Gregor Zattler <[EMAIL PROTECTED]> writes: > Dear Michael,
Hi, > O.K. I understand this now and also did find the paragraph in the > manual which says so. But I do not understand why this optimization > takes place regardless of its preconditions. Since there are no > inline transfer methods to use, they should not be used any way. > > At least to me it seems then the default is the wrong way around: It > should be set to "0" in order to function under every circumstances > and the manual should hint for possible optimazions and their > drawbacks. A default value shall serve the majority. It is a rare case, that there is no encoding routine on the remote host. >> Additionally, you need to apply the appended patch (there was an >> error, indeed). > > With tramp-copy-size-limit set to "0" and you patch applied it now > works very well. The patch is already committed to the repositories. So you shouldn't have a problem in the future. I'm thinking of setting tramp-copy-size-limit to 0 automatically, when there is no remote encoding routine. But this isn't trivial. > Thank you for your support and time spend on my problem. > > Regards, Gregor Best regards, Michael. _______________________________________________ Tramp-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/tramp-devel
