X-Spam-Check-By: sourceware.org Date: Thu, 18 May 2006 09:49:55 -0400 (EDT) From: Igor Peshansky Reply-To: cygwin AT cygwin DOT com To: "Bryan D. Thomas" cc: cygwin AT cygwin DOT com Subject: Re: "fork problem" debugging In-Reply-To: Message-ID: References: <20060518023929 DOT 65449 DOT qmail AT web31310 DOT mail DOT mud DOT yahoo DOT com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm Precedence: bulk List-Unsubscribe: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com On Thu, 18 May 2006, Bryan D. Thomas wrote: > "Igor Peshansky" wrote: > > I've been plagued by these problems for a while (since you didn't provide > > full links in your message, I don't know whether you cited my message > > among them without a lot of cutting-and-pasting). > > I didn't cite your message[1], though it was one of the sources of the > suggestion to factor out antivirus and firewall software. Thanks for that. > Also, I will use full links from now on, since my mail software seems to > wrap the lines haphazardly no matter how I try to trim them. Actually, I was referring to ... FYI, the above is shortest real link you can give to a mailing list archived message. > > unloading and reloading just Cygwin1.dll (by > > exiting all Cygwin processes) fixes the problem for me (until the next > > time I run that resource-intensive script that reproduces the problem). > > This workaround is effective for me, also. Well, the fact that only Cygwin is unloaded and reloaded pretty much seems to exclude any pure Windows cause, doesn't it? > > a patch (attached) that makes it print out more of the seemingly > > relevant information > > If I understand the patch correctly, it changes fork.cc and pinfo.cc in > winsup/cygwin. Those who wish to use it would download the winsup-src > snapshot package, apply this patch, then compile a new cygwin1.DLL > something like [2]? Exactly. Once you've built a version of cygwin1.dll from CVS, apply the patch and rebuild. > > I can say that I've been running with > > SharedSection=2048,6144,1024 (double my original values), which had no > > noticeable effect on the frequency of the fork errors (i.e., I still get > > them reproducibly). > > Thank you for this data point. Hopefully we can look elsewhere with > confidence, at least for the most common reason(s). > > When I was looking for your message, I found another from Dave Korn[3] > in which he suggested using -x option to bash to see details of the > failure. This is also used in what I called "nonsense testcase"[4]. It > seems that -x would be a useful first pass at gathering data, less > volume than strace? Umm, -x will only show you which command fails, but not the way in which it fails. As I said above, I don't even think strace has enough info to debug this. > [1] http://sourceware.org/ml/cygwin/2006-04/msg00127.html > [2] http://sourceware.org/ml/cygwin/2006-05/msg00262.html > [3] http://sourceware.org/ml/cygwin/2006-04/msg00125.html > [4] http://sourceware.org/ml/cygwin/2006-05/msg00422.html Sigh, I wish I had more time to track this down... Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_ pechtcha AT cs DOT nyu DOT edu | igor AT watson DOT ibm DOT com ZZZzz /,`.-'`' -. ;-;;,_ Igor Peshansky, Ph.D. (name changed!) |,4- ) )-,_. ,\ ( `'-' old name: Igor Pechtchanski '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! "Las! je suis sot... -Mais non, tu ne l'es pas, puisque tu t'en rends compte." "But no -- you are no fool; you call yourself a fool, there's proof enough in that!" -- Rostand, "Cyrano de Bergerac" -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/