Firmwares 4.39.23027 and 4.38.23038

[JSWolf]

Originally Posted by DNSB

Sorry Jon, but that advice was not all that useful before and even less useful now. My experience was similar to @iacallaci's. Using SSH or telnet still showed the database write-ahead logging and share memory files present until after a reboot.

How is it I never had an issue with the database becoming corrupted? I went to the home screen and then connected to USB. Not once did I have a database corruption problem.

[JSWolf]

@geek1011 Could you please figure out how to make Don't uppercase header/footer text and change page number text work in 4/38.23038? Thanks.

[DrChiper]

Originally Posted by geek1011

I just noticed another thing... the libra color is now kobo13 instead of kobo11...

Device wise this makes indeed more sense.
OTOH, my Libra 2 (kobo9) also still does not got updated OTA.
Do I sense some "Libra" manipulations here by Kobo?

[JSWolf]

Originally Posted by DrChiper

Device wise this makes indeed more sense.
OTOH, my Libra 2 (kobo9) also still does not got updated OTA.
Do I sense some "Libra" manipulations here by Kobo?

If you update the firmware manually, you can do it on your time and not have to wait for it.

[DrChiper]

Originally Posted by JSWolf

If you update the firmware manually, you can do it on your time and not have to wait for it.

Nice that you are concerned about me wasting my time, but I've the time on my side
But, as always, there is a reason for my madness:

• Via OTA I can see whether Kobo is really endorsing the update for a particular device;
• Via OTA I get the release notes on the device, which is important for my users;

Likewise, I never install a Windows / calibre / etc. update immediately when it becomes available, but wait until I deem it fit for purpose.

[JSWolf]

Originally Posted by DrChiper

Nice that you are concerned about me wasting my time, but I've the time on my side
But, as always, there is a reason for my madness:

• Via OTA I can see whether Kobo is really endorsing the update for a particular device;
• Via OTA I get the release notes on the device, which is important for my users;

Likewise, I never install a Windows / calibre / etc. update immediately when it becomes available, but wait until I deem it fit for purpose.

Via not OTA.

• Fixed: The eReader may crash while scrolling or swiping in the 'Discover' tab
• Fixed: The eReader database may be left in an incorrect state after connecting and disconnecting the eReader from a computer

Kobo is not endorsing or endorsing a given firmware. That's ridiculous.

[DrChiper]

Originally Posted by JSWolf

Via not OTA.

• Fixed: The eReader may crash while scrolling or swiping in the 'Discover' tab
• Fixed: The eReader database may be left in an incorrect state after connecting and disconnecting the eReader from a computer

Yes, I know and you know. But not everybody who is using an Kobo ereader knows how to find this information or is capable or has the means to use the internet.

Originally Posted by JSWolf

Kobo is not endorsing or endorsing a given firmware. That's ridiculous.

Using OTA is using Kobo's infrastructure, hence Kobo controls/allows you to download their particular software or prohibits it. That you (and I) know a short-cut is irrelevant for casual users.

[Cootey]

Originally Posted by DNSB

I upgraded a Mini to 4.38.23038 and didn't see any issues in very limited testing (4 repetitions of send book, send metadata, open book, read 10-15 pages, skip to end).

Thanks for the update. After making a backup, I went ahead and installed 4.38.23038 manually on my Mini without any issues. I did notice that the menu items feel a tad laggy now, but otherwise, I’m happy to have a fixed firmware on it.

[icallaci]

Originally Posted by JSWolf

How is it I never had an issue with the database becoming corrupted? I went to the home screen and then connected to USB. Not once did I have a database corruption problem.

It could be any number of things. There are many personalized settings and patches that might interfere with each other. I've also wondered if opening several books in sequence (rather than reading just one book to the end before starting another) might have something to do with it, but I wasn't about to experiment. Restoring everything was more trouble than it was worth. Once I discovered that rebooting before connecting solved the problem, I did that. Well, except for the few times I forgot and verified that, yep, the problem still existed. In any case, Kobo seems to have figured out what was causing the problem and fixed it. I have not had a single instance of database corruption with the new firmware in more than twenty connections.

[JSWolf]

Originally Posted by icallaci

It could be any number of things. There are many personalized settings and patches that might interfere with each other. I've also wondered if opening several books in sequence (rather than reading just one book to the end before starting another) might have something to do with it, but I wasn't about to experiment. Restoring everything was more trouble than it was worth. Once I discovered that rebooting before connecting solved the problem, I did that. Well, except for the few times I forgot and verified that, yep, the problem still existed. In any case, Kobo seems to have figured out what was causing the problem and fixed it. I have not had a single instance of database corruption with the new firmware in more than twenty connections.

I have switched books many times before finishing. Sometimes one after another trying to decide what to read. So if that was the problem, I'd have had it.

[DNSB]

Originally Posted by JSWolf

How is it I never had an issue with the database becoming corrupted? I went to the home screen and then connected to USB. Not once did I have a database corruption problem.

On the Kobo devices I had access to, I ran 100 of my open a book, read a couple of pages, close the book, USB connect cycles on each device. The Clara HD came up with 1 corrupt database, the Forma managed 9 corrupt databases, the Mini and Aura H2O managed no corrupt databases and the Sage managed 87 corrupt databases. One friend of mine ran similar tests on his two Elipsas. The original Elipsa gave 63 corrupt databases while the Elipsa 2E managed 4.

From what I was told, returning to the Home or Books screen has no effect on closing the database write-ahead logging and shared memory files. Those are closed when the last connection to the database is removed. The issue seemed to be that the files were not being flushed completely before the USB connection was established hence the <database_name>.sqlite-wal and <database_name>.sqlite-shm files being visible in the files when looking at the .kobo directory.

To me, your experience on a single device is pretty much meaningless.

2.2. Write-Ahead Log (WAL) Files

A write-ahead log or WAL file is used in place of a rollback journal when SQLite is operating in WAL mode. As with the rollback journal, the purpose of the WAL file is to implement atomic commit and rollback. The WAL file is always located in the same directory as the database file and has the same name as the database file except with the 4 characters "-wal" appended. The WAL file is created when the first connection to the database is opened and is normally removed when the last connection to the database closes. However, if the last connection does not shutdown cleanly, the WAL file will remain in the filesystem and will be automatically cleaned up the next time the database is opened.

2.3. Shared-Memory Files

When operating in WAL mode, all SQLite database connections associated with the same database file need to share some memory that is used as an index for the WAL file. In most implementations, this shared memory is implemented by calling mmap() on a file created for this sole purpose: the shared-memory file. The shared-memory file, if it exists, is located in the same directory as the database file and has the same name as the database file except with the 4 characters "-shm" appended. Shared memory files only exist while running in WAL mode.

The shared-memory file contains no persistent content. The only purpose of the shared-memory file is to provide a block of shared memory for use by multiple processes all accessing the same database in WAL mode. If the VFS is able to provide an alternative method for accessing shared memory, then that alternative method might be used rather than the shared-memory file. For example, if PRAGMA locking_mode is set to EXCLUSIVE (meaning that only one process is able to access the database file) then the shared memory will be allocated from heap rather than out of the shared-memory file, and the shared-memory file will never be created.

The shared-memory file has the same lifetime as its associated WAL file. The shared-memory file is created when the WAL file is created and is deleted when the WAL file is deleted. During WAL file recovery, the shared memory file is recreated from scratch based on the contents of the WAL file being recovered.

[Nemo1984]

Originally Posted by geek1011

I just noticed another thing... the libra color is now kobo13 instead of kobo11... this pair of updates is becoming a really big PITA...


Change the BaseAddress value to 0x198201E.

Thank you so much!

[ghira]

OTA for affiliate "kobo" now possible and my Sage is upgrading. Will do Aura One later.

[DrChiper]

Originally Posted by ghira

OTA for affiliate "kobo" now possible and my Sage is upgrading. Will do Aura One later.

Yes, my Libra 2 is now also OTA updating.

[geek1011]

Originally Posted by icallaci

It could be any number of things. There are many personalized settings and patches that might interfere with each other.

FWIW, it's nearly impossible for the patches to interfere with database stuff in any way, as they don't touch any database stuff or affect the control flow of anything using the database in a meaningful way. NickelMenu is unlikely to (the main thing would be if it doesn't clean things up properly if using it to switch views). NickelDBus has a small chance depending on how you use it, but if it isn't actively doing anything, it shouldn't cause any issues. Kobo-UNCaGED shouldn't cause issues since it doesn't do anything when not running. NickelClock shouldn't be causing issues, but it could theoretically do so. Other non-nickel mods wouldn't be the cause of it.

Nevertheless, since the corruption is still an issue on un-modded devices, mods are highly unlikely to affect how/if the bug is triggered.

I would agree with your guess that it has something to do with switching between books, and that being the reason why JSWolf hasn't seen the issue.