Just making sure

[desertgrandma]

When I plug in my Kindle, and Calibre is running, its "getting list of books".

On my K1 I had so many it took a long time to do that....there was a post that told what to disable so that it bypassed this.....

Was it

preference--advanced--plugins--Kindle device interface by John Schimber--communicate with the kindle ebook reader?

I'd like to do that with my K3 and just want to make sure I'm disabling the right thing (of course I'd disable for the k2/3)....Thanks!

[chaley]

It should be slow only once. Once it finishes the first (long) time, calibre keeps a copy of all the data it needs and can check very quickly.

Try letting if finish. Might take hours, but only once.

[desertgrandma]

Originally Posted by chaley

It should be slow only once. Once it finishes the first (long) time, calibre keeps a copy of all the data it needs and can check very quickly.

Try letting if finish. Might take hours, but only once.

Thanks for the reply.

[pghaworth]

Originally Posted by chaley

It should be slow only once. Once it finishes the first (long) time, calibre keeps a copy of all the data it needs and can check very quickly.

Since I upgraded beyond 0.7.19 it's slow whenever connecting to my K2US. It mostly hangs for about 5 minutes [seems to hang]. It never took this long before and I only have about 20 books.

Would you please post the instructions the OP was trying to confirm? Thanks.

[itimpi]

It was probably the PDF reader plugin that was disabled to improve performance.

[chaley]

Originally Posted by pghaworth

Since I upgraded beyond 0.7.19 it's slow whenever connecting to my K2US. It mostly hangs for about 5 minutes [seems to hang]. It never took this long before and I only have about 20 books.

It shouldn't take that long to recognize a device containing 20 books even in the worst of cases.

Is it possible you could run calibre in debug mode when you connect the device? That would tell us what calibre is doing for that time, and might give enough information to fix whatever is broken.

To run calibre in debug mode, do the following:

1) be sure that the kindle is not connected.
2) open a command line box however your OS does that, then type

calibre-debug -g > caldebug.txt

3) once the GUI has opened, plug in the kindle
4) wait until it is recognized and calibre is stable. Don't do anything else
5) quit calibre
6) post the file caldebug.txt here

Feel free to look inside the file. There shouldn't be anything personal or private inside, but you might feel differently.

[pghaworth]

I think I should clarify. When I said I have 20 books, I meant on my Kindle. In my Calibre Catalog I have over 2,000, and a lot of pdfs. Would that be the problem? Too large of a database?

I will do the debug but wanted to clarify before I went any further.

[chaley]

Originally Posted by pghaworth

I think I should clarify. When I said I have 20 books, I meant on my Kindle. In my Calibre Catalog I have over 2,000, and a lot of pdfs. Would that be the problem? Too large of a database?

I will do the debug but wanted to clarify before I went any further.

A 2,000-book library is not particularly large. And in any event, the delay should be entirely dependent on the number of books on the device. If it isn't, then it would be very good to know why.

[pghaworth]

Well, when I used the debug method it connected perfectly and took less than a minute.

Last night after I posted initially about the issue I tried it again three times and each time the job list screen showed the jobs taking at least four and a half minutes, and once was just a couple of seconds under five minutes.

Since this faster connection was using the debug method I'll try again without. Something else it's started doing is not recognizing the device at all. I have to unplug it and plug it in again. I did try restarting my computer and Calibre when this behavior is exhibited but it doesn't seem to affect anything.

Okay, tried it again without debug. The first time it didn't recognize the device at all. Second time [after unplugging and reconnecting] it took about a minute and a half to know that the device was there and then was pretty fast with the jobs. Well under the five minutes it's been taking, but still around two minutes by the time it sees the device and then processes the jobs. That still is much better than the five minutes I was dealing with yesterday.

Since I hadn't tried to connect this morning before trying the debug I can't speak to anything that would have made a difference of this sort with the connection time. Nothing had changed since last night so it's a bit of a mystery. Thanks for your help.

[pghaworth]

Originally Posted by chaley

A 2,000-book library is not particularly large. And in any event, the delay should be entirely dependent on the number of books on the device. If it isn't, then it would be very good to know why.

Here's the debug text:

Starting up...
Started up in 6.1850001812
Job: 1 Get device information finished
No details available.
DEBUG: 0.0 USBMS: Fetching list of books from device. oncard= None
DEBUG: 0.0 USBMS: dirs are: G:\ documents
DEBUG: 0.8 USBMS: count found in cache: 38, count of files in metadata: 39, need_sync: True
DEBUG: 0.8 USBMS: starting sync_booklists
DEBUG: 0.9 USBMS: finished sync_booklists
DEBUG: 0.9 USBMS: Finished fetching list of books from device. oncard= None
DEBUG: 0.9 USBMS: Fetching list of books from device. oncard= carda
DEBUG: 0.9 USBMS: Fetching list of books from device. oncard= cardb
Job: 2 Get list of books on device finished
No details available.
DEBUG: 7.0 USBMS: starting sync_booklists
DEBUG: 7.2 USBMS: finished sync_booklists
Job: 3 Send metadata to device finished
No details available.
get_metadata cache: 3176 calls, 99.94% misses

As I mentioned this connection was extremely fast, like I'm used to, and was unremarkable. The connection afterward was substantially slower but still an improvement.

[chaley]

Originally Posted by pghaworth

Here's the debug text:

Starting up...
Started up in 6.1850001812
Job: 1 Get device information finished
No details available.
DEBUG: 0.0 USBMS: Fetching list of books from device. oncard= None
DEBUG: 0.0 USBMS: dirs are: G:\ documents
DEBUG: 0.8 USBMS: count found in cache: 38, count of files in metadata: 39, need_sync: True

This line indicates that a book showed up that wasn't there the last time calibre connected. Is this reasonable/expected?

DEBUG: 0.8 USBMS: starting sync_booklists
DEBUG: 0.9 USBMS: finished sync_booklists
DEBUG: 0.9 USBMS: Finished fetching list of books from device. oncard= None
DEBUG: 0.9 USBMS: Fetching list of books from device. oncard= carda
DEBUG: 0.9 USBMS: Fetching list of books from device. oncard= cardb
Job: 2 Get list of books on device finished
No details available.
DEBUG: 7.0 USBMS: starting sync_booklists

The six seconds from when it finished getting books to this sync is somewhat disturbing. What is your metadata management setting?

DEBUG: 7.2 USBMS: finished sync_booklists
Job: 3 Send metadata to device finished
No details available.
get_metadata cache: 3176 calls, 99.94% misses

As I mentioned this connection was extremely fast, like I'm used to, and was unremarkable. The connection afterward was substantially slower but still an improvement.

Would you consider running with debug on for a while, as a matter of course? I would like to understand where the time differences come from, and you might catch one. No urgency, really.

[pghaworth]

Originally Posted by chaley

This line indicates that a book showed up that wasn't there the last time calibre connected. Is this reasonable/expected?

On device or in Calibre. There was nothing new on the device. In Calibre, yes, possible.

Originally Posted by chaley

The six seconds from when it finished getting books to this sync is somewhat disturbing. What is your metadata management setting?

Where would I find this? I poked around without much success. Running Vista 64, if it matters.

Originally Posted by chaley

Would you consider running with debug on for a while, as a matter of course? I would like to understand where the time differences come from, and you might catch one. No urgency, really.

Sure, but I don't connect very often. I keep just what I might want to read on the device and only connect when I need to add a new book. I could do a few unnecessary connections, now and again. How long would I run it on debug?

[pghaworth]

Did a reconnect and here's what I got that is similar to the last time I connected [after the first debug]. Jobs shows roughly 2 3/4 minutes.

Starting up...
Started up in 4.64199995995
Job: 1 Get device information finished
No details available.
DEBUG: 0.0 USBMS: Fetching list of books from device. oncard= None
DEBUG: 0.0 USBMS: dirs are: G:\ documents
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Failed to read MOBI cover
Unknown codepage 235802126. Assuming cp1252
Unknown codepage 235802126. Assuming cp1252
DEBUG: 129.5 USBMS: count found in cache: 0, count of files in metadata: 39, need_sync: True
DEBUG: 129.5 USBMS: starting sync_booklists
DEBUG: 133.3 USBMS: finished sync_booklists
DEBUG: 133.3 USBMS: Finished fetching list of books from device. oncard= None
DEBUG: 133.3 USBMS: Fetching list of books from device. oncard= carda
DEBUG: 133.3 USBMS: Fetching list of books from device. oncard= cardb
Job: 2 Get list of books on device finished
No details available.
DEBUG: 139.3 USBMS: starting sync_booklists
DEBUG: 140.4 USBMS: finished sync_booklists
Job: 3 Send metadata to device finished
No details available.

[chaley]

Originally Posted by pghaworth

On device or in Calibre. There was nothing new on the device. In Calibre, yes, possible.

The debug log indicated that there was one new book on the device.

[Referring to metadata management settings]
Where would I find this? I poked around without much success. Running Vista 64, if it matters.

preferences -> import/export -> sending books to device.

[chaley]

Originally Posted by pghaworth

Started up in 4.64199995995
Job: 1 Get device information finished
No details available.
DEBUG: 0.0 USBMS: Fetching list of books from device. oncard= None
DEBUG: 0.0 USBMS: dirs are: G:\ documents
Failed to read MOBI cover
...
Failed to read MOBI cover
Unknown codepage 235802126. Assuming cp1252
Unknown codepage 235802126. Assuming cp1252
DEBUG: 129.5 USBMS: count found in cache: 0, count of files in metadata: 39, need_sync: True

Now this indicates that something is truly broken. It says that between the last time you connected and this time, *all* the books were changed/deleted. This is not reasonable, but that is what calibre is seeing when it looks at your Kindle's files in main memory.

I am beginning to suspect a hardware problem. Have you recently changed computers? USB cables? USB ports where you plug in the Kindle? Anything?