I am attempting to use rsync to copy a large filesystem from an HP-UX server to a Linux server with more than enough filespace. This operation fails. A small directory from the same HP-UX server can be transfered just as expected. The HP-UX server is the source. It has 1Gb RAM - the output of bdf for the volume the source files is on is: Filesystem kbytes used avail %used Mounted on /dev/vg02/apps 61440000 52722896 8652256 86% /apps On the target Linux server, I have 1Gb RAM and 1TB of free space: Filesystem 1k-blocks Used Available Use% Mounted on /dev/tcvg1/tcvol1 1007898916 33408 956667148 1% /vol1 I have "apps" configured properly as a service in the rsyncd.conf file. I started rsync in daemon mode on the HP-UX server, and from the Linux server I invoked: rsync --numeric-ids -vvva root@hpuxsrv::apps /vol1/asic_apps/ The file list get transferred, but then I just get repeated lines of errors: recv_file_name(some_dir/fm_v200209) opendir(some_dir/fm_v200209): Not enough space I have tried with --blocking-io and --no-blocking-io, with --bwlimitXXX. Same problem each time. Any insight is appreciated. -- Bill Geddes bill_geddes@agilent.com -- UNIX is user friendly, it's just picky about who its friends are.
More error data: The usage I have described returns an error code of 12 - Error in rsync protocol data stream Also, on the tail end of the std out: opendir(somedir): Not enough space done somedir/ somedir/.somefile is uptodate somedir/somefile.HTML ERROR: out of memory in map_ptr rsync error: error allocating core memory buffers (code 22) at util.c(232) somedir/doc/somefile/ rsync: read error: Connection reset by peer rsync error: error in rsync protocol data stream (code 12) at io.c(162) I know that rsync works only as well as the NIC card drivers in use. I hope that this is not another problem like that. On Fri, 24 Jan 2003 11:54:13 -0700 Bill Geddes <bill_geddes@agilent.com> wrote:> > I am attempting to use rsync to copy a large filesystem from an > HP-UX server to a Linux server with more than enough filespace. > This operation fails. A small directory from the same HP-UX server > can be transfered just as expected. > > The HP-UX server is the source. It has 1Gb RAM - the output of bdf for > the volume the source files is on is: > > Filesystem kbytes used avail %used Mounted on > /dev/vg02/apps 61440000 52722896 8652256 86% /apps > > On the target Linux server, I have 1Gb RAM and 1TB of free space: > > Filesystem 1k-blocks Used Available Use% Mounted on > /dev/tcvg1/tcvol1 1007898916 33408 956667148 1% /vol1 > > I have "apps" configured properly as a service in the rsyncd.conf file. > I started rsync in daemon mode on the HP-UX server, and from the Linux > server I invoked: > > rsync --numeric-ids -vvva root@hpuxsrv::apps /vol1/asic_apps/ > > The file list get transferred, but then I just get repeated lines of errors: > recv_file_name(some_dir/fm_v200209) > opendir(some_dir/fm_v200209): Not enough space > > I have tried with --blocking-io and --no-blocking-io, with --bwlimitXXX. > Same problem each time. > > Any insight is appreciated. > > -- > Bill Geddes > bill_geddes@agilent.com > > -- > > UNIX is user friendly, it's just picky about who its friends are. > -- > To unsubscribe or change options: http://lists.samba.org/mailman/listinfo/rsync > Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html-- Bill Geddes bill_geddes@agilent.com -- UNIX is user friendly, it's just picky about who its friends are.
Even though it was a pain in the *ss, I broke up the transfer into several chunks (logically organized by sub-directories of the source) and wrote a script to batch the job. It worked, better and faster than via NFS. Mounting the source directory via NFS is not a solution that makes one feel good - it ended up with errors. _____ Bill On Fri, 24 Jan 2003 11:54:13 -0700 Bill Geddes <bill_geddes@agilent.com> wrote:> > I am attempting to use rsync to copy a large filesystem from an > HP-UX server to a Linux server with more than enough filespace. > This operation fails. A small directory from the same HP-UX server > can be transfered just as expected. > > The HP-UX server is the source. It has 1Gb RAM - the output of bdf for > the volume the source files is on is: > > Filesystem kbytes used avail %used Mounted on > /dev/vg02/apps 61440000 52722896 8652256 86% /apps > > On the target Linux server, I have 1Gb RAM and 1TB of free space: > > Filesystem 1k-blocks Used Available Use% Mounted on > /dev/tcvg1/tcvol1 1007898916 33408 956667148 1% /vol1 > > I have "apps" configured properly as a service in the rsyncd.conf file. > I started rsync in daemon mode on the HP-UX server, and from the Linux > server I invoked: > > rsync --numeric-ids -vvva root@hpuxsrv::apps /vol1/asic_apps/ > > The file list get transferred, but then I just get repeated lines of errors: > recv_file_name(some_dir/fm_v200209) > opendir(some_dir/fm_v200209): Not enough space > > I have tried with --blocking-io and --no-blocking-io, with --bwlimitXXX. > Same problem each time. > > Any insight is appreciated. > > -- > Bill Geddes > bill_geddes@agilent.com > > -- > > UNIX is user friendly, it's just picky about who its friends are. > -- > To unsubscribe or change options: http://lists.samba.org/mailman/listinfo/rsync > Before posting, read: http://www.tuxedo.org/~esr/faqs/smart-questions.html-- Bill Geddes bill_geddes@agilent.com -- UNIX is user friendly, it's just picky about who its friends are.
Apparently Analagous Threads
- rsync: error writing NNNN unbuffered bytes - exiting:Connection reset by peer
- rsync /somedir work@backups::somearchive/ gets stuck in huge maildirs, rsync /somedir root@backups:/some/path/ works
- [PATCH] fchdir syscall, opendir function
- [PATCH 3/5] COM32: Improve opendir() to deal with no '/' at end of string
- [PATCH 1/1] COM32/opendir: free for malloc