S63, Fingerprints and false virus checker alerts

[bcn]

In the last days we received reports from users that under Windows their anti-virus blocks the download and/or execution of the fingerprint executable identifying it as a "generic trojan".
This is a false positive. Dot.
From the sparse information we have found this false positive is triggered by the fact that the fingerprinted is generating a file when running: exactly what it has to do...

What can we do? What can you do?

- change something in the executable and compile it again.
We won't do that. First one would implicitly admit that something is wrong with the file. Secondly as the alert is most probably driven by heuristics or behavioral analysis we would have the same issue three days later again.

- We tried to contact the anti-virus manufacturer but they just ignore us as we are not customers of theirs. And we will not buy such a product just in order to be able to report a fault to them.

- So we ask you: if you have that issue, please report a "false positive" back to the manufacturer of your anti-virus program

- If you can't download the executable because it gets blocked, drop us a mail and we will send a zipped version with a different extension.

- To execute the fingerprinted you will most probably have to stop your anti-virus program for that instant.

Welcome to the Windows world :bang head:

And thank you for your patience!

Hubert & o-charts.org

[Gilletarom]

Hello Dave, Hello Hubert,

Hum ..... I see from reading your post, some anger. I can understand you.

That said, if you can not get your software is no longer analyzed as containing malware, it will be quite difficult, and even impossible, to obtain the trust of users.

Which can be accepted by beta testers, with confidence you will not be accepted by the user "alpha". Ask them not to trust their anti-virus software, is not reasonable.

And how the editors of websites, defending OpenCPN, will it themselves if they want to retain the trust of their readers?

We are facing a particularly difficult situation to manage.

It is hoped that a good solution will be found.


Best regards. Gilletarom.

[CarCode]

Gilletrom,
it is not new that anti-virus software often makes false alarms and this often with well-known software. There are several such examples on the internet. After all it is the wonderful world of Windows where even a good anti-virus software has not found malware for years. So both cases are possible.
Never trust anybody...
Gerhard

[bcn]

For me the only "solution" is the one I've posted:

- So we ask you: if you have that issue, please report a "false positive" back to the manufacturer of your anti-virus program

The rest is fighting against wind mills

Hubert

[Gilletarom]

hello Gerhard,

Originally Posted by CarCode

Gilletrom,
..... So both cases are possible.
Never trust anybody...
Gerhard

Here's a sentence in both directions .... Which will have the confidence of users ... O-Charts or anti virus software ....

It's "heads or tails"

Gilletarom.

[bcn]

So just argue why one would put a trojan on it's own web site with full address and so on...
It just does not make sense.

We are not Google Inc.

[fgd3]

Originally Posted by Gilletarom

hello Gerhard,



Here's a sentence in both directions .... Which will have the confidence of users ... O-Charts or anti virus software ....

It's "heads or tails"

Gilletarom.

I can't speak for others but I am considerably more confident in OpenCPN than many of the anti-virus programs on the market.

Fabbian

[CarCode]

Originally Posted by fgd3

I can't speak for others but I am considerably more confident in OpenCPN than many of the anti-virus programs on the market.

Fabbian

Correct, the ones works hard to get money from their users no matter which way while the OpenCPN people perhaps only wants to get back their investments.
However it is always possible some hacker injected an otherwise clean website...
Gerhard

[bcn]

Originally Posted by CarCode

However it is always possible some hacker injected an otherwise clean website...
Gerhard

That's why we added the MD5 check sum - and this one we are comparing with the file compiled on the standard Microsoft environment on a system well separated from the o-charts site where the copy for download lives.
The upload is done from a third system with a different OS.

But such is life ......

[CarCode]

Hubert,
huh, you shouldn't have mentioned Microsoft. It's the first source of all devil work...
Gerhard

[CarCode]

Addendum:
Hubert, please read here:
Zero-Day-Lücke in Windows | heise online
Gerhard

[bcn]

Well, that remembers me about a message from the same magazine some days ago:
"Several Thousands of MACs infected by iWorm"
iWorm infiziert tausende Macs | Mac & i

And Apple is not known to be very agile regarding security issues.....

If there are sufficient users, there are crooks - see Android

[rgleason]

In compiling for windows I used to have Norton 360 issue dire warnings and delete the file I had just compiled because it was "unknown and not used by many people". It was one or two files in climatology. The solution was to override Norton and reverse the deletion. Eventually I had done it so many times Norton gave up or I made an exception. One day soon I will get rid of this piggy app and get mcaffee when the subscription runs out. Others have said Norton is bad but I knoe nothing else since 1985.

Sent from my GT-N5110 using Cruisers Sailing Forum mobile app

[bcn]

seems that others - Adobe in this case - have sometimes problems with the virus checkers as well: