ESET that many of their customers are posting on their support forum board that this is a major problem.. There was no response which means ESET considers this forum a clown operation.,,,,,,,,,,,, and that is really sad. I'm back to Kaspersky.

 

Yes and no. The problem had actually been there since the mid of Dec but started manifesting at the end of the year when first 5 cases were reported and the symptoms were varying. ESET began investigating the issue more deeply and collecting memory dumps from affected users after these initial reports. However, due to the nature of the issue and the randomness of symptoms, coupled with their dependency on affected computers' environment, it was not until January 4th when a fix was released to all users.

All customer care representatives were acquainted with this issue so I guess it must have been the randomness of the symptoms that confused them and didn't think of a connection with this issue. I wonder if you could PM me your correspondence with the customer care representative so as to have more information about the symptoms that you experienced and eventually check if the response was appropriate or not.

 

Does it apply also to all the previous signatures, or do these additional tests apply only to the new signatures (>5759)?

 

You can count me as another unhappy user. I have currently two systems which are unusable as a result of your flawed updates. A Workstation laptop that has never been able to start up since last year's fiasco that you unleashed on us. A system downstairs which can't connect to the internet with the firewall enabled after your recent series of flawed updates and my workstation which has suffered several freezes requiring shutdown via the on/off switch since the beginning of the year, culminating in the anti-spam and on-demand scanner disappearing altogether in Outlook and goodness knows what other shortcoming which may exist without my knowledge.

I am just about done with this.

 

My computer has been doing the same thing since the last part of Oct. 2010. Eset needs to close and has encountered a problem. It's always been ekrn.exe. I've went through all the things to do and try but it will still crash/freeze my computer. It got to the point it was crashing 6 times or more a day and the only way to get my computer working again was turn off the power switch and restart. I even uninstalled version 4.2.67.10 and went back to version 3.x but even that version would crash my computer. Uninstalled version 4.2.67.10 and reinstalled and still same thing. When Nod32 was unininstalled, no problems, no crashes. I have used Nod32 for 3 years and have had only a couple of minor problems. The problem has been resolved, I uninstalled Nod32 and gone with another antivirus, no more problems.

 

Hello,

It applies to all virus signature database entries, effective January 4, 2011.

Regards,

Aryeh Goretsky

 

well my 4 computers are still running lickity split. Thanks Eset.

 

Hello,

Download the latest version of ESET NOD32 Antivirus (or ESET Smart Security, as applicable) as well as the ESET Uninstall Tool. Uninstall ESET NOD32 Antivirus as you normally would via the Uninstall icon off the ESET NOD32 Antivirus entry in the Start Menu, reboot, run the ESET Uninstall Tool, reboot once more and install ESET NOD32 Antivirus. You should now have a clean, working install of ESET NOD32 Antivirus on the computers in question.

Regards,

Aryeh Goretsky

 

It is a problem has concerned only to version 4.2.67.10
Or prior versions too?

 

It has nothing to do with the different version(s) I think.

The problem was due to a code that was inside an AV update.
Wich has already been solved.