On Thu, Aug 29, 2002 at 12:28:11PM -0400, Gene Parks wrote: > Well I had a similar problem but no one answered me. So I will be kind > enough to answer you. At this stage in the game the system is not > working correctly with multiple hosts as the server. You can make one > host the server and every other one the client and it should do like you > expect it to. Not sure if the developers are going to take another look > at the way it performs but at this point this has been my experience. >
Sorry that I missed your mail Gene. An anoying bug in both pbsadd and pbscheck mad it impossible to use multiple servers. I have debuged the problem and a fix will be comming soon. As a hint, the port value is modified after the first send and so all other request normaly land in nirvana. There seems to be another bug in pbscheck. The random function returns sometimes a value to big and so a pbs request is sent to 0.0.0.0 (again nirvana). This should be fixed in the upcoming 20020901 patch. > Also note that pbs* will not work with the control patch applied. You > actually have to compile twice. Once with the control patch for the > stuff you want to work and once without the control patch to get pbs > stuff working off of control files. Kind of the round about way to get > it working but it does. > Why do the pbs* tools not work with the control patch? Is it a compile error or a more fundamental flaw? > I have pbs running on 4 machines with one as the host and the other > three as clients. I would like to have a little more redundancy but > since I did not write the software I can not complain. > If it is an obvious bug, like in your case, you should probably repost the report. It happens that I miss some important qmail-ldap messges. -- :wq Claudio
