[I trimmed the CC list and changed the subject.] >>>>> "ZJ" == Zbigniew Jędrzejewski-Szmek <zbyszek@xxxxxxxxx> writes: ZJ> IIUC, the code does a sleep loop, and does not use polling. So it's ZJ> not particularly efficient, but it should work. You are probably right; I had forgotten how I'd written it but I thought I had it sleep waiting for new events. At least, that's how I'd wanted to write it. But that code is kind of a mess, and it's been a long time since I wrote it. Now I think I may even be mischaracterizing the nature of the failure since after a while I just stopped running that code. Anyway, you're right; it just iterates over the journal object and then goes to sleep for a bit and then wakes up iterates over the journal object again. What appears to be happening is that at some point there are simply no new events present when it does this, regardless of what is actually going into the log. Obviously I can nuke the journal object occasionally and reinitialize it; that would probably work around the problem but certainly seems to be a worse hack ZJ> That should not be a problem, but maybe this is tickling some bug in ZJ> the journal code? That's what it feels like. I might be able to hack something up where I force a journal rotation. Rotation of the journal doesn't invalidate a cursor, does it? ZJ> One suggestion would be to attach gdb to the hanging script and look ZJ> the sd_journal object. I can do that. Is there any way to get at any useful debugging state from within python? - J< _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
