Revisiting the Cambridge Model 20 OLC failures...

[Westbender]

Has anyone figured out a way to submit valid igc files from these
devices? For the last few seasons, everything was working fine. This
year, about 3 out of every 4 OLC postings produce an invalid status.
The OLC folks have been nice enough to fix the log file status, but
they cannot continue to do this. What's the secret? We've tried See
You as well as the DOS command line utilities. Same results no matter
which method we use.

[Darryl Ramm]

Westbender > wrote:
> Has anyone figured out a way to submit valid igc files from these
> devices? For the last few seasons, everything was working fine. This
> year, about 3 out of every 4 OLC postings produce an invalid status.
> The OLC folks have been nice enough to fix the log file status, but
> they cannot continue to do this. What's the secret? We've tried See
> You as well as the DOS command line utilities. Same results no matter
> which method we use.
>

What exact error do you get? is it only OLC complaining or does VALI-CAM
fail as well?

Have you run VALI-CAM on every file before submission?

Darryl

[Ian Reekie]

After each flight I perform the download from the Cambridge to an Ipaq
using Connectme from Naviter. When I get home I sync the IGC file created
by Connectme to the PC and do a direct file claim to the OLC site.

It has worked every time this year, I get a green V (validated OK).

The IGC file created by Connectme has the binary .CAI file contents
appended at the end. This allows OLC validation.

I believe you cannot submit flights to the OLC directly from SeeYou
anymore.

[Paul Remde]

Hi,

How exactly are you using SeeYou? You should be able to open the .cai file
(not the .igc file) in SeeYou and then save it as a .igc file.

Best Regards,

Paul Remde
Cumulus Soaring, Inc.

"Westbender" > wrote in message
...
> Has anyone figured out a way to submit valid igc files from these
> devices? For the last few seasons, everything was working fine. This
> year, about 3 out of every 4 OLC postings produce an invalid status.
> The OLC folks have been nice enough to fix the log file status, but
> they cannot continue to do this. What's the secret? We've tried See
> You as well as the DOS command line utilities. Same results no matter
> which method we use.

[Westbender]

The problem is that we get the red "V" on the OLC site after
submitting. All the output from vali-cam shows everything to be
correct. All validation passes....except for the OLC.

I corresponded with the folks at the OLC and they recommend I use the
command line utilities (vali-cam, cai2igc, etc.). However that method
does not work 100% of the time either. They claim it's an issue with
the legacy cambridge loggers and they recommended yesterday that we
should upgrade our logger.

I personally have a 302, which works flawlessly. A fellow club member
has the model 20 that we can't seem to get valid OLC submittals out
of. I've been doing the conversions and submittals for him. The
interesting thing is for the last 4 seasons or so, the files were
accepted by the OLC 100% of the time. This problem just started this
year.

[Westbender]

Ok, I could use a little schooling on these converted cai files.
Here's the message from Juergen @ OLC regarding the problems with the
files. He sent this about 3 or 4 files ago.

"The signature in the .igc file was missing. The validation procedure
only can validate the binary .cai file. Therefore the .igc file for
the OLC consists of three parts: The plain igc records, the signature
and the embedded .cai file. The server splits this combined file and
checks the .cai file and the signature. Without that signature the
file is invalid. I extracted the .cai file from the .igc file,
converted it again with my own procedure (you can find it attached)
and replaced the file in the flight claim with this new one."


Vali-cam shows the cai file to be fine, so I'm assuming that's not the
problem. When I look at the igc file, I can see two "G" records. Are
these the "signature" records that Juergen is referring to? What other
kind of signature file is there besides what's in the original cai
file?

[Westbender]

Update...

After looking at the different files from different conversion
methods, the See You converted files are in fact missing the "G"
records. However, the DOS cmd line utilities are creating "G" records,
but the files are still invalid when submitting to the OLC. Once in a
while one will work, so this is an intermittent problem, but most of
the time we get a failure.

I have sent another email to the OLC help group with more details to
see if they can tell me precisely what is wrong with the latest files
that do have the "G" records.

[Dan Marotta]

I, too, have been having problems submitting my files recorded by a Model 20
GPS. In the past, Juergen has fixed them for me, too.

I have found through experimentation that, if I validate and submit the same
file THREE TIMES, it will fail validation on the server twice and will be
accepted the third time. Of course, I have to delete the first two
submissions after they fail and before resubmitting. I have no explanation
but it seems to work for me.

I have to ask: Why do I need to upgrade my hardware? What was done to the
server software which makes my logger no longer create valid files? This
would never be acceptable in a business setting...


"Paul Remde" > wrote in message
...
> Hi,
>
> How exactly are you using SeeYou? You should be able to open the .cai
> file (not the .igc file) in SeeYou and then save it as a .igc file.
>
> Best Regards,
>
> Paul Remde
> Cumulus Soaring, Inc.
>
> "Westbender" > wrote in message
> ...
>> Has anyone figured out a way to submit valid igc files from these
>> devices? For the last few seasons, everything was working fine. This
>> year, about 3 out of every 4 OLC postings produce an invalid status.
>> The OLC folks have been nice enough to fix the log file status, but
>> they cannot continue to do this. What's the secret? We've tried See
>> You as well as the DOS command line utilities. Same results no matter
>> which method we use.
>

[Westbender]

Latest reply from Hannes @ OLC:


"the file attached to Your mail is valid. The g-record may consist of
1 or more lines starting with a G. I really don´t understand, why the
file sometimes is found invalid and without any change it sometimes is
found valid. A mysterium of Cambridge files.

Our program is working with following steps:

1. It finds a cambridge file derived from a Cambridge 10/20/25 logger
(this file must contain also the cai file (added by means of Carl
Ekdahls program c2i4o11.exe or manually with cai2igc.exe
2. It excerpts the cai file and
3. It compares the cai file and the igc file
4. If both files are found congruent, the file will be marked as igc
valid
5. If it finds any differences it will be marked as invalid

So the first condition is, that the CAI File itself is valid, which is
checked by vali-cam.exe
The second condition is, that the CAI file is converted into a normal
IGC file with conv-cam.exe
The third condition is, that the CAI file is added to the normal IGC
file with cai2igc.exe

So far clear?

Now the program of Carl Ekdahl c2i4o11.exe still produces an igc file
which has a wrong file name, because it sets a B (for 2011) in the
beginning of the file name, instead of a 1 (for 2011). So if You use
the Ekdahl program, please change the file name from e.g. b8scxxxx.igc
to 18scxxxx.igc and this file should be igc valid for olc,"

[Roy Clark]

On Aug 30, 5:32*am, Westbender > wrote:
> Latest reply from Hannes @ OLC:
>
> "the file attached to Your mail is valid. The g-record may consist of
> 1 or more lines starting with a G. I really don´t understand, why the
> file sometimes is found invalid and without any change it sometimes is
> found valid. A mysterium of Cambridge files.
>
> Our program is working with following steps:
>
> 1. It finds a cambridge file derived from a Cambridge 10/20/25 logger
> (this file must contain also the cai file (added by means of Carl
> Ekdahls program c2i4o11.exe or manually with cai2igc.exe
> 2. It excerpts the cai file and
> 3. It compares the cai file and the igc file
> 4. If both files are found congruent, the file will be marked as igc
> valid
> 5. If it finds any differences it will be marked as invalid
>
> So the first condition is, that the CAI File itself is valid, which is
> checked by vali-cam.exe
> The second condition is, that the CAI file is converted into a normal
> IGC file with conv-cam.exe
> The third condition is, that the CAI file is added to the normal IGC
> file with cai2igc.exe
>
> So far clear?
>
> Now the program of Carl Ekdahl c2i4o11.exe still produces an igc file
> which has a wrong file name, because it sets a B (for 2011) in the
> beginning of the file name, instead of a 1 (for 2011). So if You use
> the Ekdahl program, please change the file name from e.g. b8scxxxx.igc
> to 18scxxxx.igc and this file should be igc valid for olc,"

-----Forwarded Message-----
From: OLC-Help
Sent: Jun 1, 2011 10:09 PM
To: Roy Clark
Subject: Re: OLC flight claim - Status Report

Hello Roy,

the old cai2igc4olc.exe does not work anymore since beginning 2010,
please use the attached cai4olc.vbe vor conversion of cai files into
valid igc files, I set Your flights to green manually,

regards
Hannes

[Roy Clark]

The file mentioned in my prior post is: cai4olc.zip and came be
downloaded via a link
on the Geelong Gliding Club website http://www.gliding-in-melbourne.org
thanks to
Jeff Farrow.

[Westbender]

It appears that our only option is to keep resubmitting the same file
until it validates successfully. If that truly does work, then it's
obvious that the problem is in the OLC validator software. Once I have
a track record of proof, I'll contact the OLC folks again and try to
get them to look into it. Although at this point, they don't seem to
have any enthusiasm at all for dealing with this problem. We'll see
how it goes.

[[email protected]]

On Aug 31, 10:03*am, Westbender > wrote:
> It appears that our only option is to keep resubmitting the same file
> until it validates successfully. If that truly does work, then it's
> obvious that the problem is in the OLC validator software. Once I have
> a track record of proof, I'll contact the OLC folks again and try to
> get them to look into it. Although at this point, they don't seem to
> have any enthusiasm at all for dealing with this problem. We'll see
> how it goes.

This is a crazy problem.
This is a moving target.
How can a bad file be fixed with a program?
How can a good file be bad and then fixed with a program?
If I convert the cai file to igc with the old dos program it gets
scored and put in order in light print with a red v. It all pays the
same - ($0). I for one have spent my last evening trying to get rid
of the dreaded red v.

Bill Snead

[Westbender]

Last two flight logs validated fine without issue. Not sure if
something changed with the OLC flight log validator. Hopefully it's
fixed.

[Dan Marotta]

Same here, though I've only had one. It validated on the first try (after
running the batch file).


"Westbender" > wrote in message
...
> Last two flight logs validated fine without issue. Not sure if
> something changed with the OLC flight log validator. Hopefully it's
> fixed.

[Westbender]

On Sep 12, 10:53*am, "Dan Marotta" > wrote:
> Same here, though I've only had one. *It validated on the first try (after
> running the batch file).
>
> "Westbender" > wrote in message
>
> ...
>
>
>
>
>
>
>
> > Last two flight logs validated fine without issue. Not sure if
> > something changed with the OLC flight log validator. Hopefully it's
> > fixed.

Three in a row! It's looking more and more like the problem was fixed.
Yay!

[Westbender]

On Sep 15, 11:23*pm, Westbender > wrote:
> On Sep 12, 10:53*am, "Dan Marotta" > wrote:
>
>
>
>
>
>
>
>
>
> > Same here, though I've only had one. *It validated on the first try (after
> > running the batch file).
>
> > "Westbender" > wrote in message
>
> ....
>
> > > Last two flight logs validated fine without issue. Not sure if
> > > something changed with the OLC flight log validator. Hopefully it's
> > > fixed.
>
> Three in a row! It's looking more and more like the problem was fixed.
> Yay!

Hmmm.... I see there's a couple of recent failures by other pilots
submitting logs from legacy Cambridge recorders. Looks like the
problem is not solved.

[Dan Marotta]

I submitted a log from a Model 20 yesterday and it went through perfectly
first time. Previously, I've had to submit the same log three times for it
to be accepted.


"Westbender" > wrote in message
...
On Sep 15, 11:23 pm, Westbender > wrote:
> On Sep 12, 10:53 am, "Dan Marotta" > wrote:
>
>
>
>
>
>
>
>
>
> > Same here, though I've only had one. It validated on the first try
> > (after
> > running the batch file).
>
> > "Westbender" > wrote in message
>
> ...
>
> > > Last two flight logs validated fine without issue. Not sure if
> > > something changed with the OLC flight log validator. Hopefully it's
> > > fixed.
>
> Three in a row! It's looking more and more like the problem was fixed.
> Yay!

Hmmm.... I see there's a couple of recent failures by other pilots
submitting logs from legacy Cambridge recorders. Looks like the
problem is not solved.