Questions[master](server)

I have an Ubuntu machine (18.04) with PTPd installed. I want to write a script that can determine what PTP settings have been set previously.

Specifically, I want to find information on the master PTP server that this machine is locked to (if it is locked at all), such as its IP address.

As far as I can tell, PTPd does not store that sort of info in a conf file anywhere (in fact, my PTPd doesn't even have a conf file because I haven't created one for it); if it did, I could parse that conf file for the information.

Is there an existing command, or a simple script that can get this information for me?

I have a PowerDNS 3.3 server installed on Ubuntu 14.04 (for testing).
It should be master only. No slave exists.

I keep on getting this error:

Received NOTIFY for mydomain.com from 82.91.23.176 but slave support is disabled in the configuration
Received unsuccessful notification report for 'mydomain.com' from 82.91.23.176:53, rcode: 4

82.91.23.176 (modified in this post) is the WAN IP of the PowerDNS server.

Here my pdns.conf

allow-axfr-ips=127.0.0.1
allow-recursion=127.0.0.1
config-dir=/etc/powerdns
daemon=yes
disable-axfr=no
guardian=yes
include-dir=/etc/powerdns/pdns.d
local-address=0.0.0.0
local-port=53
master=yes
module-dir=/usr/lib/powerdns
setgid=pdns
setuid=pdns
socket-dir=/var/run
version-string=powerdns
webserver=yes
webserver-address=192.168.178.33

To my understanding, slave=no is per default and no slave is allowed for axfr, so PowerDNS should not receive any notification at all.

Or what is this message about?

I have the following setup:

M1 = Master 1
S1 = Slave 1
S2 = Slave 2
S3 = Slave 3

Where all slaves are replicating directly from M1. I would like to move S2 and S3 to be slaves of S1.

I'm not sure about how to do it without any downtime and with replication picking up where it left off. How to I ensure that the slaves continue from the correct point in the logs from their new master without missing any writes or trying to repeat writes that they had already done before the change.

One of the slaves is at an offsite location on a relatively slow connection, and it's quite a large DB, so rebuilding the slaves from a dump or copying data files would take too long, and seems to me as though it should be unnecessary!

Any help or advice greatly appreciated!

Simon

I am writing code to extract the IP address and username of all slave servers in a MySQL replication environment.

Does anyone know any function, variable, or something else that I can run in the console and retrieve these infos without looking into the my.cnf file?

After setting up mysql replications with master/slave db scheme, I noticed that the slave db server is not read-only. Of cause, after that I configured it manually in /etc/my.cnf file.

I just can't understand: isn't it the default behavior that the slaves are always running in the read-only mode or it should be always configured by hand?