Hi all. I'm just reporting some strange (fatal) behaviour when running rsync --daemon on an IPv6-enabled box. It gets a fatal error when trying to bind to the same address twice. strace output: bind(4, {sa_family=AF_INET6, sin6_port=htons(873), inet_pton(AF_INET6, "::", &sin6_addr), sin6_flowinfo=0, sin6_scope_id=0}, 28) = 0 bind(5, {sa_family=AF_INET, sin_port=htons(873), sin_addr=inet_addr("0.0.0.0")}, 16) = 0 listen(4, 5) = 0 listen(5, 5) = -1 EADDRINUSE (Address already in use) The workaround is very simple; use --address 0.0.0.0 (or "::") to keep it from binding twice. Thanks, Vidar
On Mon, Jun 07, 2004 at 12:23:00AM +0200, Vidar Madsen wrote:> I'm just reporting some strange (fatal) behaviour when running > rsync --daemon on an IPv6-enabled box.This should only affect older systems that don't have a proper IPv6 implementation, such as some older 2.4 Linux kernels. The fix that I suggested earlier is to use either the -6 or the -4 option to tell rsync to only bind only one of the duplicitous addresses. ..wayne..
Possibly Parallel Threads
- [Bug 1071] New: nftables: set does not work within inet table with option flags interval
- Linux 4.2p1 crash during reverse name lookup
- rsync failure with error 12
- Bug#793921: tftpd-hpa: IPv6 address cannonization breaks IPv4
- [Bug 1950] New: sshd tries to bind over and over to ::1 for several seconds