Bugzilla – Bug 4558
Server crash in Slim/Networking/Select.pm
Last modified: 2008-12-18 11:12:53 UTC
SlimScrobbler plugin user reported http response from Last.FM crashing SlimServer repeatedly. Clearing his submissions stopped the problem but thought I should report this in case it's obvious what the problem is. Here's some d_async_http log: 10:13:53.8743 Async::HTTP: Headers read. code: 200 status: OK 2006-11-30 10:13:53.8753 bless({ "cache-control" => "no-cache", connection => "close", "content-type" => "text/plain;charset=ISO-8859-1", date => "Thu, 30 Nov 2006 18:13:53 GMT", pragma => "no-cache", server => "Apache-Coyote/1.1", }, "HTTP::Headers") Not a SCALAR reference at /usr/local/slimserver/Slim/Networking/Select.pm line 274.
Looks like writeNoBlock is called with a bad $chunkRef? Do you call this directly or rely on the async code?
I am just using the async code!
We're kind of trying not to fix 6.5.1 bugs much and allow Dan and everyone to focus on 7.0. If this seems to be a serious problem we can have a deeper look.
this crash is already avoided in 7.0, as part of the plugin api rework. it is probably some sort of intermittent connection/response problem. if avoiding the crash is important for 6.5.2, and simple eval might be useful just to trap the crash. Otherwise, this can probably be retargetted for 7.0 and marked as fixed (in the interest of accuracy).
That would be fine by me. This was only every reported once that I know of.
Let's do that, then!
This bug is being closed since it was resolved for a version which is now released! Please download the new version of SqueezeCenter (formerly SlimServer) at http://www.slimdevices.com/su_downloads.html If you are still seeing this bug, please re-open it and we will consider it for a future release.