bmendez wrote:I think what we are seeing now is that there are in fact two errors happening that result in the same javascript error message: (1) The random failed requests to Amazon and (2) corrupted images in some of the movies. We will begin working on having the code simply try again whenever a request fails without warning users of the error. But for corrupt images we will want to keep the error message to be able to identify and fix them.
I disagree. I have seen no evidence of either random failures from Amazon nor of corrupted images. The only time I have seen a refresh appear to fix the problem is when the refresh results in a different movie being loaded. This will always occur if the error is in the movie being preloaded (the error message says ondeck movie). It may occur with the current movie (the error message says next movie) if the current URL does not specify a current movie (e.g., the URL from the virtual microscope button or from the get new movie after the unusual path message). But if a movie generates an error, that appears to be very consistent. I've checked some of the reported movies here (the better reports, with links and assurance that the reported ID is the ID of the movie in error and not of the preceding movie in a random sequence) and the errors persist at least over several days.
The internet being what it is I suppose there are occasional cases of a file download that fails due to random dropped packets. But all the evidence is that these are quite rare, and I've seen no evidence that they have actually been observed.
Further I've tried the failing URL (from the error message) directly in the browser. The result is an error message from the Amazon server indicating that the file cannot be found. This is not consistent with a corrupted file, which would cause an error from the browser when it attempted to process the file. The file is never sent from the server, and I don't think the server checks the contents of the file for validity.
I've been continuing to post the results because the problem with the Java errors is great. and as well, frustrating. However, I have been getting some rather harsh e-mails regarding my posts here. I am keeping track of most of the corrupt movies I see. So. Do the staff members at Berkeley want a list or no? Is yes you you want it publc posted or in a private e-mail/pm? I am very very eager to help resolve this error and will do anything within my means to resolve it. So is there any staff member that would be able to advise me as to what to do here?
I'm sorry you are getting PMs. I would think that as long as your posts are not offensive and in line with the topic, any topic, then they should be acceptable. I would hope anyone with complaints would take it up with the Stardust team or pass the concerns on to the mods instead of taking it to you.
Bryan has given instructions on how to handle the problems earlier in the thread so perhaps he has enough data. But until we hear otherwise, I would assume that this is an active thread and you can post to it if you want. If someone does not like to see these posts, I would think they can chose not to view the thread.
Thank you for taking the time to get the data and good luck with the search.
it's just that the problem is really past the point of being just a nucence. I find it very very frustrating especially wih two windows open and the errors keep taking over other apps. not only that but when you have an error in the buffer movie you get it twice. I am also very patient. I work on a lot of computers every day and understand the issues with them. but if the data can help even remotly, I have no problem recording the information and posting it it only takes about 1 second per error to do it.
I re-read the message posted and that sounds like exactly what we were doing all along. ignore, bad focus, etc.
so, I can still continue to gather data. no biggie. I just want to help in any way I can to resolve the issuse at hand.