From: David B. <dav...@en...> - 2010-06-07 22:05:52
|
On Mon, Jun 7, 2010 at 4:18 PM, David Baelde <dav...@en...> wrote: > This is a bug, for sure. Not so sure actually... > You've hit a rare situation: when icecast > realizes that it lost connection, and sets up attempts to reconnect, > the clock realizes that it's too late and restarts icecast. This does happen, and does trigger a reconnection attempt before the expected 3 seconds, but that's not a big problem. > It's > likely that we're having a race condition here, resulting in the > failure to reconnect. (Looking at the log, it sounds like there are > two parallel threads of reconnection attempts.) On closer inspection the reconnection logs are perfectly normal, and there can't be a race condition in that code. Are you sure that the network was properly back, including DNS and all? |