1
0
mirror of https://github.com/samba-team/samba.git synced 2024-12-24 21:34:56 +03:00

Merge commit 'origin/master'

(This used to be ctdb commit 135b72828fc76856fa8f6d7f9c820120de05596b)
This commit is contained in:
Martin Schwenke 2009-06-02 16:29:25 +10:00
commit 1c2e7871eb
15 changed files with 269 additions and 72 deletions

View File

@ -1,11 +1,11 @@
.\" Title: ctdb
.\" Author:
.\" Generator: DocBook XSL Stylesheets v1.73.2 <http://docbook.sf.net/>
.\" Date: 03/24/2009
.\" Date: 06/02/2009
.\" Manual:
.\" Source:
.\"
.TH "CTDB" "1" "03/24/2009" "" ""
.TH "CTDB" "1" "06/02/2009" "" ""
.\" disable hyphenation
.nh
.\" disable justification (adjust text to left margin only)
@ -469,9 +469,9 @@ This command will kill the specified TCP connection by issuing a TCP RST to the
This command will send out a gratious arp for the specified interface through the specified interface\. This command is mainly used by the ctdb eventscripts\.
.SS "reloadnodes"
.PP
This command is used when adding new nodes to an existing cluster and to reduce the disruption of this operation\. This command should never be used except when expanding an existing cluster\. This can only be used to expand a cluster\. To remove a node from the cluster you still need to shut down ctdb on all nodes, edit the nodes file and restart ctdb\.
This command is used when adding new nodes, or removing existing nodes from an existing cluster\.
.PP
Procedure:
Procedure to add a node:
.PP
1, To expand an existing cluster, first ensure with \'ctdb status\' that all nodes are up and running and that they are all healthy\. Do not try to expand a cluster unless it is completely healthy!
.PP
@ -479,11 +479,24 @@ Procedure:
.PP
3, Verify that all the nodes have identical /etc/ctdb/nodes files after you edited them and added the new node!
.PP
4, Run \'ctdb reloadnodes\' to force all nodes to reaload the nodesfile\.
4, Run \'ctdb reloadnodes\' to force all nodes to reload the nodesfile\.
.PP
5, Use \'ctdb status\' on all nodes and verify that they now show the additional node\.
.PP
6, Install and configure the new node and bring it online\.
.PP
Procedure to remove a node:
.PP
1, To remove a node from an existing cluster, first ensure with \'ctdb status\' that all nodes, except the node to be deleted, are up and running and that they are all healthy\. Do not try to remove a node from a cluster unless the cluster is completely healthy!
.PP
2, Shutdown and poerwoff the node to be removed\.
.PP
3, On all other nodes, edit the /etc/ctdb/nodes file and comment out the node to be removed\. Do not delete the line for that node, just comment it out by adding a \'#\' at the beginning of the line\.
.PP
4, Run \'ctdb reloadnodes\' to force all nodes to reload the nodesfile\.
.PP
5, Use \'ctdb status\' on all nodes and verify that the deleted node no longer shows up in the list\.\.
.PP
.SS "tickle <srcip:port> <dstip:port>"
.PP
This command will will send a TCP tickle to the source host for the specified TCP connection\. A TCP tickle is a TCP ACK packet with an invalid sequence and acknowledge number and will when received by the source host result in it sending an immediate correct ACK back to the other end\.

View File

@ -367,14 +367,9 @@ CTDB version 1
through the specified interface. This command is mainly used by the
ctdb eventscripts.
</p></div><div class="refsect2" lang="en"><a name="id2529124"></a><h3>reloadnodes</h3><p>
This command is used when adding new nodes to an existing cluster and
to reduce the disruption of this operation. This command should never
be used except when expanding an existing cluster.
This can only be used to expand a cluster. To remove a node from the
cluster you still need to shut down ctdb on all nodes, edit the nodes file
and restart ctdb.
This command is used when adding new nodes, or removing existing nodes from an existing cluster.
</p><p>
Procedure:
Procedure to add a node:
</p><p>
1, To expand an existing cluster, first ensure with 'ctdb status' that
all nodes are up and running and that they are all healthy.
@ -385,12 +380,27 @@ CTDB version 1
</p><p>
3, Verify that all the nodes have identical /etc/ctdb/nodes files after you edited them and added the new node!
</p><p>
4, Run 'ctdb reloadnodes' to force all nodes to reaload the nodesfile.
4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
</p><p>
5, Use 'ctdb status' on all nodes and verify that they now show the additional node.
</p><p>
6, Install and configure the new node and bring it online.
</p></div><div class="refsect2" lang="en"><a name="id2529176"></a><h3>tickle &lt;srcip:port&gt; &lt;dstip:port&gt;</h3><p>
</p><p>
Procedure to remove a node:
</p><p>
1, To remove a node from an existing cluster, first ensure with 'ctdb status' that
all nodes, except the node to be deleted, are up and running and that they are all healthy.
Do not try to remove a node from a cluster unless the cluster is completely healthy!
</p><p>
2, Shutdown and poerwoff the node to be removed.
</p><p>
3, On all other nodes, edit the /etc/ctdb/nodes file and comment out the node to be removed. Do not delete the line for that node, just comment it out by adding a '#' at the beginning of the line.
</p><p>
4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
</p><p>
5, Use 'ctdb status' on all nodes and verify that the deleted node no longer shows up in the list..
</p><p>
</p></div><div class="refsect2" lang="en"><a name="id2529208"></a><h3>tickle &lt;srcip:port&gt; &lt;dstip:port&gt;</h3><p>
This command will will send a TCP tickle to the source host for the
specified TCP connection.
A TCP tickle is a TCP ACK packet with an invalid sequence and
@ -402,10 +412,10 @@ CTDB version 1
TCP connection has been disrupted and that the client will need
to reestablish. This greatly speeds up the time it takes for a client
to detect and reestablish after an IP failover in the ctdb cluster.
</p></div><div class="refsect2" lang="en"><a name="id2529200"></a><h3>gettickles &lt;ip&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529233"></a><h3>gettickles &lt;ip&gt;</h3><p>
This command is used to show which TCP connections are registered with
CTDB to be "tickled" if there is a failover.
</p></div><div class="refsect2" lang="en"><a name="id2529212"></a><h3>repack [max_freelist]</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529244"></a><h3>repack [max_freelist]</h3><p>
Over time, when records are created and deleted in a TDB, the TDB list of free space will become fragmented. This can lead to a slowdown in accessing TDB records.
This command is used to defragment a TDB database and pruning the freelist.
</p><p>
@ -420,7 +430,7 @@ CTDB version 1
Example: ctdb repack 1000
</p><p>
By default, this operation is issued from the 00.ctdb event script every 5 minutes.
</p></div><div class="refsect2" lang="en"><a name="id2529258"></a><h3>vacuum [max_records]</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529290"></a><h3>vacuum [max_records]</h3><p>
Over time CTDB databases will fill up with empty deleted records which will lead to a progressive slow down of CTDB database access.
This command is used to prune all databases and delete all empty records from the cluster.
</p><p>
@ -436,12 +446,12 @@ CTDB version 1
Example: ctdb vacuum
</p><p>
By default, this operation is issued from the 00.ctdb event script every 5 minutes.
</p></div></div><div class="refsect1" lang="en"><a name="id2529295"></a><h2>Debugging Commands</h2><p>
</p></div></div><div class="refsect1" lang="en"><a name="id2529327"></a><h2>Debugging Commands</h2><p>
These commands are primarily used for CTDB development and testing and
should not be used for normal administration.
</p><div class="refsect2" lang="en"><a name="id2529305"></a><h3>process-exists &lt;pid&gt;</h3><p>
</p><div class="refsect2" lang="en"><a name="id2529337"></a><h3>process-exists &lt;pid&gt;</h3><p>
This command checks if a specific process exists on the CTDB host. This is mainly used by Samba to check if remote instances of samba are still running or not.
</p></div><div class="refsect2" lang="en"><a name="id2529317"></a><h3>getdbmap</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529350"></a><h3>getdbmap</h3><p>
This command lists all clustered TDB databases that the CTDB daemon has attached to. Some databases are flagged as PERSISTENT, this means that the database stores data persistently and the data will remain across reboots. One example of such a database is secrets.tdb where information about how the cluster was joined to the domain is stored.
</p><p>
Most databases are not persistent and only store the state information that the currently running samba daemons need. These databases are always wiped when ctdb/samba starts and when a node is rebooted.
@ -460,25 +470,25 @@ dbid:0x2672a57f name:idmap2.tdb path:/var/ctdb/persistent/idmap2.tdb.0 PERSISTEN
dbid:0xb775fff6 name:secrets.tdb path:/var/ctdb/persistent/secrets.tdb.0 PERSISTENT
dbid:0xe98e08b6 name:group_mapping.tdb path:/var/ctdb/persistent/group_mapping.tdb.0 PERSISTENT
dbid:0x7bbbd26c name:passdb.tdb path:/var/ctdb/persistent/passdb.tdb.0 PERSISTENT
</pre></div><div class="refsect2" lang="en"><a name="id2529364"></a><h3>catdb &lt;dbname&gt;</h3><p>
</pre></div><div class="refsect2" lang="en"><a name="id2529396"></a><h3>catdb &lt;dbname&gt;</h3><p>
This command will dump a clustered TDB database to the screen. This is a debugging command.
</p></div><div class="refsect2" lang="en"><a name="id2529374"></a><h3>getmonmode</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529407"></a><h3>getmonmode</h3><p>
This command returns the monutoring mode of a node. The monitoring mode is either ACTIVE or DISABLED. Normally a node will continously monitor that all other nodes that are expected are in fact connected and that they respond to commands.
</p><p>
ACTIVE - This is the normal mode. The node is actively monitoring all other nodes, both that the transport is connected and also that the node responds to commands. If a node becomes unavailable, it will be marked as DISCONNECTED and a recovery is initiated to restore the cluster.
</p><p>
DISABLED - This node is not monitoring that other nodes are available. In this mode a node failure will not be detected and no recovery will be performed. This mode is useful when for debugging purposes one wants to attach GDB to a ctdb process but wants to prevent the rest of the cluster from marking this node as DISCONNECTED and do a recovery.
</p></div><div class="refsect2" lang="en"><a name="id2529405"></a><h3>setmonmode &lt;0|1&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529438"></a><h3>setmonmode &lt;0|1&gt;</h3><p>
This command can be used to explicitely disable/enable monitoring mode on a node. The main purpose is if one wants to attach GDB to a running ctdb daemon but wants to prevent the other nodes from marking it as DISCONNECTED and issuing a recovery. To do this, set monitoring mode to 0 on all nodes before attaching with GDB. Remember to set monitoring mode back to 1 afterwards.
</p></div><div class="refsect2" lang="en"><a name="id2529421"></a><h3>attach &lt;dbname&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529453"></a><h3>attach &lt;dbname&gt;</h3><p>
This is a debugging command. This command will make the CTDB daemon create a new CTDB database and attach to it.
</p></div><div class="refsect2" lang="en"><a name="id2529432"></a><h3>dumpmemory</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529464"></a><h3>dumpmemory</h3><p>
This is a debugging command. This command will make the ctdb
daemon to write a fill memory allocation map to standard output.
</p></div><div class="refsect2" lang="en"><a name="id2529444"></a><h3>rddumpmemory</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529476"></a><h3>rddumpmemory</h3><p>
This is a debugging command. This command will dump the talloc memory
allocation tree for the recovery daemon to standard output.
</p></div><div class="refsect2" lang="en"><a name="id2529455"></a><h3>freeze</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529487"></a><h3>freeze</h3><p>
This command will lock all the local TDB databases causing clients
that are accessing these TDBs such as samba3 to block until the
databases are thawed.
@ -486,20 +496,20 @@ dbid:0x7bbbd26c name:passdb.tdb path:/var/ctdb/persistent/passdb.tdb.0 PERSISTEN
This is primarily used by the recovery daemon to stop all samba
daemons from accessing any databases while the database is recovered
and rebuilt.
</p></div><div class="refsect2" lang="en"><a name="id2529473"></a><h3>thaw</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529506"></a><h3>thaw</h3><p>
Thaw a previously frozen node.
</p></div><div class="refsect2" lang="en"><a name="id2529483"></a><h3>eventscript &lt;arguments&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529515"></a><h3>eventscript &lt;arguments&gt;</h3><p>
This is a debugging command. This command can be used to manually
invoke and run the eventscritps with arbitrary arguments.
</p></div><div class="refsect2" lang="en"><a name="id2529495"></a><h3>backupdb &lt;database&gt; &lt;file&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529527"></a><h3>backupdb &lt;database&gt; &lt;file&gt;</h3><p>
This command can be used to copy the entire content of a database out to a file. This file can later be read back into ctdb using the restoredb command.
This is mainly useful for backing up persistent databases such as secrets.tdb and similar.
</p></div><div class="refsect2" lang="en"><a name="id2529509"></a><h3>restoredb &lt;file&gt;</h3><p>
</p></div><div class="refsect2" lang="en"><a name="id2529541"></a><h3>restoredb &lt;file&gt;</h3><p>
This command restores a persistent database that was previously backed up using backupdb.
</p></div></div><div class="refsect1" lang="en"><a name="id2529521"></a><h2>SEE ALSO</h2><p>
</p></div></div><div class="refsect1" lang="en"><a name="id2529553"></a><h2>SEE ALSO</h2><p>
ctdbd(1), onnode(1)
<a class="ulink" href="http://ctdb.samba.org/" target="_top">http://ctdb.samba.org/</a>
</p></div><div class="refsect1" lang="en"><a name="id2529534"></a><h2>COPYRIGHT/LICENSE</h2><div class="literallayout"><p><br>
</p></div><div class="refsect1" lang="en"><a name="id2529566"></a><h2>COPYRIGHT/LICENSE</h2><div class="literallayout"><p><br>
Copyright (C) Andrew Tridgell 2007<br>
Copyright (C) Ronnie sahlberg 2007<br>
<br>

View File

@ -679,15 +679,10 @@ CTDB version 1
<refsect2><title>reloadnodes</title>
<para>
This command is used when adding new nodes to an existing cluster and
to reduce the disruption of this operation. This command should never
be used except when expanding an existing cluster.
This can only be used to expand a cluster. To remove a node from the
cluster you still need to shut down ctdb on all nodes, edit the nodes file
and restart ctdb.
This command is used when adding new nodes, or removing existing nodes from an existing cluster.
</para>
<para>
Procedure:
Procedure to add a node:
</para>
<para>
1, To expand an existing cluster, first ensure with 'ctdb status' that
@ -702,7 +697,7 @@ CTDB version 1
3, Verify that all the nodes have identical /etc/ctdb/nodes files after you edited them and added the new node!
</para>
<para>
4, Run 'ctdb reloadnodes' to force all nodes to reaload the nodesfile.
4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
</para>
<para>
5, Use 'ctdb status' on all nodes and verify that they now show the additional node.
@ -710,6 +705,29 @@ CTDB version 1
<para>
6, Install and configure the new node and bring it online.
</para>
<para>
Procedure to remove a node:
</para>
<para>
1, To remove a node from an existing cluster, first ensure with 'ctdb status' that
all nodes, except the node to be deleted, are up and running and that they are all healthy.
Do not try to remove a node from a cluster unless the cluster is completely healthy!
</para>
<para>
2, Shutdown and poerwoff the node to be removed.
</para>
<para>
3, On all other nodes, edit the /etc/ctdb/nodes file and comment out the node to be removed. Do not delete the line for that node, just comment it out by adding a '#' at the beginning of the line.
</para>
<para>
4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
</para>
<para>
5, Use 'ctdb status' on all nodes and verify that the deleted node no longer shows up in the list..
</para>
<para>
</para>
</refsect2>
<refsect2><title>tickle &lt;srcip:port&gt; &lt;dstip:port&gt;</title>

View File

@ -101,6 +101,11 @@ struct ctdb_call_info {
*/
#define CTDB_SRVID_PUSH_NODE_FLAGS 0xF900000000000000LL
/*
a message ID to get the recovery daemon to reload the nodes file
*/
#define CTDB_SRVID_RELOAD_NODES 0xFA00000000000000LL
/* used on the domain socket, send a pdu to the local daemon */

View File

@ -198,7 +198,8 @@ struct ctdb_node {
#define NODE_FLAGS_PERMANENTLY_DISABLED 0x00000004 /* administrator has disabled node */
#define NODE_FLAGS_BANNED 0x00000008 /* recovery daemon has banned the node */
#define NODE_FLAGS_DISABLED (NODE_FLAGS_UNHEALTHY|NODE_FLAGS_PERMANENTLY_DISABLED)
#define NODE_FLAGS_INACTIVE (NODE_FLAGS_DISCONNECTED|NODE_FLAGS_BANNED)
#define NODE_FLAGS_DELETED 0x00000010 /* this node has been deleted */
#define NODE_FLAGS_INACTIVE (NODE_FLAGS_DELETED|NODE_FLAGS_DISCONNECTED|NODE_FLAGS_BANNED)
uint32_t flags;
/* used by the dead node monitoring */

View File

@ -4,7 +4,7 @@ Summary: Clustered TDB
Vendor: Samba Team
Packager: Samba Team <samba@samba.org>
Name: ctdb
Version: 1.0.82
Version: 1.0.83
Release: 1
Epoch: 0
License: GNU GPL version 3
@ -131,6 +131,30 @@ fi
%{_libdir}/pkgconfig/ctdb.pc
%changelog
* Tue Jun 2 2009 : Version 1.0.83
- Document how to remove a ndoe from a running cluster.
- Hide all deleted nodes from ctdb output.
- Lower the loglevel on some eventscript related items
- Dont queue packets to deleted nodes
- When building initial vnnmap, ignode any nonexisting nodes
- Add a new nodestate : DELETED that is used when deleting a node from an
existing cluster.
- dont remove the ctdb socket when shutting down. This prevents a race in the
initscripts when restarting ctdb quickly after stopping it.
- TDB nesting reworked.
- Remove obsolete ipmux
- From Flavio Carmo Junior: Add eventscript and documentation for ClamAV antivirus engine
- From Sumit Bose: fix the regex in the test to handle the new ctdb
statistics output that was recently added.
- change the socket type we use for grauitious arps from the obsolete
AF_INET/SOCK_PACKET to instead use PF_PACKET/SOCK_RAW.
- Check return codes for some functions, from Sumit Bose, based on codereview by Jim Meyering.
- Sumit Bose: Remove structure memeber node_list_file that is no longer used.
- Sumit Bose: fix configure warning for netfilter.h
- Updates to the webpages by Volker.
- Remove error messages about missing /var/log/log.ctdb file from ctdb_diagnostics.sh from christian Ambach
- Additional error logs if hte eventscript switching from dameon to client mode fails.
- track how long it takes for ctdbd and the recovery daemon to perform the rec-lock fcntl() lock attemt and show this in the ctdb statistics output.
* Thu May 14 2009 : Version 1.0.82
- Update the "ctdb lvsmaster" command to return -1 on error.
- Add a -Y flag to "ctdb lvsmaster"

View File

@ -37,6 +37,11 @@ static void ctdb_check_for_dead_nodes(struct event_context *ev, struct timed_eve
/* send a keepalive to all other nodes, unless */
for (i=0;i<ctdb->num_nodes;i++) {
struct ctdb_node *node = ctdb->nodes[i];
if (node->flags & NODE_FLAGS_DELETED) {
continue;
}
if (node->pnn == ctdb->pnn) {
continue;
}

View File

@ -242,6 +242,10 @@ ctdb_reload_nodes_event(struct event_context *ev, struct timed_event *te,
continue;
}
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
/* any new or different nodes must be added */
if (ctdb->methods->add_node(ctdb->nodes[i]) != 0) {
DEBUG(DEBUG_CRIT, (__location__ " methods->add_node failed at %d\n", i));
@ -253,6 +257,9 @@ ctdb_reload_nodes_event(struct event_context *ev, struct timed_event *te,
}
}
/* tell the recovery daemon to reaload the nodes file too */
ctdb_daemon_send_message(ctdb, ctdb->pnn, CTDB_SRVID_RELOAD_NODES, tdb_null);
talloc_free(tmp_ctx);
return;
}

View File

@ -1332,12 +1332,6 @@ static int do_recovery(struct ctdb_recoverd *rec,
DEBUG(DEBUG_NOTICE, (__location__ " Starting do_recovery\n"));
if (ctdb->num_nodes != nodemap->num) {
DEBUG(DEBUG_ERR, (__location__ " ctdb->num_nodes (%d) != nodemap->num (%d) reloading nodes file\n", ctdb->num_nodes, nodemap->num));
reload_nodes_file(ctdb);
return -1;
}
/* if recovery fails, force it again */
rec->need_recovery = true;
@ -1803,6 +1797,21 @@ DEBUG(DEBUG_ERR, ("recovery master memory dump\n"));
talloc_free(tmp_ctx);
}
/*
handler for reload_nodes
*/
static void reload_nodes_handler(struct ctdb_context *ctdb, uint64_t srvid,
TDB_DATA data, void *private_data)
{
struct ctdb_recoverd *rec = talloc_get_type(private_data, struct ctdb_recoverd);
DEBUG(DEBUG_ERR, (__location__ " Reload nodes file from recovery daemon\n"));
reload_nodes_file(rec->ctdb);
}
/*
handler for recovery master elections
*/
@ -2371,6 +2380,9 @@ static void monitor_cluster(struct ctdb_context *ctdb)
/* register a message port for vacuum fetch */
ctdb_set_message_handler(ctdb, CTDB_SRVID_VACUUM_FETCH, vacuum_fetch_handler, rec);
/* register a message port for reloadnodes */
ctdb_set_message_handler(ctdb, CTDB_SRVID_RELOAD_NODES, reload_nodes_handler, rec);
again:
if (mem_ctx) {
talloc_free(mem_ctx);
@ -2591,14 +2603,16 @@ again:
goto again;
}
for (j=0; j<nodemap->num; j++) {
if (nodemap->nodes[j].flags & NODE_FLAGS_INACTIVE) {
continue;
}
/* release any existing data */
if (ctdb->nodes[j]->public_ips) {
talloc_free(ctdb->nodes[j]->public_ips);
ctdb->nodes[j]->public_ips = NULL;
}
if (nodemap->nodes[j].flags & NODE_FLAGS_INACTIVE) {
continue;
}
/* grab a new shiny list of public ips from the node */
if (ctdb_ctrl_get_public_ips(ctdb, CONTROL_TIMEOUT(),
ctdb->nodes[j]->pnn,

View File

@ -45,6 +45,9 @@ int ctdb_ip_to_nodeid(struct ctdb_context *ctdb, const char *nodeip)
int nodeid;
for (nodeid=0;nodeid<ctdb->num_nodes;nodeid++) {
if (ctdb->nodes[nodeid]->flags & NODE_FLAGS_DELETED) {
continue;
}
if (!strcmp(ctdb->nodes[nodeid]->address.address, nodeip)) {
return nodeid;
}
@ -89,7 +92,7 @@ int ctdb_set_tdb_dir_persistent(struct ctdb_context *ctdb, const char *dir)
}
/*
add a node to the list of active nodes
add a node to the list of nodes
*/
static int ctdb_add_node(struct ctdb_context *ctdb, char *nstr)
{
@ -136,6 +139,46 @@ static int ctdb_add_node(struct ctdb_context *ctdb, char *nstr)
return 0;
}
/*
add an entry for a "deleted" node to the list of nodes.
a "deleted" node is a node that is commented out from the nodes file.
this is used to prevent that subsequent nodes in the nodes list
change their pnn value if a node is "delete" by commenting it out and then
using "ctdb reloadnodes" at runtime.
*/
static int ctdb_add_deleted_node(struct ctdb_context *ctdb)
{
struct ctdb_node *node, **nodep;
nodep = talloc_realloc(ctdb, ctdb->nodes, struct ctdb_node *, ctdb->num_nodes+1);
CTDB_NO_MEMORY(ctdb, nodep);
ctdb->nodes = nodep;
nodep = &ctdb->nodes[ctdb->num_nodes];
(*nodep) = talloc_zero(ctdb->nodes, struct ctdb_node);
CTDB_NO_MEMORY(ctdb, *nodep);
node = *nodep;
if (ctdb_parse_address(ctdb, node, "0.0.0.0", &node->address) != 0) {
DEBUG(DEBUG_ERR,("Failed to setup deleted node %d\n", ctdb->num_nodes));
return -1;
}
node->ctdb = ctdb;
node->name = talloc_strdup(node, "0.0.0.0:0");
/* this assumes that the nodes are kept in sorted order, and no gaps */
node->pnn = ctdb->num_nodes;
/* this node is permanently deleted/disconnected */
node->flags = NODE_FLAGS_DELETED|NODE_FLAGS_DISCONNECTED;
ctdb->num_nodes++;
node->dead_count = 0;
return 0;
}
/*
setup the node list from a file
*/
@ -143,7 +186,7 @@ int ctdb_set_nlist(struct ctdb_context *ctdb, const char *nlist)
{
char **lines;
int nlines;
int i;
int i, j, num_present;
talloc_free(ctdb->nodes);
ctdb->nodes = NULL;
@ -158,7 +201,8 @@ int ctdb_set_nlist(struct ctdb_context *ctdb, const char *nlist)
nlines--;
}
for (i=0;i<nlines;i++) {
num_present = 0;
for (i=0; i < nlines; i++) {
char *node;
node = lines[i];
@ -167,6 +211,10 @@ int ctdb_set_nlist(struct ctdb_context *ctdb, const char *nlist)
node++;
}
if (*node == '#') {
if (ctdb_add_deleted_node(ctdb) != 0) {
talloc_free(lines);
return -1;
}
continue;
}
if (strcmp(node, "") == 0) {
@ -176,19 +224,26 @@ int ctdb_set_nlist(struct ctdb_context *ctdb, const char *nlist)
talloc_free(lines);
return -1;
}
num_present++;
}
/* initialize the vnn mapping table now that we have num_nodes setup */
/* initialize the vnn mapping table now that we have the nodes list,
skipping any deleted nodes
*/
ctdb->vnn_map = talloc(ctdb, struct ctdb_vnn_map);
CTDB_NO_MEMORY(ctdb, ctdb->vnn_map);
ctdb->vnn_map->generation = INVALID_GENERATION;
ctdb->vnn_map->size = ctdb->num_nodes;
ctdb->vnn_map->size = num_present;
ctdb->vnn_map->map = talloc_array(ctdb->vnn_map, uint32_t, ctdb->vnn_map->size);
CTDB_NO_MEMORY(ctdb, ctdb->vnn_map->map);
for(i=0;i<ctdb->vnn_map->size;i++) {
ctdb->vnn_map->map[i] = i;
for(i=0, j=0; i < ctdb->vnn_map->size; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
ctdb->vnn_map->map[j] = i;
j++;
}
talloc_free(lines);
@ -219,9 +274,8 @@ uint32_t ctdb_get_num_active_nodes(struct ctdb_context *ctdb)
{
int i;
uint32_t count=0;
for (i=0;i<ctdb->vnn_map->size;i++) {
struct ctdb_node *node = ctdb->nodes[ctdb->vnn_map->map[i]];
if (!(node->flags & NODE_FLAGS_INACTIVE)) {
for (i=0; i < ctdb->num_nodes; i++) {
if (!(ctdb->nodes[i]->flags & NODE_FLAGS_INACTIVE)) {
count++;
}
}
@ -437,7 +491,10 @@ static void ctdb_broadcast_packet_all(struct ctdb_context *ctdb,
struct ctdb_req_header *hdr)
{
int i;
for (i=0;i<ctdb->num_nodes;i++) {
for (i=0; i < ctdb->num_nodes; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
hdr->destnode = ctdb->nodes[i]->pnn;
ctdb_queue_packet(ctdb, hdr);
}
@ -463,7 +520,10 @@ static void ctdb_broadcast_packet_connected(struct ctdb_context *ctdb,
struct ctdb_req_header *hdr)
{
int i;
for (i=0;i<ctdb->num_nodes;i++) {
for (i=0; i < ctdb->num_nodes; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
if (!(ctdb->nodes[i]->flags & NODE_FLAGS_DISCONNECTED)) {
hdr->destnode = ctdb->nodes[i]->pnn;
ctdb_queue_packet(ctdb, hdr);
@ -500,7 +560,12 @@ void ctdb_queue_packet(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
node = ctdb->nodes[hdr->destnode];
if (hdr->destnode == ctdb->pnn) {
if (node->flags & NODE_FLAGS_DELETED) {
DEBUG(DEBUG_ERR, (__location__ " Can not queue packet to DELETED node %d\n", hdr->destnode));
return;
}
if (node->pnn == ctdb->pnn) {
ctdb_defer_packet(ctdb, hdr);
} else {
if (ctdb->methods == NULL) {

View File

@ -673,6 +673,10 @@ create_merged_ip_list(struct ctdb_context *ctdb, TALLOC_CTX *tmp_ctx)
for (i=0;i<ctdb->num_nodes;i++) {
public_ips = ctdb->nodes[i]->public_ips;
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
/* there were no public ips for this node */
if (public_ips == NULL) {
continue;

View File

@ -405,7 +405,7 @@ static int ctdb_event_script_v(struct ctdb_context *ctdb, const char *options)
continue;
}
if (!(st.st_mode & S_IXUSR)) {
DEBUG(DEBUG_ERR,("Event script %s is not executable. Ignoring this event script\n", str));
DEBUG(DEBUG_INFO,("Event script %s is not executable. Ignoring this event script\n", str));
continue;
}

View File

@ -296,7 +296,11 @@ static int ctdb_tcp_listen_automatic(struct ctdb_context *ctdb)
return -1;
}
for (i=0;i<ctdb->num_nodes;i++) {
for (i=0; i < ctdb->num_nodes; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
/* if node_ip is specified we will only try to bind to that
ip.
*/

View File

@ -69,7 +69,10 @@ static int ctdb_tcp_initialise(struct ctdb_context *ctdb)
exit(1);
}
for (i=0; i<ctdb->num_nodes; i++) {
for (i=0; i < ctdb->num_nodes; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
if (ctdb_tcp_add_node(ctdb->nodes[i]) != 0) {
DEBUG(DEBUG_CRIT, ("methods->add_node failed at %d\n", i));
return -1;
@ -135,7 +138,10 @@ static int ctdb_tcp_start(struct ctdb_context *ctdb)
{
int i;
for (i=0; i<ctdb->num_nodes; i++) {
for (i=0; i < ctdb->num_nodes; i++) {
if (ctdb->nodes[i]->flags & NODE_FLAGS_DELETED) {
continue;
}
ctdb_tcp_connect_node(ctdb->nodes[i]);
}

View File

@ -82,6 +82,10 @@ static void verify_node(struct ctdb_context *ctdb)
DEBUG(DEBUG_ERR, ("Node %u does not exist\n", options.pnn));
exit(ERR_NONODE);
}
if (nodemap->nodes[options.pnn].flags & NODE_FLAGS_DELETED) {
DEBUG(DEBUG_ERR, ("Node %u is DELETED\n", options.pnn));
exit(ERR_DISNODE);
}
if (nodemap->nodes[options.pnn].flags & NODE_FLAGS_DISCONNECTED) {
DEBUG(DEBUG_ERR, ("Node %u is DISCONNECTED\n", options.pnn));
exit(ERR_DISNODE);
@ -472,6 +476,9 @@ static int control_status(struct ctdb_context *ctdb, int argc, const char **argv
if(options.machinereadable){
printf(":Node:IP:Disconnected:Banned:Disabled:Unhealthy:\n");
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DELETED) {
continue;
}
printf(":%d:%s:%d:%d:%d:%d:\n", nodemap->nodes[i].pnn,
ctdb_addr_to_str(&nodemap->nodes[i].addr),
!!(nodemap->nodes[i].flags&NODE_FLAGS_DISCONNECTED),
@ -492,9 +499,14 @@ static int control_status(struct ctdb_context *ctdb, int argc, const char **argv
{ NODE_FLAGS_PERMANENTLY_DISABLED, "DISABLED" },
{ NODE_FLAGS_BANNED, "BANNED" },
{ NODE_FLAGS_UNHEALTHY, "UNHEALTHY" },
{ NODE_FLAGS_DELETED, "DELETED" },
};
char *flags_str = NULL;
int j;
if (nodemap->nodes[i].flags & NODE_FLAGS_DELETED) {
continue;
}
for (j=0;j<ARRAY_SIZE(flag_names);j++) {
if (nodemap->nodes[i].flags & flag_names[j].flag) {
if (flags_str == NULL) {
@ -644,6 +656,9 @@ static int control_natgwlist(struct ctdb_context *ctdb, int argc, const char **a
/* print the pruned list of nodes belonging to this natgw list */
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DELETED) {
continue;
}
printf(":%d:%s:%d:%d:%d:%d:\n", nodemap->nodes[i].pnn,
ctdb_addr_to_str(&nodemap->nodes[i].addr),
!!(nodemap->nodes[i].flags&NODE_FLAGS_DISCONNECTED),
@ -925,6 +940,9 @@ control_get_all_public_ips(struct ctdb_context *ctdb, TALLOC_CTX *tmp_ctx, struc
ip_tree = trbt_create(tmp_ctx, 0);
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DELETED) {
continue;
}
if (nodemap->nodes[i].flags & NODE_FLAGS_DISCONNECTED) {
continue;
}
@ -986,7 +1004,7 @@ find_other_host_for_public_ip(struct ctdb_context *ctdb, ctdb_sock_addr *addr)
}
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DISCONNECTED) {
if (nodemap->nodes[i].flags & NODE_FLAGS_INACTIVE) {
continue;
}
if (nodemap->nodes[i].pnn == options.pnn) {
@ -1103,7 +1121,7 @@ static int control_delip_all(struct ctdb_context *ctdb, int argc, const char **a
/* remove it from the nodes that are not hosting the ip currently */
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DISCONNECTED) {
if (nodemap->nodes[i].flags & NODE_FLAGS_INACTIVE) {
continue;
}
if (ctdb_ctrl_get_public_ips(ctdb, TIMELIMIT(), nodemap->nodes[i].pnn, tmp_ctx, &ips) != 0) {
@ -1131,7 +1149,7 @@ static int control_delip_all(struct ctdb_context *ctdb, int argc, const char **a
/* remove it from every node (also the one hosting it) */
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DISCONNECTED) {
if (nodemap->nodes[i].flags & NODE_FLAGS_INACTIVE) {
continue;
}
if (ctdb_ctrl_get_public_ips(ctdb, TIMELIMIT(), nodemap->nodes[i].pnn, tmp_ctx, &ips) != 0) {
@ -2782,6 +2800,9 @@ static int control_listnodes(struct ctdb_context *ctdb, int argc, const char **a
}
for(i=0;i<nodemap->num;i++){
if (nodemap->nodes[i].flags & NODE_FLAGS_DELETED) {
continue;
}
printf("%s\n", ctdb_addr_to_str(&nodemap->nodes[i].addr));
}