Thing is, if this is that frequent there ottabee many reports of stray locks. I can say we are not seeing that at all but will admit it used to show up with a product called SmartCVS - not pointing the finger at that product, just conveying our experience.
-----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of S I Sent: Wednesday, June 22, 2005 2:01 PM To: info-cvs@gnu.org Subject: CVS Lock Files Hi I've setup a cronjob on our unix server to run nightly to clean up CVS lock files before my build and before tagging CVS. A user using Tortoise accidentally caused certain folders in the repository to lock up during my build (about 3x's in 1 week) and tagging took about 600 minutes (pending) to complete until I manually removed the offensive "#cvs.*.*" files on the server side and we cleaned up her Windows PC of any cvs.lock.tmp or cvs.lock.folder pattern files and folders. We seem to be OK now since I deleted the files. During my research I've found out that there're PERHAPS 3 ways files COULD be locked up UNINTENTIONALLY (NOT executing cvs admin -l): 1. Multiple users concurrently doing checkout, commit, or update. 2. Tagging? 3. And by means of an IDE such as IntelliJ or Eclipse? Do you see anything wrong or dangerous with my unix script below? #!/bin/sh # Delete lock debris on cvs find /usr/local/cvs -name "#cvs.lock*" -print | xargs rm -fdr find /usr/local/cvs -name "#cvs.lock.*" -print | xargs rm -f find /usr/local/cvs -name "#cvs.rfl.*" -print | xargs rm -f find /usr/local/cvs -name "#cvs.wfl.*" -print | xargs rm -f find /usr/local/cvs -name "#cvs.pfl.*" -print | xargs rm -f find /usr/local/cvs -name "#cvs.tfl.*" -print | xargs rm -f find /usr/local/cvs -name "cvsloc" -print | xargs rm -fdr Thank you Steve _______________________________________________ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs _______________________________________________ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
