Uploading of completed scan DB file

[mtotheikle]

We need a way to up upload the DB file to a server so that people reviewing the data can see what we have. We will need to research how we can upload large files and also when do we upload a file?

We don't want DB files staying on the computer long. What I am thinking is that we don't even save a file on the HDD and we do a scan when the user has an internet connection, that way we can communicate to an external site over SSL the results we find and therefore not leaving a trace on the computer.

We will need to discuss this with James, this relates to showing the pattern found in the DB.

[sturakov]

Uploading the results is an interesting idea...Wouldn't we be doing it over Eastern's network anyway? So I'm thinking speed might not be too big of an issue. We'd definitely want it secured though.

My concern is the stability of the scanner and the length of time it takes to scan all the files.
If the Scanner dies or hits some case we didn't expect...what kind of contingency would we have in place to handle it?
It sounds like we might have to some way of saving the files that had been scanned already, a sort of manageable restore point.

Overall, this sounds like a feature we might want to implement when we are certain our Scanner is working at least to Beta standards...The engine recognizes well and the scan time is not glacial.

[sturakov]

We should definitely bring it up with James and hear what he has to say. He might tell us to prioritize other stuff in the mean-time.

[mtotheikle]

James has talked a little about the saving of a checkpoint and it would be very difficult to both create and use the checkpoints so we will have to discuss this with him in more detail.

I am not sure how the scanner handles exceptions besides if its on a file it will just skip over the file and insert the error. But exceptions such as memory problems I am not sure how those are handled yet and will need to look into it.

[sturakov]

I wonder what he'll have to say about how he wants the information
handled..either uploaded or scanned entirely first..and then uploaded?

You're right about the scanner skipping over. In terms of memory
exceptions, the scanner seems to quit scanning (this is based on my one
test run).
I'll have to look into the code as well.

On Sun, Apr 22, 2012 at 3:54 PM, Michael Williams <
reply@reply.github.com

wrote:

James has talked a little about the saving of a checkpoint and it would be
very difficult to both create and use the checkpoints so we will have to
discuss this with him in more detail.

I am not sure how the scanner handles exceptions besides if its on a file
it will just skip over the file and insert the error. But exceptions such
as memory problems I am not sure how those are handled yet and will need to
look into it.

---

Reply to this email directly or view it on GitHub:

#11 (comment)

[mtotheikle]

Well he was first talking about completing a full scan and then uploading the file, the problem I think is that the file could remain on the HDD if the program crashes and that file could have sensitive information.

[sturakov]

Oh yeah, that's a definite concern. We would definitely need to look into
that.

On Sun, Apr 22, 2012 at 10:19 PM, Michael Williams <
reply@reply.github.com

wrote:

Well he was first talking about completing a full scan and then uploading
the file, the problem I think is that the file could remain on the HDD if
the program crashes and that file could have sensitive information.

---

Reply to this email directly or view it on GitHub:

#11 (comment)