[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 b...@mogwai.be changed: What|Removed |Added Status|CONFIRMED |RESOLVED Resolution|--- |FIXED Latest Commit||https://invent.kde.org/mult ||imedia/kasts/commit/c5f4fd2 ||3fc8358f8a3f5c1649b1a00c14b ||b4eb57 --- Comment #7 from b...@mogwai.be --- Git commit c5f4fd23fc8358f8a3f5c1649b1a00c14bb4eb57 by Bart De Vries. Committed on 28/02/2023 at 13:50. Pushed by bdevries into branch 'master'. Set database to WAL mode on startup Using WAL mode avoids having to lock the database for certain transactions. Therefore, deadlocks between read/writes from different threads should not happen anymore. These were rare, but happened sometimes on slower hardware or slow storage devices. M +22 -0src/database.cpp M +1-0src/database.h https://invent.kde.org/multimedia/kasts/commit/c5f4fd23fc8358f8a3f5c1649b1a00c14bb4eb57 -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 b...@mogwai.be changed: What|Removed |Added Status|REPORTED|CONFIRMED Ever confirmed|0 |1 --- Comment #6 from b...@mogwai.be --- I've implemented WAL mode for the sqlite database, which should avoid any kind of database locking issues. However, with this mode the database has to be located on a local filesystem. -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 --- Comment #5 from b...@mogwai.be --- Just re-checked the sqlite docs, and it's saying that - while they don't forbid it - sqlite databases should actually never be put on NFS because of a broken locking implementation. -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 --- Comment #4 from b...@mogwai.be --- Well, to be honest, having the config directory on NFS is going to be the root cause to your problems. The refresh action heavily relies on sqlite transactions to and from the database from multiple threads. Putting the sqlite DB on the network is asking for trouble since it was never designed for that. I have very occasionally also seen the DB transaction errors on very slow ARM systems. If reads/writes across several threads are getting very slow, they can start blocking each other. That way you can get into a deadlock where multiple threads are waiting on each other to finish, which will never happen. I was already looking into using even more performant sqlite settings that don't have those deadlock problems. However, with that solution it would simply refuse to open a connection to a DB on a non-local filesystem. -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 linus.kardell+kdeb...@gmail.com changed: What|Removed |Added Resolution|WAITINGFORINFO |--- Status|NEEDSINFO |REPORTED -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 --- Comment #3 from linus.kardell+kdeb...@gmail.com --- Created attachment 156801 --> https://bugs.kde.org/attachment.cgi?id=156801&action=edit kasts-log.txt Sorry for the delay. Here is the log, though I've tried to redact the lines for each podcast and episode (otherwise it was 35 MiB). I have 83 podcasts with over 1 episodes total. There are a lot of lines about "database is locked". Here I have the config directory on NFS; with the config directory on a local drive the SQL errors go away, and it performs a lot better, though even then it does hang for a few seconds when refreshing. -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 --- Comment #2 from Bug Janitor Service --- Dear Bug Submitter, This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging If you have already provided the requested information, please mark the bug as REPORTED so that the KDE team knows that the bug is ready to be confirmed. Thank you for helping us make KDE software even better for everyone! -- You are receiving this mail because: You are watching all bug changes.
[kasts] [Bug 465110] Unresponsive while refreshing podcasts
https://bugs.kde.org/show_bug.cgi?id=465110 b...@mogwai.be changed: What|Removed |Added Status|REPORTED|NEEDSINFO Resolution|--- |WAITINGFORINFO --- Comment #1 from b...@mogwai.be --- Could you perhaps start the kasts flatpak from the command line using the following command: `QT_LOGGING_RULES=org.kde.kasts.fetcher=true flatpak run org.kde.kasts >& kasts-log.txt` and then make Kasts refresh the podcasts. After you've done that, could you share the output file `kasts-log.txt`? Please also include any warnings or errors that still show up on the command line. That will give some debug output relating to the update routine. Of course, do first check whether that output file contains data that you wouldn't want to share. -- You are receiving this mail because: You are watching all bug changes.
