[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Problems with r15909 (ra_svn connect timeout)

From: Malcolm Rowe <malcolm-svn-dev_at_farside.org.uk>
Date: 2005-12-08 21:39:49 CET

On Thu, Dec 08, 2005 at 01:48:32PM -0500, Greg Hudson wrote:
> * The patch was predicated on the notion that OSX has some horrible
> bug related to interrupting blocking connects. *Lots and lots* of
> programs do blocking connects. If OSX really has or had an issue
> related to doing that (and I think we pretty much took the patch
> submitter's word for that), I don't think the answer is for every
> piece of code to change to doing non-blocking connects.

I didn't really agree with the substance of the patch either, but just to
note that, yes, OS X did have a horrible problem with blocking connects,
at least in 10.4.2.

It's fixed in 10.4.3, but prior to that fix, retrying a blocking connect()
call after a signal caused it to return EINTR would actually take out
the entire network stack. I broke my machine several times setting up
a testcase for Apple's bugreporter.


To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Dec 8 21:57:33 2005

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.