Hi all, I've been trying to figure out what's wrong with my dates but I just can't figure out where the problem lies. Both source and destination box are set at the right time via ntp. I tried to split the logs so I could have a "per module log" with syslog-ng but I'm not quite there yet. Still, this is what I get when an rsync is run... The time starts right 23:02 but then I lose an hour. /var/log/rsyncd/rsync.log Nov 15 23:02:49 mysrvr rsyncd[27827]: 192.168.2.16 is not a known address for "test.domain.com": spoofed address? Nov 15 23:02:49 mysrvr rsyncd[27827]: connect from UNKNOWN (192.168.2.16) /var/log/rsyncd/foo.log Nov 15 22:05:55 mysrvr rsyncd[27827]:... Nov 15 22:05:55 mysrvr rsyncd[27827]: ... /var/log/rsyncd/rsync.log Nov 15 22:05:55 mysrvr rsyncd[27827]: sent 69 bytes received 3931999 bytes total size 4146898508 Does anyone have a clue what the reason could be? Is it staring me in the face and I can't see it? Thank you all for your feedback. -Ed
On Thu, Nov 16, 2006 at 12:17:32AM +0100, Ed wrote:> The time starts right 23:02 but then I lose an hour.That's caused by the chroot() call calling your system library away from the timezone routines. In 2.6.9, rsync calls strftime() once prior to the chroot(), so it should hopefully avoid this problem. If it doesn't, you may need to set a TZ environment varible, or copy the necessary timezone files into each chroot() area that will make your system libraries happy. ..wayne..
Seemingly Similar Threads
- rsync 2.5.1 server - strange logs
- Timezone error with chrooted rsync version 2.6.3pre1
- [Bug 8198] New: rsync does not set TZ variable after chroot(), which confuses logging timestamps
- Rsync 2.5.1 on Solaris 8 reverse lookup failures
- Conflicting timestamps in rsyncd.log