Hello,
Maybe I wasn't clear enough, but I know that the version in Lenny is
not supposed to be affected.
I forgot to tag It as Etch specific (I tough reportbug would do It for
me, but I missed that It didn't).
If you don't want to fix this bug, then at least mark It as wontfix
with the reason, but not closed. And no, people with production systems
won't move to Lenny before It is released as Debian Stable.
I've one system where Slony seams to eats a bit over 100MB of ram per
month per instance of the daemon.
I've noticed when the DB started to use the swap, while there was
normally almost one GB of free RAM. I wasn't curious enough to wait for
the OOM killer to quick-in and see what would happens.
I think I should increase the priority of the bug to at least
"important", because if we ever reach OOM conditions, lots of bad things
could happens.
The simple fact that people use Slony is a good hint that they really
care about their DB staying available.
As I said, It might not be serious enough to release a patch just for
this bug, but I believe the bug needs to be kept open and patched at the
same time if there is a security update to It.
Simon Valiquette
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
