FTP – sorry not closed

Socrates

I spoke prematurely.

The FTP works with FTP + SSL.

But there is a huge problem.

Once I use it in a pbs job, it resets the profile to SFTP – which does NOT work.

As long as I don’t try to use a pbs, all is fine. But once I do, it’s all over.

You might want to fix this before the new patchbeam.

ADDITION: I thought maybe complete redoing the pbs files from scatch would help. Nope. FTP + SSL will not stick in a pbs file. It defaults to SFTP which does not work.

Mili

Reproduced this, logging it in. Thanks!

Socrates

Thanks.

Socrates

I tried running a PBS with encryption.

The progress bar went to 100%, but never said it was done – not even after 2 minutes after it reach the end of the bar.

What would cause this?

Ideas?

spwolf

send the pbs to support and they will try to reproduce the issue, i have no idea… btw - does the new version work fine with FTP in backup?

Socrates

After waiting for 2 minutes I finally stopped it — and it had worked (although the update is much slower (by at least a factor of 2) than the same transfer with FileZilla using the same encryption.

So I will try it a couple more times and if isn’t fixed, I can just use FZ.

Having the FTP is nice, but its not the main thing I like about PA. Your program already does so much. I cannot expect it to do everything well.

Thanks for trying.

spwolf

oh you meant the sftp transfer was slow? i didnt quite understand… it should be exactly the same speed, maybe there was something else going on? ssl overhead is pretty small.

Socrates

I realize I was unclear. It created the archive just fine.

And it uploaded the file (albeit, very slowly). But THAT is where the progress bar showed that the upload was 100% complete, but I never saw the “it’s finished” (sorry, don’t remember the wording) dialogue box. Finally I manually stopped it (2+ minutes later). Although there was the no notice that the PBS has finished uploading, the upload was successful.

Socrates

After two reboots, everything went smoothly.

spwolf

we have tested it ourselves in the meantime, same speed between the two… it was probably something specific with connection at that particular time.