PNG download stops after page 156

[teckpatt]

I see there had been problem with PNG download before and I noted that me last download too stopped after page 156.

If the backup is done into a cbz container the extension crashes. If done into single files the backup runs to the end but the last page actually written to the filesystem is again 156.

I tried both Chrome and Opera Next with the same result.

Single file download into JPG worked fine.

And thanks for the extension. Buying books so much more fun when you can make DRM free backups.

[Cortys]

This bug should be fixed in the next release. Large files are handled more efficiently there.

[Cortys]

Hi! Could you try to reproduce the issue with the new 2.0 release?

Hopefully this is fixed now.

[benjbubu]

Hello ! I try the last release of comixology backup (2.0) but unfortunetaly, it doesn't work it stucked randomly between 95 et 98%. At this status, there is the "downloading comic" and black screen. I try on several OS and chrome/chromium but didn't work. It's a large comic (more than 200 pages)

[Cortys]

@benjbubu Could you try the latest 2.0.1 beta? The bug is probably fixed there (if it's a timing and not a memory issue).

[ghost]

I've used 2.0.1 but I'm still having similar issues. I've tried to backup the A vs X collected edition which is over 1100 pages, but it always freezes or crashes at 97-98%. This has also happened with 2.0 and earlier.

[Cortys]

Zipping such large comics isn't something I've tested yet and I guess there is not much I can do about that, since there is a 500 Mib memory limit per extension in Chrome.

For now you could just enable single file downloads in the options. That should not cause any crashes.

[ghost]

Thanks for the extension! It adds so much value to Comixology.

For large books would it be possible to have the extension work as normal, but then automatically output a cbz at a user defined percentage and then subsequently continue downloading the remainder of the book? Basically break up the book into two or more cbzs?

[sidick]

I wonder about how practical it would be to write them out to a temporary file and then zip/rar them on the fly rather than hitting the memory limit in Chrome.

On 27 August 2014 09:34, phencer notifications@github.com wrote:

Thanks for the extension! It adds so much value to Comixology.

For large books would it be possible to have the extension work as normal, but then automatically output a cbz at a user defined percentage and then subsequently continue downloading the remainder of the book? Basically break up the book into two or more cbzs?

— Reply to this email directly or view it on GitHub https://github.com/Cortys/comixology-backup/issues/7#issuecomment-53540997 .

[Cortys]

I just noticed, that there is a way to zip very large cbz files by using the HDD instead of RAM for temporary memory. It would be quite slow though. I'll try that.

[sidick]

I think it would be worth having as an option at least, personally all the machines I use nowadays have SSDs in so I probably won't notice much difference.

On 27 August 2014 10:00, Clemens notifications@github.com wrote:

I just noticed, that there is a way to zip very large cbz files by using the HDD instead of RAM for temporary memory. It would be quite slow though. I'll try that.

— Reply to this email directly or view it on GitHub https://github.com/Cortys/comixology-backup/issues/7#issuecomment-53543480 .

[Cortys]

The option is available in the 2.0.1 branch now including some other optimizations.

It should (at least theoretically) be possible to backup really big comics now. I do not own such comics though, so I can't promise anything.

@phencer It would be great if you could test the new version and give some feedback.

[ghost]

OK, just tested on A vs X again and unfortunately it doesn't work. First I tried with the RAM option and it went all the way through but produced a corrupt cbz that wouldn't open. Then I tried the FS option and this time it stopped at 61%.

Both the RAM and the FS options work on smaller books though.

[Cortys]

Thanks for your feedback!

Too bad the issue isn't solved yet. As I've said before: Single image download should definitely work. So if you just want your backup, that's the way to go.

Of course I'll continue improving the zipping backend. I really have no idea why the backup stops and it's quite hard to debug without any comic that big.

Maybe you could redo the backup with FS option enabled and send me the error logs after the download stops at 61%. If you do not know how to do this:

1. (The extension has to be installed in unpacked mode.)
2. Click on the hidden reader tab that was opened when you clicked the "Scan Comic" button.
3. Hit ctrl (or cmd on a Mac)+alt+J. The JS console should be opened.
4. Copy the whole log (if there is any). A screenshot will do as well (if there is no part of the comic on it, we had major issues with cmxlgy regarding such images in the past).
5. Open the extensions page in the settings. Click on the background page link next to the comixology extension. A new window should open.
6. Click on "Console" (if not already enabled).
7. Copy the whole log as you did in step 4.

That's basically all and it would help me a lot finding the cause of this issue.

Thanks again for your support!

[ghost]

Hi,

So I tried to back up A vs X again, so I uninstalled the extension and reinstalled it in unpacked mode. This time it got to 64% and then it froze. I’m not sure if it’s a latency issue or what. When I used 2.0.1 RC 4 and below, sometimes the extension would actually crash and sometimes it would freeze.

I tried to do the steps in your prior post, but when I clicked on the hidden reader tab and opened the console, no log was there. Then when I went to the background page and went to the console tab, and again no log was there. Not sure what is going on.

Thanks again for making this awesome extension!

On Aug 27, 2014, at 6:58 PM, Clemens notifications@github.com wrote:

Thanks for your feedback!

Too bad the issue isn't solved yet. As I've said before: Single image download should definitely work. So if you just want your backup, that's the way to go.

Of course I'll continue improving the zipping backend. I really have no idea why the backup stops and it's quite hard to debug without any comic that big.

Maybe you could redo the backup with FS option enabled and send me the error logs after the download stops at 61%. If you do not know how to do this:

• (The extension has to be installed in unpacked mode.) • Click on the hidden reader tab that was opened when you clicked the "Scan Comic" button. • Hit ctrl (or cmd on a Mac)+alt+J. The JS console should be opened. • Copy the whole log (if there is any). A screenshot will do as well (if there is no part of the comic on it, we had major issues with cmxlgy regarding such images in the past). • Open the extensions page in the settings. Click on the background page link next to the comixology extension. A new window should open. • Click on "Console" (if not already enabled). • Copy the whole log as you did in step 4. That's basically all and it would help me a lot finding the cause of this issue.

Thanks again for your support!

— Reply to this email directly or view it on GitHub.

[ghost]

This time I tried single issue download, and got about 80% of the way before it made Chrome crash altogether. This was with 2.0.1 RC6 using the RAM option. Haven't tried it with the FS option yet.

I tried to do the FS option again, and again it froze, this time at around 57%. I tried to get the logs, but again nothing was there.

I guess A vs X is way too big.

[Cortys]

@phencer Thanks a lot for your reports!

If Chrome itself crashes using single image download, there is pretty much nothing I can do about that. Downloading some thousand images in a row probably isn't a thing they test a lot.

To finish the single image download you can still use the start backup at: 'currently opened page' option. This will continue the backup where it stopped (as long as you've enabled the 'continue automatically' option in the ComiXology reader).

That said I have no idea how to actually fix your issue. I'm quite sure it isn't a memory problem, because I recently fixed all leaks I could find. If anyone has an idea, I'll happily merge pull requests, that can fix this problem.

Since we can consider the "download stops after page 156" issue to be fixed and a solution for your extreme case currently seems to be beyond reach I will close this issue for now.

Maybe we can tackle the problem in a later release.

[benjbubu]

Hey again,

i try with the last 2.0.1, in single image. It stuck at 97%. I try RAM and HDD but the problem is still here. Anyway i got the scanned page from 0 to 97% for a total of 208 pages. 7 pages is missing.

If this problem cannot be solved maybe you can add a feature to scan just 1 page ? For example, with a specific button on the reader to enable a single scan page. That could resolve the problem.

[Cortys]

Hi!

You can set "start backup at: currently opened page" in the options, then navigate to the page where you left off in the reader, close the reader and start a new scan.

This should solve your problem.