Issues with importing Channel Configuration File

[Amie]

We have not been able to go live yet because our songs are messed up. I have finally narrowed it down to at least one problem in S3, maybe two.

1) If I have a group of channels in the first track, when I import the channel configuration file, it sometimes skips a channel and puts the data to one channel on the channel below it. If I don't catch it, and have to import again, it will shift the information down another channel, in the same spot. (I can recreate this issue in a test file, fwiw.)

2) If I have multiple tracks, all created by copying channels from the first track, sometimes the channels that are supposed to be exact duplicates of the channels in the first track end up becoming separate channels. In other words, if I turn a channel on in a track, it should turn on in the first track as well because they are exact duplicates. However, it is not turning on in the first track. When I look at the xml file, there are two separate channels, with the same name, but different savedIndex numbers. The second one will be a higher number than we have physical channels, but it will have the same unit and channel number that the channel in track one has ... so maybe this is not what is messing up our show.

This one is harder to recreate because I have no idea what caused these anomalies. I have my megatree in 4 different tracks, but it is only messed up in one of those tracks. In three of the tracks, if I make a change, the change is reflected in all of the other three tracks. I think it comes down to groups again.

[jstorms]

I ran into a similar issue. By making my track 1 my Master track with no channel groups, it worked itself out. When loading onto that first track it doesn't handle hitting a channel group so well. When I did this all my other tracks worked just fine. It was a tense couple of hours though.

[Amie]

jstorms wrote:

I ran into a similar issue. By making my track 1 my Master track with no channel groups, it worked itself out. When loading onto that first track it doesn't handle hitting a channel group so well. When I did this all my other tracks worked just fine. It was a tense couple of hours though.

A tense couple of hours?! I've been sweating this for 3 days now. Thank you for your help. Now I need to decide if I should go back to 3.0.2 or not since I am running on xp.

[ClydeL]

I found that when I imported the new channel config to the old sequence, what was messed up was the channels I had added new for the 2011 show. So, All i did was insert a blank channel into the old sequence in the exact place I had my new channel config arranged...

Worked like a charm. I was able to completely quit using the copy and paste method which took nearly 1 hour per sequence. After doing 3 or 4 sequences, I was able to remember exactly where I had inserted a new channel and was able to convert about 4 sequences in an hour with perfect results.

The only thing left was to add sequencing to the new display items.

[Amie]

I appreciate your comment Clydel, I know exactly what you are talking about. However, this is a one-to-one copy...all channels are exactly the same, in the same location and there are no additional or missing channels. The problem is coming from the fact that I am using groups and multiple tracks with S3.

[bob]

Importing and exporting channel configuration can be complicated, and the more complicated the sequences the more complicated the operations.

One of the things that's done during import is that the tracks in the sequences are (temporarily) "flattened" to only contain channels. If a track indirectly contains the same channel multiple times, the flattened track will directly contain the channel only once. If you're not expecting that, it's possible that this could lead to things like the channels seeming "off".

If you send me the following:

(1) A sequence to export from;

(2) A sequence to import to;

(3) An explicit and specific explanation of what happens that you weren't expecting to happen;

I'll check it out and see what the deal is.

[Amie]

Bob, I would be more than happy to create test scenarios if/when we finally get a show running, but right now, I am trying to rewrite all of my sequences and don't have time to deal with this anymore. Thank you for your offer though.

The problem I am having was completely outlined in this thread. I just wish that I had found it before it was too late. This really is a big deal and I think everyone should know about this issue so they don't make a mess out of their sequences like I have. It didn't happen until I had to start adding the unit and channel info to my sequences, and that was just a few days ago...after all of the sequences were programmed..

[Amie]

I have finally found some time to set up a small test case. There are worse scenarios than these, but I will start slow.


1. Create an animation sequence with several channels that you can remember the order of those channels.





2) Export the channel configuration file.





3) Create a group in track 1 and copy it into a new track, lock track 1








4) Turn on a few cells in New Track, to verify that the channels are connected in both tracks.





5) Rearrange the channels in the New Track. Note the order of Track 1 now.




6) Select to ‘Remove Channel From Channel Group List’ of one of the channels in a group in New Track. Note that channel was removed from the group in both the New Track, and Track 1, even though Track 1 is locked.




7) Add a new channel in NewTrack called BEAT, at the top of the group.




8) Finally, Import the Channel Configuration File, which every always does, in order to update their unit and channel values before a show.



Attached files

[Paul Roberson]

Problem still the same in the latest release 3.1.4 and the latest beta version 3.1.112

This is well documented in this thread and this one also.

Please submit a wish to change this to wishlist@lightorama.com . More information about what I sent to wishlist in this thread. (same as the hyperlink above)

[Amie]

Yes, I am aware that it is well documented in the thread that you posted. That is an excellent thread. However, Bob asked for more info, so I gave it to him. Until this is fixed, and until the LOR Developers realize the potential that these bugs have to completely destroy sequences, I will continue to bring this issue up and to show more examples and testcase scenarios, even if it is well documented somewhere else. This should not just be a wishlist, this should be a MAJOR bug fix. The fact that Track 1 can be changed, rearranged, channels can be deleted, etc. even when it is locked, is not simply something that I am going to put on a wish list for them to add when they feel like getting around to it. I have reported it and was sent a message back stating that it was fixed in the latest release. Obviously, they don't understand the problem yet. I referenced your thread in the message that I sent, btw.

I am seeing more and more people report that they are seeing strange things happen and I know that it comes down to this issue. I can create a scenario, that is not mentioned in your link, that will delete 1/2 of a channel's programming information and keep the other 1/2 perfectly intact. It is because of how groups are linked together and what happens when you import the channel configuration file.

Here is code for a test case where I have 1 channel, "Green", that has been duplicated and shouldn't have been. Now, channel 'Green' has different data stored in 2 locations, but only 1/2 of the data is stored in Track 1 and so that is the only information that will play in the sequence.

<?xml version="1.0" encoding="UTF-8" standalone="no"?>

I can fix this by editing the xml file and transferring the sequence data from savedIndex=1 over to savedIndex=4. Guess what I have been doing for the past week to try and fix my sequences?

It is frustrating to see LOR developers blame the users and say that they are doing something wrong when, in fact, it is a software issue.

[Paul Roberson]

I was certainly not pointing out that it was well documented in the other thread and shouldn't have been in this one.
I think if you re-read what I posted you will see I agree with you 100%. I think this is a serious issue and it should be fixed ASAP. I was told to send it to the wishlist (and feel the same as you about that) and am trying to get more people to send it to the wishlist so it gets fixed quicker.
I think the more you or anyone post about it the better.

[Paul Roberson]

Btw. It's not considered a "bug" because it is working as it was designed. Although I dont care for the design!

[Amie]

Thank you for clarifying, Paul.

Paul Roberson wrote:

Btw. It's not considered a "bug" because it is working as it was designed. Although I dont care for the design!

You've gotta be kidding me. How can something that can completely destroy our sequences that we spend tons of hours creating, be 'as designed'. I watched my programmed sequences outside on our lights and they were so screwed up that our show hasn't started yet. I am still rewriting all of my sequences. I am really ticked if this is really 'as designed'.

I can submit a wishlist, but my wishlist will be that they fix this bug!

[bob]

I have reported it and was sent a message back stating that it was fixed in the latest release.

Where were you told that? Exactly what was said?

[Amie]

I opened a support ticket on the LOR site. I linked to the thread that Paul has listed and explained that there was a bug in the software that had caused serious damage to my sequences.

The next day, I received an email back stating:

Please download the updated software that was posted just a few hours ago (version 3.1.4).  Seems the problem causing grief was a memory leak in the LOR Control panel.  Very tough to find.  http://www.lightorama.com/SoftwareDownloadPage.html.

[bob]

The fix made in 3.1.4 is unrelated to what you have pointed out. It was at least potentially related to a bunch of strange symptoms, so I assume that the person who wrote that saw a strange symptom and thought that perhaps it was one of the strange symptoms in question.

[Amie]

Yes, I agree. I need to make another report.

[Amie]

Report submitted and I think I received a pretty good response. They recognized the errors as actual errors, and said it would get fixed but not for this season, which I wouldn't expect anyway.

[Paul Roberson]

Good to hear!!

[semme]

could it not be a real idea to

create a tool as in aurora sequencer it works

import options

match channels by channel name
match channels by net/id/chan sitting only
match channels by name then by net/id/chan:D