Uploaded Kismet '.gps' files missing SSID's in Wigle databas

[mungewell]

Hi all.
I've uploaded a few Kismet '.gps' over the past few days, they seem to have been accepted and reported new networks found - however the database (and web maps) show a lot (most) of points with '<no ssid>'.

If I process the same files locally with gpsmap the SSID's are there, and the resultant image shows them OK.

I tried upload the '.net' files to see if these fixed the problem, however all of these were processed with no new networks found.

What's up with that?
Simon.

[mungewell]

What's up with that?

Forget it.... thought to upload XML and that's fixed it.
Simon

[bobzilla]

I pray for kismet to output a file that has all data, unfortunately to get everything you really need at least the csv and gps. xml has another cross section of info.

[DrDeke]

What is the best way to upload data gathered from Kismet then? Should I just send the .csv or .xml files? One of those and also the .gps? All three?

[mrmeebles]

GPS contains the most thorough amount of data. Sometimes XML and CSV will have file that GPS doesn't have. In my experience, my GPS always have more recorded networks that the XML.

[argh]

Assuming Linux, some variant on this will work:

Code: Select all

tar cfvz filename_to_upload.tar.gz *.xml *.csv *.gps

This way you have one file to upload, and Wigle will keep the original names straight. This will of course tar.gz all filenames in the directory with .xml, .csv and .gps suffixes, so adjust filenames and/or path to be what works for you.

[DrDeke]

Will sending all three (xml, csv, gps) not confuse the wigle system?

Alternatively, I have looked at the .gps files, and while they definitely contain the most position information, I don't see any SSIDs in them.

My goal is to provide wigle with the greatest amount of accurate data possible, and it seems like having SSIDs reported is an important part of that. However, I would hate to not upload the gps files at all, because they contain a lot of data that could be useful in triangulating the probable position of each base station...

[uhtu]

we are fairly easily confused, but do our best to get as reasonable a set of data as possible out of posted files (and we support getting data from all the many-flavored kismet files)

[jointer]

Hi,
i hope i found the right thread.
I uploaded two scans from Kismet.
Both scans this way:
1st XML file (it have ESSID)
2nd GPS file (it have more data about locations).
4 files total.
The problem is that on the map, i see some of the APs twice, separated by distance of "half tile at max zoom". (probably 10 to 20 meters)
Is this correct way how to upload or is this caused only by doing two scans in the area?

[uhtu]

if you can give a slightly less vague description, we can probably help more.
do you see single instances of the network in question on the query page, and its just a display issue on the web maps? are you sure its the same network, "linksys" is a very common name, etc.

[jointer]

It is a bit hard to describe, so i made screenshot:
http://img244.imageshack.us/img244/5116/error.gif
It is in the online map, when using last two levels of zoom, i see the APs twice, four or six times, placed in pattern.
In case i export the area to .kml and view in Google Earth or use lower zoom than last two levels, it is displayed correctly.
I found this right now, so i suppose it is not problem of multiple records in database, but map rendering issue.
I hope the upload process by using first .XML and then .GPS is fine.

[uhtu]

if you don't see the issue when zoomed out, its just a tiling issue on the map server, and not anything wrong with your posts