Closed tassa-yoniso-manasi-karoto closed 2 months ago
Another one on processing ~500 notes with HyperTTS with lastest official build
[165023.247122] anki[31103]: segfault at 0 ip 0000000000000000 sp 00007ffe8f16acc8 error 14 in anki[55c85c0d6000+19a000] likely on CPU 2 (core 2, socket 0)
[165023.247145] Code: Unable to access opcode bytes at 0xffffffffffffffd6.
pinging @dae too, although this occurs only when using the addons I can't tell if this is also relevant to Anki itself or how it is build.
Which Linux distribution are you using? In case it matters. In my experience crashes in the past have happened due to making gui changes on the wrong thread, I'll have to look into it.
Voidlinux 6.6.32_1
Seems gone as of anki 24.06.3, hypertts 1.11.1 , langtools 3.33.
OK thank you for the update.
I initially thought this was bug in the anki flatpak package. Here is the copy pasted report I had posted on their issue tracker:
[The bug] occurs when translating / generating audio for a large batch of notes. Once all of the translations for the fields have been loaded, Anki will just crash with no warning popup, nothing logged in console when running from console.
The bug appeared about 2 month ago. It also affects HyperTTS but it doesn't matter because hypertts adds the content on the fly whereas with lang tool when a crash occurs, everything fetched is lost.I was using only google translate each time.
Here is the outcome I have just had with various amount of note & releases:
Flatpak Anki 24.04.1 Python 3.9.18 Qt 6.6.2 PyQt 6.6.1 94 OK 146 CRASH
110 OK 125 OK 135 OK 92 OK 146 CRASH (different set than the prev 146 notes)
As I said the bug wasn't there just ~2 month ago so I tried an old build: Official Anki 23.12.1-linux-qt6 146 OK (the same set which had just crashed)
Official Anki 24.06.2-linux-qt6 Python 3.9.18 Qt 6.6.2 PyQt 6.6.1 146 OK (the same set which had just crashed) 369 OK ~78 CRASH
Reproduced in the official bug → this is indeed a bug related to the addon