In addition: according to README in timsieved directory, proxying is not implemented yet indeed :(((
I've found more recent version of timsieved in CVS: last updated at 10 March 2003. I hope proxying is working there ... how can I compile the CVS sources: there is no makefiles, configure etc ... ? Any help would be so great. On Friday 20 June 2003 16:47, Dmitry Novosjolov wrote: > Hello, > > I'm finishing my Murder cyrus-2.2a installation: now I want that timsieved > running on a frontend server would be able to proxy requests to a proper > backend. > > I'm accessing a sieve service with a websieve application (perl) (I also > tried to work with sieve via sieve plugin for SquirrelMail php based web > interface) > > When I'm connecting to the backend's timsieved directly with websieve all > is fine. > But when I connect to the frontend nothing works. But sieve should look the > same via both frontend and backend connections, like imap or pop3 do. > > In documentation about murder installation is said that sieve does not need > a proxy, for timsieved supports refferrals and can work as proxy without > any actions from the client side. Seem like it's not so, for when I'm > connecting with SquirrelMail plugin it shows: > ----- begin ----- > UNKNOWN ERROR (Please report this line to [EMAIL PROTECTED] to include > in future releases): BYE (REFERRAL "gw") "Try Remote." > Unable to load script to server. See server response below: > > BYE (REFERRAL "gw") "Try Remote." > UNKNOWN ERROR (Please report this line to [EMAIL PROTECTED] to include > in future releases): > ----- end ----- > > Previously with a help of this mailling list I discovered that pop3d also > does not need a proxy, which conflicts with documentation, so it's possible > that there is something similar relating timsieved. > > Question: is it the case ? > > In log files I see that authentication for sieve is going the way it should > (first on a frontend server, then on backend). > > Has anybody got working properly timsieved on a frontend in murder > installation? > > I'm stuck with that, any help would be great. > Thank you. -- Best regards, Novosjolov Dmitry
