mirror of
git://git.proxmox.com/git/pve-ha-manager.git
synced 2025-02-02 13:47:16 +03:00
fixing typos, also whitespace cleanup in PVE2 env class
fix typos through the whole project, used codespell to find most of them. Also do a big whitespace cleanup in the PVE2 enviorment class. Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
This commit is contained in:
Thomas Lamprecht
Dietmar Maurer
parent
4877c0312a
commit
63f6a08c82
6
README
6
README
@ -41,7 +41,7 @@ The Proxmox 'pmxcfs' implements this on top of corosync.
|
||||
=== Watchdog ===
|
||||
|
||||
We need a reliable watchdog mechanism, which is able to provide hard
|
||||
timeouts. It must be guaranteed that the node reboot withing specified
|
||||
timeouts. It must be guaranteed that the node reboots within the specified
|
||||
timeout if we do not update the watchdog. For me it looks that neither
|
||||
systemd nor the standard watchdog(8) daemon provides such guarantees.
|
||||
|
||||
@ -52,7 +52,7 @@ provides that service to several other daemons using a local socket.
|
||||
|
||||
== Self fencing ==
|
||||
|
||||
A node needs to aquire a special 'ha_agent_${node}_lock' (one separate
|
||||
A node needs to acquire a special 'ha_agent_${node}_lock' (one separate
|
||||
lock for each node) before starting HA resources, and the node updates
|
||||
the watchdog device once it get that lock. If the node loose quorum,
|
||||
or is unable to get the 'ha_agent_${node}_lock', the watchdog is no
|
||||
@ -63,7 +63,7 @@ This makes sure that the node holds the 'ha_agent_${node}_lock' as
|
||||
long as there are running services on that node.
|
||||
|
||||
The HA manger can assume that the watchdog triggered a reboot when he
|
||||
is able to aquire the 'ha_agent_${node}_lock' for that node.
|
||||
is able to acquire the 'ha_agent_${node}_lock' for that node.
|
||||
|
||||
=== Problems with "two_node" Clusters ===
|
||||
|
||||
|
||||
@ -105,14 +105,14 @@ sub log {
|
||||
return $self->{plug}->log($level, @args);
|
||||
}
|
||||
|
||||
# aquire a cluster wide manager lock
|
||||
# acquire a cluster wide manager lock
|
||||
sub get_ha_manager_lock {
|
||||
my ($self) = @_;
|
||||
|
||||
return $self->{plug}->get_ha_manager_lock();
|
||||
}
|
||||
|
||||
# aquire a cluster wide node agent lock
|
||||
# acquire a cluster wide node agent lock
|
||||
sub get_ha_agent_lock {
|
||||
my ($self, $node) = @_;
|
||||
|
||||
|
||||
@ -47,7 +47,7 @@ sub read_manager_status {
|
||||
|
||||
sub write_manager_status {
|
||||
my ($self, $status_obj) = @_;
|
||||
|
||||
|
||||
PVE::HA::Config::write_manager_status($status_obj);
|
||||
}
|
||||
|
||||
@ -63,7 +63,7 @@ sub write_lrm_status {
|
||||
my ($self, $status_obj) = @_;
|
||||
|
||||
my $node = $self->{nodename};
|
||||
|
||||
|
||||
PVE::HA::Config::write_lrm_status($node, $status_obj);
|
||||
}
|
||||
|
||||
@ -81,7 +81,7 @@ sub read_crm_commands {
|
||||
|
||||
sub service_config_exists {
|
||||
my ($self) = @_;
|
||||
|
||||
|
||||
return PVE::HA::Config::resources_config_exists();
|
||||
}
|
||||
|
||||
@ -89,7 +89,7 @@ sub read_service_config {
|
||||
my ($self) = @_;
|
||||
|
||||
my $res = PVE::HA::Config::read_resources_config();
|
||||
|
||||
|
||||
my $vmlist = PVE::Cluster::get_vmlist();
|
||||
my $conf = {};
|
||||
|
||||
@ -116,7 +116,7 @@ sub read_service_config {
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
return $conf;
|
||||
}
|
||||
|
||||
@ -147,7 +147,7 @@ sub get_node_info {
|
||||
my ($self) = @_;
|
||||
|
||||
my ($node_info, $quorate) = ({}, 0);
|
||||
|
||||
|
||||
my $nodename = $self->{nodename};
|
||||
|
||||
$quorate = PVE::Cluster::check_cfs_quorum(1) || 0;
|
||||
@ -156,11 +156,11 @@ sub get_node_info {
|
||||
|
||||
foreach my $node (keys %$members) {
|
||||
my $d = $members->{$node};
|
||||
$node_info->{$node}->{online} = $d->{online};
|
||||
$node_info->{$node}->{online} = $d->{online};
|
||||
}
|
||||
|
||||
|
||||
$node_info->{$nodename}->{online} = 1; # local node is always up
|
||||
|
||||
|
||||
return ($node_info, $quorate);
|
||||
}
|
||||
|
||||
@ -187,7 +187,7 @@ sub get_pve_lock {
|
||||
|
||||
my $retry = 0;
|
||||
my $retry_timeout = 100; # fixme: what timeout
|
||||
|
||||
|
||||
eval {
|
||||
|
||||
mkdir $lockdir;
|
||||
@ -219,15 +219,15 @@ sub get_pve_lock {
|
||||
# $self->log('err', $err) if $err; # for debugging
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
$last_lock_status->{$lockid} = $got_lock ? $ctime : 0;
|
||||
|
||||
if (!!$got_lock != !!$last) {
|
||||
if ($got_lock) {
|
||||
$self->log('info', "successfully aquired lock '$lockid'");
|
||||
$self->log('info', "successfully acquired lock '$lockid'");
|
||||
} else {
|
||||
my $msg = "lost lock '$lockid";
|
||||
$msg .= " - $err" if $err;
|
||||
$msg .= " - $err" if $err;
|
||||
$self->log('err', $msg);
|
||||
}
|
||||
} else {
|
||||
@ -245,7 +245,7 @@ sub get_ha_manager_lock {
|
||||
|
||||
sub get_ha_agent_lock {
|
||||
my ($self, $node) = @_;
|
||||
|
||||
|
||||
$node = $self->nodename() if !defined($node);
|
||||
|
||||
return $self->get_pve_lock("ha_agent_${node}_lock");
|
||||
@ -255,10 +255,10 @@ sub quorate {
|
||||
my ($self) = @_;
|
||||
|
||||
my $quorate = 0;
|
||||
eval {
|
||||
$quorate = PVE::Cluster::check_cfs_quorum();
|
||||
eval {
|
||||
$quorate = PVE::Cluster::check_cfs_quorum();
|
||||
};
|
||||
|
||||
|
||||
return $quorate;
|
||||
}
|
||||
|
||||
@ -290,7 +290,7 @@ sub loop_start_hook {
|
||||
my ($self) = @_;
|
||||
|
||||
PVE::Cluster::cfs_update();
|
||||
|
||||
|
||||
$self->{loop_start} = $self->get_time();
|
||||
}
|
||||
|
||||
@ -298,7 +298,7 @@ sub loop_end_hook {
|
||||
my ($self) = @_;
|
||||
|
||||
my $delay = $self->get_time() - $self->{loop_start};
|
||||
|
||||
|
||||
warn "loop take too long ($delay seconds)\n" if $delay > 30;
|
||||
}
|
||||
|
||||
@ -313,7 +313,7 @@ sub watchdog_open {
|
||||
Type => SOCK_STREAM(),
|
||||
Peer => "/run/watchdog-mux.sock") ||
|
||||
die "unable to open watchdog socket - $!\n";
|
||||
|
||||
|
||||
$self->log('info', "watchdog active");
|
||||
}
|
||||
|
||||
@ -367,15 +367,15 @@ sub exec_resource_agent {
|
||||
my ($self, $sid, $service_config, $cmd, @params) = @_;
|
||||
|
||||
# setup execution environment
|
||||
|
||||
|
||||
$ENV{'PATH'} = '/sbin:/bin:/usr/sbin:/usr/bin';
|
||||
|
||||
PVE::INotify::inotify_close();
|
||||
|
||||
|
||||
PVE::INotify::inotify_init();
|
||||
|
||||
PVE::Cluster::cfs_update();
|
||||
|
||||
|
||||
my $nodename = $self->{nodename};
|
||||
|
||||
# fixme: return valid_exit code (instead of using die) ?
|
||||
@ -473,8 +473,7 @@ sub exec_resource_agent {
|
||||
|
||||
} elsif ($cmd eq 'error') {
|
||||
|
||||
|
||||
if($running) {
|
||||
if ($running) {
|
||||
$self->log("err", "service $sid is in an error state while running");
|
||||
} else {
|
||||
$self->log("warning", "service $sid is not running and in an error state");
|
||||
|
||||
@ -379,7 +379,7 @@ sub manage {
|
||||
|
||||
next if !$fenced_nodes->{$sd->{node}};
|
||||
|
||||
# node fence was sucessful - mark service as stopped
|
||||
# node fence was successful - mark service as stopped
|
||||
&$change_service_state($self, $sid, 'stopped');
|
||||
}
|
||||
|
||||
|
||||
@ -315,7 +315,7 @@ sub global_lock {
|
||||
}
|
||||
if (!$success) {
|
||||
close($fh);
|
||||
die "can't aquire lock '$lockfile' - $!\n";
|
||||
die "can't acquire lock '$lockfile' - $!\n";
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@ -53,7 +53,7 @@ sub pve_verify_ha_group_node {
|
||||
}
|
||||
|
||||
PVE::JSONSchema::register_standard_option('pve-ha-group-node-list', {
|
||||
description => "List of cluster node names with optional priority. We use priority '0' as default. The CRM tries to run services on the node with higest priority (also see option 'nofailback').",
|
||||
description => "List of cluster node names with optional priority. We use priority '0' as default. The CRM tries to run services on the node with highest priority (also see option 'nofailback').",
|
||||
type => 'string', format => 'pve-ha-group-node-list',
|
||||
typetext => '<node>[:<pri>]{,<node>[:<pri>]}*',
|
||||
});
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user