mirror of
https://github.com/percona/percona-toolkit.git
synced 2025-09-25 21:54:48 +00:00
Change mk- to pt- in all tools.
This commit is contained in:
Daniel Nichter
@@ -2297,7 +2297,7 @@ sub info {
|
||||
: print ts($now), " ", $message, "\n";
|
||||
}
|
||||
|
||||
# Catches signals so mk-slave-delay can exit gracefully.
|
||||
# Catches signals so pt-slave-delay can exit gracefully.
|
||||
sub finish {
|
||||
my ($signal) = @_;
|
||||
print STDERR "Exiting on SIG$signal.\n";
|
||||
@@ -2342,19 +2342,19 @@ if ( !caller ) { exit main(@ARGV); }
|
||||
|
||||
=head1 NAME
|
||||
|
||||
mk-slave-delay - Make a MySQL slave server lag behind its master.
|
||||
pt-slave-delay - Make a MySQL slave server lag behind its master.
|
||||
|
||||
=head1 SYNOPSIS
|
||||
|
||||
Usage: mk-slave-delay [OPTION...] SLAVE-HOST [MASTER-HOST]
|
||||
Usage: pt-slave-delay [OPTION...] SLAVE-HOST [MASTER-HOST]
|
||||
|
||||
mk-slave-delay starts and stops a slave server as needed to make it lag
|
||||
pt-slave-delay starts and stops a slave server as needed to make it lag
|
||||
behind the master. The SLAVE-HOST and MASTER-HOST use DSN syntax, and
|
||||
values are copied from the SLAVE-HOST to the MASTER-HOST if omitted.
|
||||
|
||||
To hold slavehost one minute behind its master for ten minutes:
|
||||
|
||||
mk-slave-delay --delay 1m --interval 15s --run-time 10m slavehost
|
||||
pt-slave-delay --delay 1m --interval 15s --run-time 10m slavehost
|
||||
|
||||
=head1 RISKS
|
||||
|
||||
@@ -2363,7 +2363,7 @@ whether known or unknown, of using this tool. The two main categories of risks
|
||||
are those created by the nature of the tool (e.g. read-only tools vs. read-write
|
||||
tools) and those created by bugs.
|
||||
|
||||
mk-slave-delay is generally very low-risk. It simply starts and stops the
|
||||
pt-slave-delay is generally very low-risk. It simply starts and stops the
|
||||
replication SQL thread. This might cause monitoring systems to think the slave
|
||||
is having trouble.
|
||||
|
||||
@@ -2373,13 +2373,13 @@ users.
|
||||
The authoritative source for updated information is always the online issue
|
||||
tracking system. Issues that affect this tool will be marked as such. You can
|
||||
see a list of such issues at the following URL:
|
||||
L<http://www.maatkit.org/bugs/mk-slave-delay>.
|
||||
L<http://www.maatkit.org/bugs/pt-slave-delay>.
|
||||
|
||||
See also L<"BUGS"> for more information on filing bugs and getting help.
|
||||
|
||||
=head1 DESCRIPTION
|
||||
|
||||
C<mk-slave-delay> watches a slave and starts and stops its replication SQL
|
||||
C<pt-slave-delay> watches a slave and starts and stops its replication SQL
|
||||
thread as necessary to hold it at least as far behind the master as you
|
||||
request. In practice, it will typically cause the slave to lag between
|
||||
L<"--delay"> and L<"--delay">+L<"--interval"> behind the master.
|
||||
@@ -2388,11 +2388,11 @@ It bases the delay on binlog positions in the slave's relay logs by default,
|
||||
so there is no need to connect to the master. This works well if the IO
|
||||
thread doesn't lag the master much, which is typical in most replication
|
||||
setups; the IO thread lag is usually milliseconds on a fast network. If your
|
||||
IO thread's lag is too large for your purposes, C<mk-slave-delay> can also
|
||||
IO thread's lag is too large for your purposes, C<pt-slave-delay> can also
|
||||
connect to the master for information about binlog positions.
|
||||
|
||||
If the slave's I/O thread reports that it is waiting for the SQL thread to
|
||||
free some relay log space, C<mk-slave-delay> will automatically connect to the
|
||||
free some relay log space, C<pt-slave-delay> will automatically connect to the
|
||||
master to find binary log positions. If L<"--ask-pass"> and L<"--daemonize">
|
||||
are given, it is possible that this could cause it to ask for a password while
|
||||
daemonized. In this case, it exits. Therefore, if you think your slave might
|
||||
@@ -2401,23 +2401,23 @@ L<"--use-master"> explicitly when daemonizing, or don't specify L<"--ask-pass">.
|
||||
|
||||
The SLAVE-HOST and optional MASTER-HOST are both DSNs. See L<"DSN OPTIONS">.
|
||||
Missing MASTER-HOST values are filled in with values from SLAVE-HOST, so you
|
||||
don't need to specify them in both places. C<mk-slave-delay> reads all normal
|
||||
don't need to specify them in both places. C<pt-slave-delay> reads all normal
|
||||
MySQL option files, such as ~/.my.cnf, so you may not need to specify username,
|
||||
password and other common options at all.
|
||||
|
||||
C<mk-slave-delay> tries to exit gracefully by trapping signals such as Ctrl-C.
|
||||
C<pt-slave-delay> tries to exit gracefully by trapping signals such as Ctrl-C.
|
||||
You cannot bypass L<"--[no]continue"> with a trappable signal.
|
||||
|
||||
=head1 PRIVILEGES
|
||||
|
||||
mk-slave-delay requires the following privileges: PROCESS, REPLICATION CLIENT,
|
||||
pt-slave-delay requires the following privileges: PROCESS, REPLICATION CLIENT,
|
||||
and SUPER.
|
||||
|
||||
=head1 OUTPUT
|
||||
|
||||
If you specify L<"--quiet">, there is no output. Otherwise, the normal output
|
||||
is a status message consisting of a timestamp and information about what
|
||||
C<mk-slave-delay> is doing: starting the slave, stopping the slave, or just
|
||||
C<pt-slave-delay> is doing: starting the slave, stopping the slave, or just
|
||||
observing.
|
||||
|
||||
=head1 OPTIONS
|
||||
@@ -2455,7 +2455,7 @@ default: yes
|
||||
Continue replication normally on exit. After exiting, restart the slave's SQL
|
||||
thread with no UNTIL condition, so it will run as usual and catch up to the
|
||||
master. This is enabled by default and works even if you terminate
|
||||
C<mk-slave-delay> with Control-C.
|
||||
C<pt-slave-delay> with Control-C.
|
||||
|
||||
=item --daemonize
|
||||
|
||||
@@ -2489,7 +2489,7 @@ Connect to host.
|
||||
|
||||
type: time; default: 1m
|
||||
|
||||
How frequently C<mk-slave-delay> should check whether the slave needs to be
|
||||
How frequently C<pt-slave-delay> should check whether the slave needs to be
|
||||
started or stopped.
|
||||
|
||||
=item --log
|
||||
@@ -2530,7 +2530,7 @@ Don't print informational messages about operation. See L<OUTPUT> for details.
|
||||
|
||||
type: time
|
||||
|
||||
How long C<mk-slave-delay> should run before exiting. The default is to run
|
||||
How long C<pt-slave-delay> should run before exiting. The default is to run
|
||||
forever.
|
||||
|
||||
=item --set-vars
|
||||
@@ -2551,10 +2551,10 @@ Socket file to use for connection.
|
||||
Get binlog positions from master, not slave. Don't trust the binlog positions
|
||||
in the slave's relay log. Connect to the master and get binlog positions
|
||||
instead. If you specify this option without giving a MASTER-HOST on the command
|
||||
line, C<mk-slave-delay> examines the slave's SHOW SLAVE STATUS to determine the
|
||||
line, C<pt-slave-delay> examines the slave's SHOW SLAVE STATUS to determine the
|
||||
hostname and port for connecting to the master.
|
||||
|
||||
C<mk-slave-delay> uses only the MASTER_HOST and MASTER_PORT values from SHOW
|
||||
C<pt-slave-delay> uses only the MASTER_HOST and MASTER_PORT values from SHOW
|
||||
SLAVE STATUS for the master connection. It does not use the MASTER_USER
|
||||
value. If you want to specify a different username for the master than the
|
||||
one you use to connect to the slave, you should specify the MASTER-HOST option
|
||||
@@ -2661,7 +2661,7 @@ installed in any reasonably new version of Perl.
|
||||
|
||||
=head1 BUGS
|
||||
|
||||
For a list of known bugs see L<http://www.maatkit.org/bugs/mk-slave-delay>.
|
||||
For a list of known bugs see L<http://www.maatkit.org/bugs/pt-slave-delay>.
|
||||
|
||||
Please use Google Code Issues and Groups to report bugs or request support:
|
||||
L<http://code.google.com/p/maatkit/>. You can also join #maatkit on Freenode to
|
||||
|
||||
Reference in New Issue
Block a user