search for: piped_child

...ter: %s\n", fname_convert_cmd); + split_on_spaces(fname_convert_cmd, args); + /* Invoke child pipe with non-blocking IO and without registering it for + autocleanup (the latter may blow up the all_pids table, and is not needed + since we have our own cleanup handler. */ + conv_pid = piped_child(args, &conv_read_fd, &conv_write_fd, 0, 0); + set_nonblocking(conv_write_fd); + set_nonblocking(conv_read_fd); + } +} + +/** + * Kills the filename converter process. Should be called when the file + * list creation is done. We assume that the converter will terminate + * soon after its s...

...ier rsync on cygwin successfully for a long. I have now moved to cygwin 1.7.0, which required moving to rsync 3.0.6-1 at the same time. (The -1 is for minor cygwin-specific patches.) When I do almost any command with rsync, such as the one below, it breaks at a dup2 after a fork in pipe.c (routine piped_child). Windows gives error code 6, which is translated by cygwin to errno 9 = EBADF, and rsync then terminates. rsync -avzuP foo roc:talks/ Even rsync -r . fails the same way, though it does not need a network connection. Steven Monai suggests that the problem may be a mismatch in protocol (30 vers...

...slower. After poking around with strace, it seems that rsync's child ssh sets its stdERR non-blocking, and that stderr has been inherited unchanged from the top-level rsync. (The rsync has supplied pipes for its child's stdin and stdout, but left the stderr alone; see rsync-2.6.2/pipe.c::piped_child().) Because of the "2>&1", the top-level stderr is a dup of the top-level stdout, so ssh has inadvertantly made rsync's stdOUT non-blocking. Rsync is not expecting that, and does not check the return code from fflush(stdout), so it can silently drop lines from stdout. (See th...

...8 05:05:53 2003 --- rsync-2.5.6-remotebatch/main.c Wed Jul 23 15:55:18 2003 *************** *** 263,270 **** } if (local_server) { - if (read_batch) - create_flist_from_batch(); /* sets batch_flist */ ret = local_child(argc, args, f_in, f_out, child_main); } else { ret = piped_child(args,f_in,f_out); --- 263,268 ---- *************** *** 505,515 **** if (delete_mode && !delete_excluded) recv_exclude_list(f_in); ! ! if (read_batch) ! flist = batch_flist; ! else ! flist = recv_file_list(f_in); if (!flist) { rprintf(FERROR,"server_recv: r...

...(FINFO,"cmd="); for (i=0;i<argc;i++) rprintf(FINFO,"%s ",args[i]); rprintf(FINFO,"\n"); } if (local_server) { if (read_batch) create_flist_from_batch(); /* sets batch_flist */ ret = local_child(argc, args, f_in, f_out, child_main); } else { ret = piped_child(args,f_in,f_out); } if (dir) free(dir); return ret; oom: out_of_memory("do_cmd"); return 0; /* not reached */ } -------------------------------------------------------------------------- -- Configure bugmail: https://bugzilla.samba.org/userprefs.cgi?tab=email ------- You are re...

..., + command[0], strerror(errno)); + exit_cleanup(RERR_IPC); + } + + return pid; +} diff -ur rsync-2.5.6/proto.h rsync-2.5.6-filtered/proto.h --- rsync-2.5.6/proto.h 2003-01-27 04:35:09.000000000 +0100 +++ rsync-2.5.6-filtered/proto.h 2003-11-16 13:20:48.000000000 +0100 @@ -181,6 +181,8 @@ pid_t piped_child(char **command, int *f_in, int *f_out); pid_t local_child(int argc, char **argv,int *f_in,int *f_out, int (*child_main)(int, char*[])); +pid_t run_filter(char *command[], int in, int *pipe_to_filter); +pid_t run_filter_on_file(char *command[], int out, int in); void end_progress(OFF_T size);...

...nical record. + */ + if (am_server) { if (am_sender) { int64 w; *************** *** 300,307 **** } if (local_server) { - if (read_batch) - create_flist_from_batch(); /* sets batch_flist */ ret = local_child(argc, args, f_in, f_out, child_main); } else { ret = piped_child(args,f_in,f_out); --- 321,326 ---- *************** *** 461,466 **** --- 480,491 ---- recv_files(f_in,flist,local_name); io_flush(FULL_FLUSH); + + /* This test for read_batch is needed because of a + * protocol dependency on am_server state, see + * report(). We are quite fortun...

> Just curious: what about the rsync+ patch? Thanks for the reminder. I've just committed Jos's rsync+ patch onto the "branch_mbp_rsyncplus_merge" branch. If it works OK and nobody screams I will move it across onto the main tree tomorrow or Wednesday. I see the patch doesn't add documentation about the new options to the man page, so we should fix that in the future.

...p; !local_server) { if (!cmd) cmd = getenv(RSYNC_RSH_ENV); if (!cmd) @@ -230,7 +230,7 @@ if (local_server) { if (read_batch) - create_flist_from_batch(); + create_flist_from_batch(); /* sets batch_flist */ ret = local_child(argc, args, f_in, f_out); } else { ret = piped_child(args,f_in,f_out); @@ -443,8 +443,8 @@ extern int am_daemon; extern int module_id; extern int am_sender; - extern int read_batch; /* dw */ - extern struct file_list *batch_flist; /* dw */ + extern int read_batch; + extern struct file_list *batch_flist; if (verbose > 2) rprintf(FIN...