Blair Zajac
2011-May-15 19:31 UTC
rsync client and server processes, all hanging in read_timeout()
I'm backing my Ubuntu server up to an HFS+ case-sensitive volume on our iMac using the below rsync command and it hangs at the end. Compiling rsync 3.0.8 by hand with CFLAGS=-g on both systems shows that all rsync processes are in read_timeout(). $ rsync \ --delete-delay \ -P \ -v \ -A \ -X \ --fake-super \ -aH \ --numeric-ids \ --exclude '/dev' \ --exclude '/home/OLD_OS' \ --exclude '/media' \ --exclude '/mnt' \ --exclude '/proc' \ --exclude '/sys' \ --exclude '/tmp/blair/.a' \ --exclude '/tmp/gsrvdir**' \ --exclude '/tmp/orbit-**' \ --exclude '/tmp/ssh-**' \ --exclude '/usr/lib/Adobe/Acrobat7.0/Reader/HowTo' \ --exclude '/var/lib/mysql' \ --exclude '/var/lib/postgresql' \ --exclude '/var/lib/vmware' \ --exclude '/var/lib/mysql' \ --exclude '/var/log/mysql' \ --exclude '/var/run/postgresql' \ root at 192.168.1.206:/ \ backup-dir/ The last output shows it's working on symlinks. These symlinks haven't changed in a while, so it's odd they are showing up here. var/svn/hot_backups/foobar-www-8/hooks/pre-unlock => var/svn/ hot_backups/foobar-www-5/hooks/pre-unlock var/svn/hot_backups/foobar-www-9/hooks/ var/svn/hot_backups/foobar-www-9/hooks/post-commit => var/svn/ hot_backups/foobar-www-6/hooks/post-commit var/svn/hot_backups/foobar-www-9/hooks/post-lock => var/svn/ hot_backups/foobar-www-6/hooks/post-lock var/svn/hot_backups/foobar-www-9/hooks/post-revprop-change => var/svn/ hot_backups/foobar-www-6/hooks/post-revprop-change var/svn/hot_backups/foobar-www-9/hooks/post-unlock => var/svn/ hot_backups/foobar-www-6/hooks/post-unlock var/svn/hot_backups/foobar-www-9/hooks/pre-commit => var/svn/ hot_backups/foobar-www-5/hooks/pre-commit var/svn/hot_backups/foobar-www-9/hooks/pre-lock => var/svn/hot_backups/ foobar-www-5/hooks/pre-lock var/svn/hot_backups/foobar-www-9/hooks/pre-revprop-change => var/svn/ hot_backups/foobar-www-5/hooks/pre-revprop-change var/svn/hot_backups/foobar-www-9/hooks/pre-unlock => var/svn/ hot_backups/foobar-www-5/hooks/pre-unlock Stack trace on the iMac for the parent rsync process: (gdb) bt #0 0x00007fff87a86e52 in select$DARWIN_EXTSN () #1 0x000000010002eaad in read_timeout (fd=3, buf=0x100800000 "\004", len=8184) at io.c:654 #2 0x000000010002f891 in readfd_unbuffered (fd=3, buf=0x7fff5fbfdee0 "", len=4) at io.c:1012 #3 0x000000010002ffae in readfd (fd=3, buffer=0x7fff5fbfdee0 "", N=4) at io.c:1157 #4 0x000000010002da8c in read_msg_fd () at io.c:338 #5 0x000000010002e7b4 in wait_for_receiver () at io.c:540 #6 0x00000001000132f8 in generate_files (f_out=4, local_name=0x0) at generator.c:2290 #7 0x0000000100020f48 in do_recv (f_in=5, f_out=4, local_name=0x0) at main.c:833 #8 0x000000010002179f in client_run (f_in=5, f_out=4, pid=64758, argc=1, argv=0x100100398) at main.c:1070 #9 0x00000001000220ff in start_client (argc=1, argv=0x100100398) at main.c:1277 #10 0x00000001000227ea in main (argc=2, argv=0x100100390) at main.c:1504 Stack trace on the iMac for the child rsync process (gdb) bt #0 0x00007fff87a86e52 in select$DARWIN_EXTSN () #1 0x000000010002eaad in read_timeout (fd=5, buf=0x7fff5fbfc420 "6", len=4) at io.c:654 #2 0x000000010002f7c5 in read_loop (fd=5, buf=0x7fff5fbfc420 "6", len=4) at io.c:989 #3 0x000000010002f989 in readfd_unbuffered (fd=5, buf=0x7fff5fbfd8a0 "", len=1) at io.c:1026 #4 0x000000010002ffae in readfd (fd=5, buffer=0x7fff5fbfd8a0 "", N=1) at io.c:1157 #5 0x00000001000319c8 in read_ndx (f=5) at io.c:1787 #6 0x000000010000ae74 in read_ndx_and_attrs (f_in=5, iflag_ptr=0x7fff5fbfdb90, type_ptr=0x7fff5fbfdb9f "?", buf=0x7fff5fbfe3a0 "var/svn/hot_backups/foobar-www-5/hooks/pre- unlock", len_ptr=0x7fff5fbfdb8c) at rsync.c:257 #7 0x000000010001449f in recv_files (f_in=5, local_name=0x0) at receiver.c:455 #8 0x0000000100020d95 in do_recv (f_in=5, f_out=4, local_name=0x0) at main.c:777 #9 0x000000010002179f in client_run (f_in=5, f_out=4, pid=64758, argc=1, argv=0x100100398) at main.c:1070 #10 0x00000001000220ff in start_client (argc=1, argv=0x100100398) at main.c:1277 #11 0x00000001000227ea in main (argc=2, argv=0x100100390) at main.c:1504 Stack trace on the Ubuntu system: (gdb) bt #0 0x00d2e416 in __kernel_vsyscall () #1 0x00a9f33d in ___newselect_nocancel () at ../sysdeps/unix/syscall-template.S:82 #2 0x08073879 in read_timeout (fd=0, buf=0xbf85ddec "I\005", len=4) at io.c:654 #3 0x0807461e in read_loop (fd=0, buf=0xbf85ddec "I\005", len=4) at io.c:989 #4 0x08074761 in readfd_unbuffered (fd=0, buf=0xbf85f444 "\016", len=1) at io.c:1026 #5 0x08074deb in readfd (fd=0, buffer=0xbf85f444 "\016", N=1) at io.c: 1157 #6 0x080764e3 in read_ndx (f=0) at io.c:1787 #7 0x08053888 in read_ndx_and_attrs (f_in=0, iflag_ptr=0xbf85f5c0, type_ptr=0xbf85f5db "\200var/svn/hot_backups/foobar-www-5/hooks/ pre-unlock", buf=0xbf85f5dc "var/svn/hot_backups/foobar-www-5/hooks/ pre-unlock", len_ptr=0xbf85f5bc) at rsync.c:257 #8 0x0805e5fa in send_files (f_in=0, f_out=1) at sender.c:187 #9 0x080671c0 in do_server_sender (f_in=0, f_out=1, argc=1, argv=0x94339c4) at main.c:726 #10 0x08067a5c in start_server (f_in=0, f_out=1, argc=2, argv=0x94339c0) at main.c:971 #11 0x08068c7b in main (argc=2, argv=0x94339c0) at main.c:1501 Any suggestions? Should I file a bug report? Blair
Seemingly Similar Threads
- Subversion causing Apache to hog memory
- [LLVMdev] Request: Changing commit message in llvm trunk
- [LLVMdev] Request: Changing commit message in llvm trunk
- [LLVMdev] Request: Changing commit message in llvm trunk
- [nut-commits] svn commit r2504 - in branches/AsciiDoc/docs: . website