|
I have a question about keepalived service checks for rsync on the real
servers. Each keepalived tcp check to port 873 causes these messages
every 2 seconds in the rsyncd log:
2003/05/25 21:17:49 [29016] rsync: writefd_unbuffered failed to write 12
bytes: phase "unknown": Connection reset by peer
2003/05/25 21:17:49 [29016] rsync error: error in rsync protocol data
stream (code 12) at io.c(515)
The volume of messages is a problem. Has anyone set up an indirect
check for a service to avoid these connection issues? According to the
keepalived documentation, it should be difficult. Apache is also
running on the real servers, so a cron-job-generated web page would work
for an http check. I'm curious, though, if someone has suggestions.
Thanks,
-John R.
|
