Christoph Klaffl
Christoph Klaffl
@thorro but this will only work if __replicate is always the most recent snapshot or you are using --no-stream option.
@jimsalterjrs i think I remember some PR with an rsh implementation and you were objected to that, right? @dvaezazizi @mac-linux-free did you try hpn-ssh patches, they allow to select null...
@jimsalterjrs ping :-) My guess is you won't like this idea as the proposed rsh PR was denied as well, right?
@sasoiliev exit code 65280, needs to be shifted by 8 bits to the right to get the actuall exit code of the ssh command for listing the snapshots -> 255...
@sasoiliev I don't think it's a race condition. But I have a theory, the first synoicd created the control socket, the other runs will us this for their SSH stuff....
@sasoiliev sorry for taking so long to reply. I think the easiest solution would be to replace `$remoteuser-$rhost` with something like `$$` ($$ = current pid of the process). This...
@gaika because it's an unexpected change if the default is changed, if you would like to use bookmarks you can add the flag. But if for example debian just upgrades...
@gaika the previous sync snapshot is deleted on each successful replication, so the bookmark won't help because the snapshot is removed on both sides.
@gaika True, and also if using the "--keep-sync-snap" flag
@gdevenyi do you intend to port the entire code? The hard part will be detection the DST crossover and handle the duplicate snapshot accordingly.