2001-10-16 20:25:28 +04:00
/*
2008-01-30 17:00:02 +03:00
* Copyright ( C ) 2001 - 2004 Sistina Software , Inc . All rights reserved .
2013-07-01 13:27:22 +04:00
* Copyright ( C ) 2004 - 2013 Red Hat , Inc . All rights reserved .
2001-10-16 20:25:28 +04:00
*
2004-03-30 23:35:44 +04:00
* This file is part of LVM2 .
2001-10-16 20:25:28 +04:00
*
2004-03-30 23:35:44 +04:00
* This copyrighted material is made available to anyone wishing to use ,
* modify , copy , or redistribute it subject to the terms and conditions
2007-08-21 00:55:30 +04:00
* of the GNU Lesser General Public License v .2 .1 .
2001-10-16 20:25:28 +04:00
*
2007-08-21 00:55:30 +04:00
* You should have received a copy of the GNU Lesser General Public License
2004-03-30 23:35:44 +04:00
* along with this program ; if not , write to the Free Software Foundation ,
* Inc . , 59 Temple Place , Suite 330 , Boston , MA 02111 - 1307 USA
2001-10-16 20:25:28 +04:00
*/
# include "tools.h"
2010-08-17 02:54:35 +04:00
/*
* Increments * count by the number of _new_ monitored devices .
*/
2007-01-20 01:21:45 +03:00
static int _monitor_lvs_in_vg ( struct cmd_context * cmd ,
2010-08-17 02:54:35 +04:00
struct volume_group * vg , int reg , int * count )
2006-05-12 23:16:48 +04:00
{
struct lv_list * lvl ;
struct logical_volume * lv ;
2010-10-30 01:15:23 +04:00
int r = 1 ;
2006-05-12 23:16:48 +04:00
2008-11-04 01:14:30 +03:00
dm_list_iterate_items ( lvl , & vg - > lvs ) {
2006-05-12 23:16:48 +04:00
lv = lvl - > lv ;
2012-03-23 13:58:04 +04:00
if ( ! lv_info ( cmd , lv , lv_is_thin_pool ( lv ) ? 1 : 0 ,
2014-09-24 12:01:14 +04:00
NULL , 0 , 0 ) )
2012-03-23 13:58:04 +04:00
continue ;
2006-05-12 23:16:48 +04:00
/*
* FIXME : Need to consider all cases . . . PVMOVE , etc
*/
2014-09-16 00:33:53 +04:00
if ( lv_is_pvmove ( lv ) )
2006-05-12 23:16:48 +04:00
continue ;
2010-08-17 20:25:32 +04:00
if ( ! monitor_dev_for_events ( cmd , lv , 0 , reg ) ) {
2010-10-30 01:15:23 +04:00
r = 0 ;
2006-05-12 23:16:48 +04:00
continue ;
2014-09-23 18:47:27 +04:00
}
( * count ) + + ;
2006-05-12 23:16:48 +04:00
}
2010-08-17 02:54:35 +04:00
return r ;
2006-05-12 23:16:48 +04:00
}
2010-01-05 23:56:51 +03:00
static int _poll_lvs_in_vg ( struct cmd_context * cmd ,
struct volume_group * vg )
{
struct lv_list * lvl ;
struct logical_volume * lv ;
struct lvinfo info ;
int lv_active ;
int count = 0 ;
dm_list_iterate_items ( lvl , & vg - > lvs ) {
lv = lvl - > lv ;
2010-08-17 20:25:32 +04:00
if ( ! lv_info ( cmd , lv , 0 , & info , 0 , 0 ) )
2010-01-05 23:56:51 +03:00
lv_active = 0 ;
else
lv_active = info . exists ;
2010-01-13 04:50:34 +03:00
if ( lv_active & &
2014-09-16 00:33:53 +04:00
( lv_is_pvmove ( lv ) | | lv_is_converting ( lv ) | | lv_is_merging ( lv ) ) ) {
2010-01-13 04:50:34 +03:00
lv_spawn_background_polling ( cmd , lv ) ;
count + + ;
}
2010-01-05 23:56:51 +03:00
}
/*
* returns the number of polled devices
* - there is no way to know if lv is already being polled
*/
return count ;
}
2012-06-27 16:59:34 +04:00
static int _activate_lvs_in_vg ( struct cmd_context * cmd , struct volume_group * vg ,
activation_change_t activate )
2002-03-01 22:08:11 +03:00
{
2003-10-16 00:02:46 +04:00
struct lv_list * lvl ;
2002-03-01 22:08:11 +03:00
struct logical_volume * lv ;
2009-11-24 19:08:49 +03:00
int count = 0 , expected_count = 0 ;
2002-03-01 22:08:11 +03:00
2011-09-07 12:41:47 +04:00
sigint_allow ( ) ;
2008-11-04 01:14:30 +03:00
dm_list_iterate_items ( lvl , & vg - > lvs ) {
2011-09-07 12:41:47 +04:00
if ( sigint_caught ( ) )
return_0 ;
2003-10-16 00:02:46 +04:00
lv = lvl - > lv ;
2002-03-01 22:08:11 +03:00
2009-11-18 20:20:18 +03:00
if ( ! lv_is_visible ( lv ) )
continue ;
2014-04-01 19:53:18 +04:00
/* Cache pool cannot be activated */
if ( lv_is_cache_pool ( lv ) )
continue ;
2011-09-14 22:20:03 +04:00
/* If LV is sparse, activate origin instead */
if ( lv_is_cow ( lv ) & & lv_is_virtual_origin ( origin_from_cow ( lv ) ) )
lv = origin_from_cow ( lv ) ;
2003-05-06 16:14:36 +04:00
/* Only request activation of snapshot origin devices */
2005-04-07 16:39:44 +04:00
if ( ( lv - > status & SNAPSHOT ) | | lv_is_cow ( lv ) )
2002-03-18 16:09:27 +03:00
continue ;
2008-06-12 15:49:46 +04:00
/* Only request activation of mirror LV */
if ( ( lv - > status & MIRROR_IMAGE ) | | ( lv - > status & MIRROR_LOG ) )
continue ;
2010-05-24 13:03:39 +04:00
/* Only request activation of the first replicator-dev LV */
/* Avoids retry with all heads in case of failure */
if ( lv_is_replicator_dev ( lv ) & & ( lv ! = first_replicator_dev ( lv ) ) )
continue ;
2008-06-12 15:49:46 +04:00
/* Can't deactivate a pvmove LV */
2004-06-16 21:13:41 +04:00
/* FIXME There needs to be a controlled way of doing this */
2014-09-16 00:33:53 +04:00
if ( lv_is_pvmove ( lv ) & & ! is_change_activating ( activate ) )
2003-05-06 16:14:36 +04:00
continue ;
2014-02-19 14:02:16 +04:00
if ( lv_activation_skip ( lv , activate , arg_count ( cmd , ignoreactivationskip_ARG ) ) )
2013-07-11 14:44:36 +04:00
continue ;
2013-04-11 15:51:08 +04:00
if ( ( activate = = CHANGE_AAY ) & &
! lv_passes_auto_activation_filter ( cmd , lv ) )
2012-09-12 11:47:40 +04:00
continue ;
2012-06-27 18:21:15 +04:00
2009-11-24 19:08:49 +03:00
expected_count + + ;
2013-04-11 15:51:08 +04:00
if ( ! lv_change_activate ( cmd , lv , activate ) ) {
2013-11-29 02:22:03 +04:00
if ( ! lv_is_active_exclusive_remotely ( lv ) )
stack ;
else {
/*
* If the LV is active exclusive remotely ,
* then ignore it here
*/
log_verbose ( " %s/%s is exclusively active on "
" a remote node " , vg - > name , lv - > name ) ;
expected_count - - ; /* not accounted */
}
2002-03-01 22:08:11 +03:00
continue ;
2010-01-06 00:08:34 +03:00
}
2002-03-01 22:08:11 +03:00
count + + ;
}
2011-09-07 12:41:47 +04:00
sigint_restore ( ) ;
2014-09-22 01:22:23 +04:00
sync_local_dev_names ( vg - > cmd ) ; /* Wait until devices are available */
2011-09-07 12:41:47 +04:00
2009-11-24 19:08:49 +03:00
if ( expected_count )
log_verbose ( " %s %d logical volumes in volume group %s " ,
2014-02-18 23:52:17 +04:00
is_change_activating ( activate ) ?
" Activated " : " Deactivated " , count , vg - > name ) ;
2009-11-24 19:08:49 +03:00
2010-10-30 01:15:23 +04:00
return ( expected_count ! = count ) ? 0 : 1 ;
2002-03-01 22:08:11 +03:00
}
2006-05-12 23:16:48 +04:00
static int _vgchange_monitoring ( struct cmd_context * cmd , struct volume_group * vg )
{
2010-10-30 01:15:23 +04:00
int r = 1 ;
2010-08-17 02:54:35 +04:00
int monitored = 0 ;
2006-05-12 23:16:48 +04:00
2010-07-26 23:03:29 +04:00
if ( lvs_in_vg_activated ( vg ) & &
2007-01-25 02:43:27 +03:00
dmeventd_monitor_mode ( ) ! = DMEVENTD_MONITOR_IGNORE ) {
2010-10-30 01:15:23 +04:00
if ( ! _monitor_lvs_in_vg ( cmd , vg , dmeventd_monitor_mode ( ) , & monitored ) )
r = 0 ;
config: add silent mode
Accept -q as the short form of --quiet.
Suppress non-essential standard output if -q is given twice.
Treat log/silent in lvm.conf as equivalent to -qq.
Review all log_print messages and change some to
log_print_unless_silent.
When silent, the following commands still produce output:
dumpconfig, lvdisplay, lvmdiskscan, lvs, pvck, pvdisplay,
pvs, version, vgcfgrestore -l, vgdisplay, vgs.
[Needs checking.]
Non-essential messages are shifted from log level 4 to log level 5
for syslog and lvm2_log_fn purposes.
2012-08-25 23:35:48 +04:00
log_print_unless_silent ( " %d logical volume(s) in volume group "
" \" %s \" %smonitored " ,
monitored , vg - > name , ( dmeventd_monitor_mode ( ) ) ? " " : " un " ) ;
2006-05-12 23:16:48 +04:00
}
2010-10-30 01:15:23 +04:00
return r ;
2006-05-12 23:16:48 +04:00
}
2010-01-05 23:56:51 +03:00
static int _vgchange_background_polling ( struct cmd_context * cmd , struct volume_group * vg )
{
int polled ;
if ( lvs_in_vg_activated ( vg ) & & background_polling ( ) ) {
polled = _poll_lvs_in_vg ( cmd , vg ) ;
2010-10-26 05:37:59 +04:00
if ( polled )
config: add silent mode
Accept -q as the short form of --quiet.
Suppress non-essential standard output if -q is given twice.
Treat log/silent in lvm.conf as equivalent to -qq.
Review all log_print messages and change some to
log_print_unless_silent.
When silent, the following commands still produce output:
dumpconfig, lvdisplay, lvmdiskscan, lvs, pvck, pvdisplay,
pvs, version, vgcfgrestore -l, vgdisplay, vgs.
[Needs checking.]
Non-essential messages are shifted from log level 4 to log level 5
for syslog and lvm2_log_fn purposes.
2012-08-25 23:35:48 +04:00
log_print_unless_silent ( " Background polling started for %d logical volume(s) "
" in volume group \" %s \" " ,
polled , vg - > name ) ;
2010-01-05 23:56:51 +03:00
}
2010-10-30 01:15:23 +04:00
return 1 ;
2010-01-05 23:56:51 +03:00
}
2012-06-27 16:59:34 +04:00
int vgchange_activate ( struct cmd_context * cmd , struct volume_group * vg ,
activation_change_t activate )
2001-10-16 20:25:28 +04:00
{
2014-02-18 23:52:17 +04:00
int lv_open , active , monitored = 0 , r = 1 ;
2012-03-09 02:43:28 +04:00
const struct lv_list * lvl ;
2014-02-18 23:52:17 +04:00
int do_activate = is_change_activating ( activate ) ;
2004-05-24 17:44:10 +04:00
2009-07-15 09:47:55 +04:00
/*
* Safe , since we never write out new metadata here . Required for
* partial activation to work .
*/
2012-06-27 16:59:34 +04:00
cmd - > handles_missing_pvs = 1 ;
2004-06-24 18:48:01 +04:00
2001-10-16 20:25:28 +04:00
/* FIXME: Force argument to deactivate them? */
2012-06-27 16:59:34 +04:00
if ( ! do_activate & & ( lv_open = lvs_in_vg_opened ( vg ) ) ) {
2012-03-09 02:43:28 +04:00
dm_list_iterate_items ( lvl , & vg - > lvs )
if ( lv_is_visible ( lvl - > lv ) & &
2014-09-24 12:05:26 +04:00
! lv_check_not_in_use ( lvl - > lv ) ) {
2012-03-09 02:43:28 +04:00
log_error ( " Can't deactivate volume group \" %s \" with %d open "
" logical volume(s) " , vg - > name , lv_open ) ;
return 0 ;
}
2001-10-16 20:25:28 +04:00
}
2005-05-17 17:44:02 +04:00
/* FIXME Move into library where clvmd can use it */
2012-06-27 16:59:34 +04:00
if ( do_activate )
2005-05-17 17:44:02 +04:00
check_current_backup ( vg ) ;
2012-06-27 16:59:34 +04:00
if ( do_activate & & ( active = lvs_in_vg_activated ( vg ) ) ) {
2002-01-30 18:04:48 +03:00
log_verbose ( " %d logical volume(s) in volume group \" %s \" "
2002-03-01 22:08:11 +03:00
" already active " , active , vg - > name ) ;
2007-01-25 02:43:27 +03:00
if ( dmeventd_monitor_mode ( ) ! = DMEVENTD_MONITOR_IGNORE ) {
2010-10-30 01:15:23 +04:00
if ( ! _monitor_lvs_in_vg ( cmd , vg , dmeventd_monitor_mode ( ) , & monitored ) )
r = 0 ;
2007-01-25 02:43:27 +03:00
log_verbose ( " %d existing logical volume(s) in volume "
" group \" %s \" %smonitored " ,
monitored , vg - > name ,
dmeventd_monitor_mode ( ) ? " " : " un " ) ;
}
2006-05-12 23:16:48 +04:00
}
2001-11-21 22:32:35 +03:00
2014-09-21 13:48:09 +04:00
if ( ! _activate_lvs_in_vg ( cmd , vg , activate ) ) {
stack ;
2010-10-30 01:15:23 +04:00
r = 0 ;
2014-09-21 13:48:09 +04:00
}
2001-10-16 20:25:28 +04:00
2010-05-24 12:59:29 +04:00
/* Print message only if there was not found a missing VG */
if ( ! vg - > cmd_missing_vgs )
config: add silent mode
Accept -q as the short form of --quiet.
Suppress non-essential standard output if -q is given twice.
Treat log/silent in lvm.conf as equivalent to -qq.
Review all log_print messages and change some to
log_print_unless_silent.
When silent, the following commands still produce output:
dumpconfig, lvdisplay, lvmdiskscan, lvs, pvck, pvdisplay,
pvs, version, vgcfgrestore -l, vgdisplay, vgs.
[Needs checking.]
Non-essential messages are shifted from log level 4 to log level 5
for syslog and lvm2_log_fn purposes.
2012-08-25 23:35:48 +04:00
log_print_unless_silent ( " %d logical volume(s) in volume group \" %s \" now active " ,
lvs_in_vg_activated ( vg ) , vg - > name ) ;
2010-10-30 01:15:23 +04:00
return r ;
}
static int _vgchange_refresh ( struct cmd_context * cmd , struct volume_group * vg )
{
log_verbose ( " Refreshing volume group \" %s \" " , vg - > name ) ;
2013-07-01 13:27:11 +04:00
if ( ! vg_refresh_visible ( cmd , vg ) )
return_0 ;
2010-10-30 01:15:23 +04:00
return 1 ;
2001-10-16 20:25:28 +04:00
}
2004-05-19 02:12:53 +04:00
static int _vgchange_alloc ( struct cmd_context * cmd , struct volume_group * vg )
{
alloc_policy_t alloc ;
2012-02-28 18:24:57 +04:00
alloc = ( alloc_policy_t ) arg_uint_value ( cmd , alloc_ARG , ALLOC_NORMAL ) ;
2004-05-19 02:12:53 +04:00
2009-07-09 14:08:54 +04:00
/* FIXME: make consistent with vg_set_alloc_policy() */
2004-05-19 02:12:53 +04:00
if ( alloc = = vg - > alloc ) {
log_error ( " Volume group allocation policy is already %s " ,
get_alloc_string ( vg - > alloc ) ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2009-09-15 02:47:49 +04:00
}
2004-05-19 02:12:53 +04:00
2010-10-30 01:15:23 +04:00
if ( ! vg_set_alloc_policy ( vg , alloc ) )
return_0 ;
2004-05-19 02:12:53 +04:00
2010-10-30 01:15:23 +04:00
return 1 ;
2004-05-19 02:12:53 +04:00
}
2003-10-22 02:06:07 +04:00
static int _vgchange_resizeable ( struct cmd_context * cmd ,
struct volume_group * vg )
2001-10-16 20:25:28 +04:00
{
2002-02-12 00:00:35 +03:00
int resizeable = ! strcmp ( arg_str_value ( cmd , resizeable_ARG , " n " ) , " y " ) ;
2001-10-16 20:25:28 +04:00
2009-09-15 22:35:13 +04:00
if ( resizeable & & vg_is_resizeable ( vg ) ) {
2002-02-12 00:00:35 +03:00
log_error ( " Volume group \" %s \" is already resizeable " ,
vg - > name ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2001-10-16 20:25:28 +04:00
}
2009-09-15 22:35:13 +04:00
if ( ! resizeable & & ! vg_is_resizeable ( vg ) ) {
2002-01-30 18:04:48 +03:00
log_error ( " Volume group \" %s \" is already not resizeable " ,
2001-10-16 20:25:28 +04:00
vg - > name ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2009-09-15 02:47:49 +04:00
}
2002-01-09 16:17:14 +03:00
2002-01-10 18:09:51 +03:00
if ( resizeable )
vg - > status | = RESIZEABLE_VG ;
2001-10-16 20:25:28 +04:00
else
2002-01-10 18:09:51 +03:00
vg - > status & = ~ RESIZEABLE_VG ;
2001-10-16 20:25:28 +04:00
2010-10-30 01:15:23 +04:00
return 1 ;
2001-10-16 20:25:28 +04:00
}
2005-03-22 01:55:12 +03:00
static int _vgchange_clustered ( struct cmd_context * cmd ,
struct volume_group * vg )
{
int clustered = ! strcmp ( arg_str_value ( cmd , clustered_ARG , " n " ) , " y " ) ;
2008-04-10 21:09:32 +04:00
if ( clustered & & ( vg_is_clustered ( vg ) ) ) {
2005-03-22 01:55:12 +03:00
log_error ( " Volume group \" %s \" is already clustered " ,
vg - > name ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2005-03-22 01:55:12 +03:00
}
2008-04-10 21:09:32 +04:00
if ( ! clustered & & ! ( vg_is_clustered ( vg ) ) ) {
2005-03-22 01:55:12 +03:00
log_error ( " Volume group \" %s \" is already not clustered " ,
vg - > name ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2009-09-15 02:47:49 +04:00
}
2005-03-22 01:55:12 +03:00
2014-06-06 07:45:19 +04:00
if ( clustered & & ! arg_count ( cmd , yes_ARG ) ) {
2014-06-06 16:21:09 +04:00
if ( ! clvmd_is_running ( ) ) {
2014-06-06 07:27:40 +04:00
if ( yes_no_prompt ( " LVM cluster daemon (clvmd) is not "
" running. \n "
" Make volume group \" %s \" clustered "
2014-06-10 12:49:14 +04:00
" anyway? [y/n]: " , vg - > name ) = = ' n ' ) {
log_error ( " No volume groups changed. " ) ;
2014-06-06 07:27:40 +04:00
return 0 ;
2014-06-10 12:49:14 +04:00
}
2014-06-06 07:27:40 +04:00
} else if ( ! locking_is_clustered ( ) & &
( yes_no_prompt ( " LVM locking type is not clustered. \n "
" Make volume group \" %s \" clustered "
2014-06-10 12:49:14 +04:00
" anyway? [y/n]: " , vg - > name ) = = ' n ' ) ) {
log_error ( " No volume groups changed. " ) ;
2014-06-06 07:27:40 +04:00
return 0 ;
2014-06-10 12:49:14 +04:00
}
2014-06-06 07:27:40 +04:00
}
2009-10-31 20:30:52 +03:00
if ( ! vg_set_clustered ( vg , clustered ) )
2010-10-30 01:15:23 +04:00
return_0 ;
2005-03-22 01:55:12 +03:00
2010-10-30 01:15:23 +04:00
return 1 ;
2005-03-22 01:55:12 +03:00
}
2003-10-22 02:06:07 +04:00
static int _vgchange_logicalvolume ( struct cmd_context * cmd ,
struct volume_group * vg )
2001-10-16 20:25:28 +04:00
{
2002-12-20 02:25:55 +03:00
uint32_t max_lv = arg_uint_value ( cmd , logicalvolume_ARG , 0 ) ;
2001-10-16 20:25:28 +04:00
2010-10-30 01:15:23 +04:00
if ( ! vg_set_max_lv ( vg , max_lv ) )
return_0 ;
2001-10-16 20:25:28 +04:00
2010-10-30 01:15:23 +04:00
return 1 ;
2001-10-16 20:25:28 +04:00
}
2002-11-18 17:04:08 +03:00
2006-08-16 18:41:42 +04:00
static int _vgchange_physicalvolumes ( struct cmd_context * cmd ,
struct volume_group * vg )
{
uint32_t max_pv = arg_uint_value ( cmd , maxphysicalvolumes_ARG , 0 ) ;
2010-10-30 01:15:23 +04:00
if ( ! vg_set_max_pv ( vg , max_pv ) )
return_0 ;
2006-08-16 18:41:42 +04:00
2010-10-30 01:15:23 +04:00
return 1 ;
2006-08-16 18:41:42 +04:00
}
2005-04-18 18:56:42 +04:00
static int _vgchange_pesize ( struct cmd_context * cmd , struct volume_group * vg )
{
uint32_t extent_size ;
2011-03-02 23:00:09 +03:00
if ( arg_uint64_value ( cmd , physicalextentsize_ARG , 0 ) > MAX_EXTENT_SIZE ) {
2012-10-16 12:14:41 +04:00
log_warn ( " Physical extent size cannot be larger than %s. " ,
display_size ( cmd , ( uint64_t ) MAX_EXTENT_SIZE ) ) ;
2011-03-02 23:00:09 +03:00
return 1 ;
}
2007-11-14 03:08:25 +03:00
extent_size = arg_uint_value ( cmd , physicalextentsize_ARG , 0 ) ;
2009-07-09 14:02:15 +04:00
/* FIXME: remove check - redundant with vg_change_pesize */
2005-04-18 18:56:42 +04:00
if ( extent_size = = vg - > extent_size ) {
2012-10-16 12:14:41 +04:00
log_warn ( " Physical extent size of VG %s is already %s. " ,
vg - > name , display_size ( cmd , ( uint64_t ) extent_size ) ) ;
2010-10-30 01:15:23 +04:00
return 1 ;
2005-04-18 18:56:42 +04:00
}
2010-10-30 01:15:23 +04:00
if ( ! vg_set_extent_size ( vg , extent_size ) )
return_0 ;
2005-04-18 18:56:42 +04:00
2013-12-12 14:26:35 +04:00
if ( ! vg_check_pv_dev_block_sizes ( vg ) ) {
log_error ( " Failed to change physical extent size for VG %s. " ,
vg - > name ) ;
return 0 ;
}
2010-10-30 01:15:23 +04:00
return 1 ;
2005-04-18 18:56:42 +04:00
}
2010-10-30 01:15:23 +04:00
static int _vgchange_addtag ( struct cmd_context * cmd , struct volume_group * vg )
{
2011-01-24 16:38:31 +03:00
return change_tag ( cmd , vg , NULL , NULL , addtag_ARG ) ;
2010-10-30 01:15:23 +04:00
}
2004-03-08 20:19:15 +03:00
2010-10-30 01:15:23 +04:00
static int _vgchange_deltag ( struct cmd_context * cmd , struct volume_group * vg )
{
2011-01-24 16:38:31 +03:00
return change_tag ( cmd , vg , NULL , NULL , deltag_ARG ) ;
2004-03-08 20:19:15 +03:00
}
2010-07-09 19:34:40 +04:00
static int _vgchange_uuid ( struct cmd_context * cmd __attribute__ ( ( unused ) ) ,
2006-05-10 01:23:51 +04:00
struct volume_group * vg )
2004-01-13 21:42:05 +03:00
{
struct lv_list * lvl ;
if ( lvs_in_vg_activated ( vg ) ) {
log_error ( " Volume group has active logical volumes " ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2009-09-15 02:47:49 +04:00
}
2004-01-13 21:42:05 +03:00
2005-01-20 21:11:53 +03:00
if ( ! id_create ( & vg - > id ) ) {
log_error ( " Failed to generate new random UUID for VG %s. " ,
vg - > name ) ;
2010-10-30 01:15:23 +04:00
return 0 ;
2005-01-20 21:11:53 +03:00
}
2004-01-13 21:42:05 +03:00
2008-11-04 01:14:30 +03:00
dm_list_iterate_items ( lvl , & vg - > lvs ) {
2004-01-13 21:42:05 +03:00
memcpy ( & lvl - > lv - > lvid , & vg - > id , sizeof ( vg - > id ) ) ;
}
2010-10-30 01:15:23 +04:00
return 1 ;
2008-12-22 12:00:51 +03:00
}
2010-06-29 00:37:37 +04:00
static int _vgchange_metadata_copies ( struct cmd_context * cmd ,
struct volume_group * vg )
{
2010-07-01 00:03:52 +04:00
uint32_t mda_copies = arg_uint_value ( cmd , vgmetadatacopies_ARG , DEFAULT_VGMETADATACOPIES ) ;
2010-06-29 00:37:37 +04:00
if ( mda_copies = = vg_mda_copies ( vg ) ) {
2010-07-01 00:03:52 +04:00
if ( vg_mda_copies ( vg ) = = VGMETADATACOPIES_UNMANAGED )
2012-10-16 12:14:41 +04:00
log_warn ( " Number of metadata copies for VG %s is already unmanaged. " ,
vg - > name ) ;
2010-07-01 00:03:52 +04:00
else
2012-10-16 12:14:41 +04:00
log_warn ( " Number of metadata copies for VG %s is already %u. " ,
vg - > name , mda_copies ) ;
2010-10-30 01:15:23 +04:00
return 1 ;
2010-06-29 00:37:37 +04:00
}
2010-10-30 01:15:23 +04:00
if ( ! vg_set_mda_copies ( vg , mda_copies ) )
return_0 ;
2010-06-29 00:37:37 +04:00
2010-10-30 01:15:23 +04:00
return 1 ;
2010-06-29 00:37:37 +04:00
}
2013-06-25 14:33:06 +04:00
static int _vgchange_profile ( struct cmd_context * cmd ,
struct volume_group * vg )
{
const char * old_profile_name , * new_profile_name ;
struct profile * new_profile ;
old_profile_name = vg - > profile ? vg - > profile - > name : " (no profile) " ;
if ( arg_count ( cmd , detachprofile_ARG ) ) {
new_profile_name = " (no profile) " ;
vg - > profile = NULL ;
} else {
config: differentiate command and metadata profiles and consolidate profile handling code
- When defining configuration source, the code now uses separate
CONFIG_PROFILE_COMMAND and CONFIG_PROFILE_METADATA markers
(before, it was just CONFIG_PROFILE that did not make the
difference between the two). This helps when checking the
configuration if it contains correct set of options which
are all in either command-profilable or metadata-profilable
group without mixing these groups together - so it's a firm
distinction. The "command profile" can't contain
"metadata profile" and vice versa! This is strictly checked
and if the settings are mixed, such profile is rejected and
it's not used. So in the end, the CONFIG_PROFILE_COMMAND
set of options and CONFIG_PROFILE_METADATA are mutually exclusive
sets.
- Marking configuration with one or the other marker will also
determine the way these configuration sources are positioned
in the configuration cascade which is now:
CONFIG_STRING -> CONFIG_PROFILE_COMMAND -> CONFIG_PROFILE_METADATA -> CONFIG_FILE/CONFIG_MERGED_FILES
- Marking configuration with one or the other marker will also make
it possible to issue a command context refresh (will be probably
a part of a future patch) if needed for settings in global profile
set. For settings in metadata profile set this is impossible since
we can't refresh cmd context in the middle of reading VG/LV metadata
and for each VG/LV separately because each VG/LV can have a different
metadata profile assinged and it's not possible to change these
settings at this level.
- When command profile is incorrect, it's rejected *and also* the
command exits immediately - the profile *must* be correct for the
command that was run with a profile to be executed. Before this
patch, when the profile was found incorrect, there was just the
warning message and the command continued without profile applied.
But it's more correct to exit immediately in this case.
- When metadata profile is incorrect, we reject it during command
runtime (as we know the profile name from metadata and not early
from command line as it is in case of command profiles) and we
*do continue* with the command as we're in the middle of operation.
Also, the metadata profile is applied directly and on the fly on
find_config_tree_* fn call and even if the metadata profile is
found incorrect, we still need to return the non-profiled value
as found in the other configuration provided or default value.
To exit immediately even in this case, we'd need to refactor
existing find_config_tree_* fns so they can return error. Currently,
these fns return only config values (which end up with default
values in the end if the config is not found).
- To check the profile validity before use to be sure it's correct,
one can use :
lvm dumpconfig --commandprofile/--metadataprofile ProfileName --validate
(the --commandprofile/--metadataprofile for dumpconfig will come
as part of the subsequent patch)
- This patch also adds a reference to --commandprofile and
--metadataprofile in the cmd help string (which was missing before
for the --profile for some commands). We do not mention --profile
now as people should use --commandprofile or --metadataprofile
directly. However, the --profile is still supported for backward
compatibility and it's translated as:
--profile == --metadataprofile for lvcreate, vgcreate, lvchange and vgchange
(as these commands are able to attach profile to metadata)
--profile == --commandprofile for all the other commands
(--metadataprofile is not allowed there as it makes no sense)
- This patch also contains some cleanups to make the code handling
the profiles more readable...
2014-05-20 16:13:10 +04:00
if ( arg_count ( cmd , metadataprofile_ARG ) )
new_profile_name = arg_str_value ( cmd , metadataprofile_ARG , NULL ) ;
else
new_profile_name = arg_str_value ( cmd , profile_ARG , NULL ) ;
if ( ! ( new_profile = add_profile ( cmd , new_profile_name , CONFIG_PROFILE_METADATA ) ) )
2013-06-25 14:33:06 +04:00
return_0 ;
vg - > profile = new_profile ;
}
log_verbose ( " Changing configuration profile for VG %s: %s -> %s. " ,
vg - > name , old_profile_name , new_profile_name ) ;
return 1 ;
}
2002-11-18 17:04:08 +03:00
static int vgchange_single ( struct cmd_context * cmd , const char * vg_name ,
2009-07-01 21:00:50 +04:00
struct volume_group * vg ,
2010-07-09 19:34:40 +04:00
void * handle __attribute__ ( ( unused ) ) )
2002-11-18 17:04:08 +03:00
{
2014-09-17 16:27:46 +04:00
int ret = ECMD_PROCESSED ;
2013-07-06 21:40:09 +04:00
unsigned i ;
2014-09-17 16:27:46 +04:00
struct lv_list * lvl ;
2010-10-30 01:15:23 +04:00
2013-07-06 21:40:09 +04:00
static const struct {
2010-10-30 01:15:23 +04:00
int arg ;
int ( * fn ) ( struct cmd_context * cmd , struct volume_group * vg ) ;
} _vgchange_args [ ] = {
{ logicalvolume_ARG , & _vgchange_logicalvolume } ,
{ maxphysicalvolumes_ARG , & _vgchange_physicalvolumes } ,
{ resizeable_ARG , & _vgchange_resizeable } ,
{ deltag_ARG , & _vgchange_deltag } ,
{ addtag_ARG , & _vgchange_addtag } ,
{ physicalextentsize_ARG , & _vgchange_pesize } ,
{ uuid_ARG , & _vgchange_uuid } ,
{ alloc_ARG , & _vgchange_alloc } ,
{ clustered_ARG , & _vgchange_clustered } ,
{ vgmetadatacopies_ARG , & _vgchange_metadata_copies } ,
config: differentiate command and metadata profiles and consolidate profile handling code
- When defining configuration source, the code now uses separate
CONFIG_PROFILE_COMMAND and CONFIG_PROFILE_METADATA markers
(before, it was just CONFIG_PROFILE that did not make the
difference between the two). This helps when checking the
configuration if it contains correct set of options which
are all in either command-profilable or metadata-profilable
group without mixing these groups together - so it's a firm
distinction. The "command profile" can't contain
"metadata profile" and vice versa! This is strictly checked
and if the settings are mixed, such profile is rejected and
it's not used. So in the end, the CONFIG_PROFILE_COMMAND
set of options and CONFIG_PROFILE_METADATA are mutually exclusive
sets.
- Marking configuration with one or the other marker will also
determine the way these configuration sources are positioned
in the configuration cascade which is now:
CONFIG_STRING -> CONFIG_PROFILE_COMMAND -> CONFIG_PROFILE_METADATA -> CONFIG_FILE/CONFIG_MERGED_FILES
- Marking configuration with one or the other marker will also make
it possible to issue a command context refresh (will be probably
a part of a future patch) if needed for settings in global profile
set. For settings in metadata profile set this is impossible since
we can't refresh cmd context in the middle of reading VG/LV metadata
and for each VG/LV separately because each VG/LV can have a different
metadata profile assinged and it's not possible to change these
settings at this level.
- When command profile is incorrect, it's rejected *and also* the
command exits immediately - the profile *must* be correct for the
command that was run with a profile to be executed. Before this
patch, when the profile was found incorrect, there was just the
warning message and the command continued without profile applied.
But it's more correct to exit immediately in this case.
- When metadata profile is incorrect, we reject it during command
runtime (as we know the profile name from metadata and not early
from command line as it is in case of command profiles) and we
*do continue* with the command as we're in the middle of operation.
Also, the metadata profile is applied directly and on the fly on
find_config_tree_* fn call and even if the metadata profile is
found incorrect, we still need to return the non-profiled value
as found in the other configuration provided or default value.
To exit immediately even in this case, we'd need to refactor
existing find_config_tree_* fns so they can return error. Currently,
these fns return only config values (which end up with default
values in the end if the config is not found).
- To check the profile validity before use to be sure it's correct,
one can use :
lvm dumpconfig --commandprofile/--metadataprofile ProfileName --validate
(the --commandprofile/--metadataprofile for dumpconfig will come
as part of the subsequent patch)
- This patch also adds a reference to --commandprofile and
--metadataprofile in the cmd help string (which was missing before
for the --profile for some commands). We do not mention --profile
now as people should use --commandprofile or --metadataprofile
directly. However, the --profile is still supported for backward
compatibility and it's translated as:
--profile == --metadataprofile for lvcreate, vgcreate, lvchange and vgchange
(as these commands are able to attach profile to metadata)
--profile == --commandprofile for all the other commands
(--metadataprofile is not allowed there as it makes no sense)
- This patch also contains some cleanups to make the code handling
the profiles more readable...
2014-05-20 16:13:10 +04:00
{ metadataprofile_ARG , & _vgchange_profile } ,
2013-06-25 14:33:06 +04:00
{ profile_ARG , & _vgchange_profile } ,
{ detachprofile_ARG , & _vgchange_profile } ,
2010-10-30 01:15:23 +04:00
} ;
2003-10-22 02:06:07 +04:00
2009-09-14 23:44:15 +04:00
if ( vg_is_exported ( vg ) ) {
2002-11-18 17:04:08 +03:00
log_error ( " Volume group \" %s \" is exported " , vg_name ) ;
return ECMD_FAILED ;
}
2010-01-06 22:08:58 +03:00
/*
* FIXME : DEFAULT_BACKGROUND_POLLING should be " unspecified " .
* If - - poll is explicitly provided use it ; otherwise polling
* should only be started if the LV is not already active . So :
* 1 ) change the activation code to say if the LV was actually activated
* 2 ) make polling of an LV tightly coupled with LV activation
2010-05-06 15:15:55 +04:00
*
* Do not initiate any polling if - - sysinit option is used .
2010-01-06 22:08:58 +03:00
*/
2010-05-06 15:15:55 +04:00
init_background_polling ( arg_count ( cmd , sysinit_ARG ) ? 0 :
arg_int_value ( cmd , poll_ARG ,
DEFAULT_BACKGROUND_POLLING ) ) ;
2010-01-05 23:56:51 +03:00
2013-07-06 21:40:09 +04:00
for ( i = 0 ; i < DM_ARRAY_SIZE ( _vgchange_args ) ; + + i ) {
2010-10-30 01:15:23 +04:00
if ( arg_count ( cmd , _vgchange_args [ i ] . arg ) ) {
2013-07-01 13:27:22 +04:00
if ( ! archive ( vg ) )
return_ECMD_FAILED ;
if ( ! _vgchange_args [ i ] . fn ( cmd , vg ) )
return_ECMD_FAILED ;
2010-10-30 01:15:23 +04:00
}
}
2013-06-30 20:03:34 +04:00
if ( vg_is_archived ( vg ) ) {
2013-07-01 13:27:22 +04:00
if ( ! vg_write ( vg ) | | ! vg_commit ( vg ) )
return_ECMD_FAILED ;
2010-10-30 01:15:23 +04:00
backup ( vg ) ;
config: add silent mode
Accept -q as the short form of --quiet.
Suppress non-essential standard output if -q is given twice.
Treat log/silent in lvm.conf as equivalent to -qq.
Review all log_print messages and change some to
log_print_unless_silent.
When silent, the following commands still produce output:
dumpconfig, lvdisplay, lvmdiskscan, lvs, pvck, pvdisplay,
pvs, version, vgcfgrestore -l, vgdisplay, vgs.
[Needs checking.]
Non-essential messages are shifted from log level 4 to log level 5
for syslog and lvm2_log_fn purposes.
2012-08-25 23:35:48 +04:00
log_print_unless_silent ( " Volume group \" %s \" successfully changed " , vg - > name ) ;
2014-09-17 16:27:46 +04:00
/* FIXME: fix clvmd bug and take DLM lock for non clustered VGs. */
if ( arg_is_set ( cmd , clustered_ARG ) & &
vg_is_clustered ( vg ) & & /* just switched to clustered */
locking_is_clustered ( ) & &
locking_supports_remote_queries ( ) )
dm_list_iterate_items ( lvl , & vg - > lvs ) {
if ( ( lv_lock_holder ( lvl - > lv ) ! = lvl - > lv ) | |
! lv_is_active ( lvl - > lv ) )
continue ;
if ( ! activate_lv_excl_local ( cmd , lvl - > lv ) | |
! lv_is_active_exclusive_locally ( lvl - > lv ) ) {
log_error ( " Can't reactive logical volume %s, "
" please fix manually. " ,
display_lvname ( lvl - > lv ) ) ;
ret = ECMD_FAILED ;
}
if ( lv_is_mirror ( lvl - > lv ) )
/* Give hint for clustered mirroring */
log_print_unless_silent ( " For clustered mirroring of %s "
" deactivation and activation is needed. " ,
display_lvname ( lvl - > lv ) ) ;
}
2010-10-30 01:15:23 +04:00
}
2012-06-27 15:48:31 +04:00
if ( arg_count ( cmd , activate_ARG ) ) {
2012-08-21 17:49:23 +04:00
if ( ! vgchange_activate ( cmd , vg , ( activation_change_t )
arg_uint_value ( cmd , activate_ARG , CHANGE_AY ) ) )
2013-07-01 13:27:22 +04:00
return_ECMD_FAILED ;
2010-10-26 05:37:59 +04:00
}
if ( arg_count ( cmd , refresh_ARG ) ) {
/* refreshes the visible LVs (which starts polling) */
2010-10-30 01:15:23 +04:00
if ( ! _vgchange_refresh ( cmd , vg ) )
2013-07-01 13:27:22 +04:00
return_ECMD_FAILED ;
2010-10-26 05:37:59 +04:00
}
2002-11-18 17:04:08 +03:00
2012-06-27 15:48:31 +04:00
if ( ! arg_count ( cmd , activate_ARG ) & &
2010-10-26 05:37:59 +04:00
! arg_count ( cmd , refresh_ARG ) & &
arg_count ( cmd , monitor_ARG ) ) {
/* -ay* will have already done monitoring changes */
2010-10-30 01:15:23 +04:00
if ( ! _vgchange_monitoring ( cmd , vg ) )
2013-07-01 13:27:22 +04:00
return_ECMD_FAILED ;
2010-10-26 05:37:59 +04:00
}
2006-05-12 23:16:48 +04:00
2010-10-26 05:37:59 +04:00
if ( ! arg_count ( cmd , refresh_ARG ) & &
2011-06-11 04:03:06 +04:00
background_polling ( ) )
2010-10-30 01:15:23 +04:00
if ( ! _vgchange_background_polling ( cmd , vg ) )
2013-07-01 13:27:22 +04:00
return_ECMD_FAILED ;
2004-03-08 20:19:15 +03:00
2014-09-17 16:27:46 +04:00
return ret ;
2002-11-18 17:04:08 +03:00
}
int vgchange ( struct cmd_context * cmd , int argc , char * * argv )
{
2010-10-30 01:15:23 +04:00
/* Update commands that can be combined */
[lv|vg]change: Allow limited metadata changes when PVs are missing
A while back, the behavior of LVM changed from allowing metadata changes
when PVs were missing to not allowing changes. Until recently, this
change was tolerated by HA-LVM by forcing a 'vgreduce --removemissing'
before trying (again) to add tags to an LV and then activate it. LVM
mirroring requires that failed devices are removed anyway, so this was
largely harmless. However, RAID LVs do not require devices to be removed
from the array in order to be activated. In fact, in an HA-LVM
environment this would be very undesirable. Device failures in such an
environment can often be transient and it would be much better to restore
the device to the array than synchronize an entirely new device.
There are two methods that can be used to setup an HA-LVM environment:
"clvm" or "tagging". For RAID LVs, "clvm" is out of the question because
RAID LVs are not supported in clustered VGs - not even in an exclusively
activated manner. That leaves "tagging". HA-LVM uses tagging - coupled
with 'volume_list' - to ensure that only one machine can have an LV active
at a time. If updates are not allowed when a PV is missing, it is
impossible to add or remove tags to allow for activation. This removes
one of the most basic functionalities of HA-LVM - site redundancy. If
mirroring or RAID is used to replicate the storage in two data centers
and one of them goes down, a server and a storage device are lost. When
the service fails-over to the alternate site, the VG will be "partial".
Unable to add a tag to the VG/LV, the RAID device will be unable to
activate.
The solution is to allow vgchange and lvchange to alter the LVM metadata
for a limited set of options - --[add|del]tag included. The set of
allowable options are ones that do not cause changes to the DM kernel
target (like --resync would) or could alter the structure of the LV
(like allocation or conversion).
2012-10-10 20:33:10 +04:00
int update_partial_safe =
arg_count ( cmd , deltag_ARG ) | |
2013-06-25 14:33:06 +04:00
arg_count ( cmd , addtag_ARG ) | |
config: differentiate command and metadata profiles and consolidate profile handling code
- When defining configuration source, the code now uses separate
CONFIG_PROFILE_COMMAND and CONFIG_PROFILE_METADATA markers
(before, it was just CONFIG_PROFILE that did not make the
difference between the two). This helps when checking the
configuration if it contains correct set of options which
are all in either command-profilable or metadata-profilable
group without mixing these groups together - so it's a firm
distinction. The "command profile" can't contain
"metadata profile" and vice versa! This is strictly checked
and if the settings are mixed, such profile is rejected and
it's not used. So in the end, the CONFIG_PROFILE_COMMAND
set of options and CONFIG_PROFILE_METADATA are mutually exclusive
sets.
- Marking configuration with one or the other marker will also
determine the way these configuration sources are positioned
in the configuration cascade which is now:
CONFIG_STRING -> CONFIG_PROFILE_COMMAND -> CONFIG_PROFILE_METADATA -> CONFIG_FILE/CONFIG_MERGED_FILES
- Marking configuration with one or the other marker will also make
it possible to issue a command context refresh (will be probably
a part of a future patch) if needed for settings in global profile
set. For settings in metadata profile set this is impossible since
we can't refresh cmd context in the middle of reading VG/LV metadata
and for each VG/LV separately because each VG/LV can have a different
metadata profile assinged and it's not possible to change these
settings at this level.
- When command profile is incorrect, it's rejected *and also* the
command exits immediately - the profile *must* be correct for the
command that was run with a profile to be executed. Before this
patch, when the profile was found incorrect, there was just the
warning message and the command continued without profile applied.
But it's more correct to exit immediately in this case.
- When metadata profile is incorrect, we reject it during command
runtime (as we know the profile name from metadata and not early
from command line as it is in case of command profiles) and we
*do continue* with the command as we're in the middle of operation.
Also, the metadata profile is applied directly and on the fly on
find_config_tree_* fn call and even if the metadata profile is
found incorrect, we still need to return the non-profiled value
as found in the other configuration provided or default value.
To exit immediately even in this case, we'd need to refactor
existing find_config_tree_* fns so they can return error. Currently,
these fns return only config values (which end up with default
values in the end if the config is not found).
- To check the profile validity before use to be sure it's correct,
one can use :
lvm dumpconfig --commandprofile/--metadataprofile ProfileName --validate
(the --commandprofile/--metadataprofile for dumpconfig will come
as part of the subsequent patch)
- This patch also adds a reference to --commandprofile and
--metadataprofile in the cmd help string (which was missing before
for the --profile for some commands). We do not mention --profile
now as people should use --commandprofile or --metadataprofile
directly. However, the --profile is still supported for backward
compatibility and it's translated as:
--profile == --metadataprofile for lvcreate, vgcreate, lvchange and vgchange
(as these commands are able to attach profile to metadata)
--profile == --commandprofile for all the other commands
(--metadataprofile is not allowed there as it makes no sense)
- This patch also contains some cleanups to make the code handling
the profiles more readable...
2014-05-20 16:13:10 +04:00
arg_count ( cmd , metadataprofile_ARG ) | |
2013-06-25 14:33:06 +04:00
arg_count ( cmd , profile_ARG ) | |
arg_count ( cmd , detachprofile_ARG ) ;
[lv|vg]change: Allow limited metadata changes when PVs are missing
A while back, the behavior of LVM changed from allowing metadata changes
when PVs were missing to not allowing changes. Until recently, this
change was tolerated by HA-LVM by forcing a 'vgreduce --removemissing'
before trying (again) to add tags to an LV and then activate it. LVM
mirroring requires that failed devices are removed anyway, so this was
largely harmless. However, RAID LVs do not require devices to be removed
from the array in order to be activated. In fact, in an HA-LVM
environment this would be very undesirable. Device failures in such an
environment can often be transient and it would be much better to restore
the device to the array than synchronize an entirely new device.
There are two methods that can be used to setup an HA-LVM environment:
"clvm" or "tagging". For RAID LVs, "clvm" is out of the question because
RAID LVs are not supported in clustered VGs - not even in an exclusively
activated manner. That leaves "tagging". HA-LVM uses tagging - coupled
with 'volume_list' - to ensure that only one machine can have an LV active
at a time. If updates are not allowed when a PV is missing, it is
impossible to add or remove tags to allow for activation. This removes
one of the most basic functionalities of HA-LVM - site redundancy. If
mirroring or RAID is used to replicate the storage in two data centers
and one of them goes down, a server and a storage device are lost. When
the service fails-over to the alternate site, the VG will be "partial".
Unable to add a tag to the VG/LV, the RAID device will be unable to
activate.
The solution is to allow vgchange and lvchange to alter the LVM metadata
for a limited set of options - --[add|del]tag included. The set of
allowable options are ones that do not cause changes to the DM kernel
target (like --resync would) or could alter the structure of the LV
(like allocation or conversion).
2012-10-10 20:33:10 +04:00
int update_partial_unsafe =
2010-10-30 01:15:23 +04:00
arg_count ( cmd , logicalvolume_ARG ) | |
arg_count ( cmd , maxphysicalvolumes_ARG ) | |
arg_count ( cmd , resizeable_ARG ) | |
arg_count ( cmd , uuid_ARG ) | |
arg_count ( cmd , physicalextentsize_ARG ) | |
arg_count ( cmd , clustered_ARG ) | |
arg_count ( cmd , alloc_ARG ) | |
arg_count ( cmd , vgmetadatacopies_ARG ) ;
[lv|vg]change: Allow limited metadata changes when PVs are missing
A while back, the behavior of LVM changed from allowing metadata changes
when PVs were missing to not allowing changes. Until recently, this
change was tolerated by HA-LVM by forcing a 'vgreduce --removemissing'
before trying (again) to add tags to an LV and then activate it. LVM
mirroring requires that failed devices are removed anyway, so this was
largely harmless. However, RAID LVs do not require devices to be removed
from the array in order to be activated. In fact, in an HA-LVM
environment this would be very undesirable. Device failures in such an
environment can often be transient and it would be much better to restore
the device to the array than synchronize an entirely new device.
There are two methods that can be used to setup an HA-LVM environment:
"clvm" or "tagging". For RAID LVs, "clvm" is out of the question because
RAID LVs are not supported in clustered VGs - not even in an exclusively
activated manner. That leaves "tagging". HA-LVM uses tagging - coupled
with 'volume_list' - to ensure that only one machine can have an LV active
at a time. If updates are not allowed when a PV is missing, it is
impossible to add or remove tags to allow for activation. This removes
one of the most basic functionalities of HA-LVM - site redundancy. If
mirroring or RAID is used to replicate the storage in two data centers
and one of them goes down, a server and a storage device are lost. When
the service fails-over to the alternate site, the VG will be "partial".
Unable to add a tag to the VG/LV, the RAID device will be unable to
activate.
The solution is to allow vgchange and lvchange to alter the LVM metadata
for a limited set of options - --[add|del]tag included. The set of
allowable options are ones that do not cause changes to the DM kernel
target (like --resync would) or could alter the structure of the LV
(like allocation or conversion).
2012-10-10 20:33:10 +04:00
int update = update_partial_safe | | update_partial_unsafe ;
2010-10-30 01:15:23 +04:00
if ( ! update & &
2012-06-27 15:48:31 +04:00
! arg_count ( cmd , activate_ARG ) & &
2010-10-30 01:15:23 +04:00
! arg_count ( cmd , monitor_ARG ) & &
! arg_count ( cmd , poll_ARG ) & &
! arg_count ( cmd , refresh_ARG ) ) {
2010-01-05 23:56:51 +03:00
log_error ( " Need 1 or more of -a, -c, -l, -p, -s, -x, "
" --refresh, --uuid, --alloc, --addtag, --deltag, "
2010-06-29 00:37:37 +04:00
" --monitor, --poll, --vgmetadatacopies or "
" --metadatacopies " ) ;
2002-11-18 17:04:08 +03:00
return EINVALID_CMD_LINE ;
}
config: differentiate command and metadata profiles and consolidate profile handling code
- When defining configuration source, the code now uses separate
CONFIG_PROFILE_COMMAND and CONFIG_PROFILE_METADATA markers
(before, it was just CONFIG_PROFILE that did not make the
difference between the two). This helps when checking the
configuration if it contains correct set of options which
are all in either command-profilable or metadata-profilable
group without mixing these groups together - so it's a firm
distinction. The "command profile" can't contain
"metadata profile" and vice versa! This is strictly checked
and if the settings are mixed, such profile is rejected and
it's not used. So in the end, the CONFIG_PROFILE_COMMAND
set of options and CONFIG_PROFILE_METADATA are mutually exclusive
sets.
- Marking configuration with one or the other marker will also
determine the way these configuration sources are positioned
in the configuration cascade which is now:
CONFIG_STRING -> CONFIG_PROFILE_COMMAND -> CONFIG_PROFILE_METADATA -> CONFIG_FILE/CONFIG_MERGED_FILES
- Marking configuration with one or the other marker will also make
it possible to issue a command context refresh (will be probably
a part of a future patch) if needed for settings in global profile
set. For settings in metadata profile set this is impossible since
we can't refresh cmd context in the middle of reading VG/LV metadata
and for each VG/LV separately because each VG/LV can have a different
metadata profile assinged and it's not possible to change these
settings at this level.
- When command profile is incorrect, it's rejected *and also* the
command exits immediately - the profile *must* be correct for the
command that was run with a profile to be executed. Before this
patch, when the profile was found incorrect, there was just the
warning message and the command continued without profile applied.
But it's more correct to exit immediately in this case.
- When metadata profile is incorrect, we reject it during command
runtime (as we know the profile name from metadata and not early
from command line as it is in case of command profiles) and we
*do continue* with the command as we're in the middle of operation.
Also, the metadata profile is applied directly and on the fly on
find_config_tree_* fn call and even if the metadata profile is
found incorrect, we still need to return the non-profiled value
as found in the other configuration provided or default value.
To exit immediately even in this case, we'd need to refactor
existing find_config_tree_* fns so they can return error. Currently,
these fns return only config values (which end up with default
values in the end if the config is not found).
- To check the profile validity before use to be sure it's correct,
one can use :
lvm dumpconfig --commandprofile/--metadataprofile ProfileName --validate
(the --commandprofile/--metadataprofile for dumpconfig will come
as part of the subsequent patch)
- This patch also adds a reference to --commandprofile and
--metadataprofile in the cmd help string (which was missing before
for the --profile for some commands). We do not mention --profile
now as people should use --commandprofile or --metadataprofile
directly. However, the --profile is still supported for backward
compatibility and it's translated as:
--profile == --metadataprofile for lvcreate, vgcreate, lvchange and vgchange
(as these commands are able to attach profile to metadata)
--profile == --commandprofile for all the other commands
(--metadataprofile is not allowed there as it makes no sense)
- This patch also contains some cleanups to make the code handling
the profiles more readable...
2014-05-20 16:13:10 +04:00
if ( ( arg_count ( cmd , profile_ARG ) | | arg_count ( cmd , metadataprofile_ARG ) ) & &
arg_count ( cmd , detachprofile_ARG ) ) {
log_error ( " Only one of --metadataprofile and --detachprofile permitted. " ) ;
2013-06-25 14:33:06 +04:00
return EINVALID_CMD_LINE ;
}
2012-06-27 15:48:31 +04:00
if ( arg_count ( cmd , activate_ARG ) & & arg_count ( cmd , refresh_ARG ) ) {
2010-10-26 05:37:59 +04:00
log_error ( " Only one of -a and --refresh permitted. " ) ;
return EINVALID_CMD_LINE ;
}
2010-05-06 15:15:55 +04:00
if ( ( arg_count ( cmd , ignorelockingfailure_ARG ) | |
2010-10-30 01:15:23 +04:00
arg_count ( cmd , sysinit_ARG ) ) & & update ) {
2010-06-23 14:22:59 +04:00
log_error ( " Only -a permitted with --ignorelockingfailure and --sysinit " ) ;
2010-05-06 15:15:55 +04:00
return EINVALID_CMD_LINE ;
}
2012-06-27 15:48:31 +04:00
if ( arg_count ( cmd , activate_ARG ) & &
2010-10-26 05:37:59 +04:00
( arg_count ( cmd , monitor_ARG ) | | arg_count ( cmd , poll_ARG ) ) ) {
2014-05-09 18:13:48 +04:00
if ( ! is_change_activating ( ( activation_change_t ) arg_uint_value ( cmd , activate_ARG , 0 ) ) ) {
2010-10-26 05:37:59 +04:00
log_error ( " Only -ay* allowed with --monitor or --poll. " ) ;
return EINVALID_CMD_LINE ;
}
}
2010-05-06 15:15:55 +04:00
if ( arg_count ( cmd , poll_ARG ) & & arg_count ( cmd , sysinit_ARG ) ) {
log_error ( " Only one of --poll and --sysinit permitted. " ) ;
2002-11-18 17:04:08 +03:00
return EINVALID_CMD_LINE ;
}
2014-02-19 00:21:06 +04:00
if ( ( arg_count ( cmd , activate_ARG ) = = 1 ) & &
arg_count ( cmd , autobackup_ARG ) ) {
2002-11-18 17:04:08 +03:00
log_error ( " -A option not necessary with -a option " ) ;
return EINVALID_CMD_LINE ;
}
2010-10-30 01:15:23 +04:00
if ( arg_count ( cmd , maxphysicalvolumes_ARG ) & &
2012-02-28 18:24:57 +04:00
arg_sign_value ( cmd , maxphysicalvolumes_ARG , SIGN_NONE ) = = SIGN_MINUS ) {
2010-10-30 01:15:23 +04:00
log_error ( " MaxPhysicalVolumes may not be negative " ) ;
return EINVALID_CMD_LINE ;
}
if ( arg_count ( cmd , physicalextentsize_ARG ) & &
2012-02-28 18:24:57 +04:00
arg_sign_value ( cmd , physicalextentsize_ARG , SIGN_NONE ) = = SIGN_MINUS ) {
2010-10-30 01:15:23 +04:00
log_error ( " Physical extent size may not be negative " ) ;
return EINVALID_CMD_LINE ;
}
2013-11-26 17:51:23 +04:00
/*
* If - - sysinit - aay is used and at the same time lvmetad is used ,
* we want to rely on autoactivation to take place . Also , we
* need to take special care here as lvmetad service does
* not neet to be running at this moment yet - it could be
* just too early during system initialization time .
*/
if ( arg_count ( cmd , sysinit_ARG ) & & lvmetad_used ( ) & &
2012-07-10 15:49:46 +04:00
arg_uint_value ( cmd , activate_ARG , 0 ) = = CHANGE_AAY ) {
2013-11-26 17:51:23 +04:00
if ( ! lvmetad_socket_present ( ) ) {
/*
* If lvmetad socket is not present yet ,
* the service is just not started . It ' ll
* be started a bit later so we need to do
* the activation without lvmetad which means
* direct activation instead of autoactivation .
*/
log_warn ( " lvmetad is not active yet, using direct activation during sysinit " ) ;
lvmetad_set_active ( 0 ) ;
} else if ( lvmetad_active ( ) ) {
/*
* If lvmetad is active already , we want
* to make use of the autoactivation .
*/
log_warn ( " lvmetad is active, skipping direct activation during sysinit " ) ;
return ECMD_PROCESSED ;
}
2012-07-10 15:49:46 +04:00
}
2013-08-13 21:20:11 +04:00
if ( arg_count ( cmd , clustered_ARG ) & & ! argc & & ! arg_count ( cmd , yes_ARG ) & &
( yes_no_prompt ( " Change clustered property of all volumes groups? [y/n]: " ) = = ' n ' ) ) {
log_error ( " No volume groups changed. " ) ;
return ECMD_FAILED ;
}
[lv|vg]change: Allow limited metadata changes when PVs are missing
A while back, the behavior of LVM changed from allowing metadata changes
when PVs were missing to not allowing changes. Until recently, this
change was tolerated by HA-LVM by forcing a 'vgreduce --removemissing'
before trying (again) to add tags to an LV and then activate it. LVM
mirroring requires that failed devices are removed anyway, so this was
largely harmless. However, RAID LVs do not require devices to be removed
from the array in order to be activated. In fact, in an HA-LVM
environment this would be very undesirable. Device failures in such an
environment can often be transient and it would be much better to restore
the device to the array than synchronize an entirely new device.
There are two methods that can be used to setup an HA-LVM environment:
"clvm" or "tagging". For RAID LVs, "clvm" is out of the question because
RAID LVs are not supported in clustered VGs - not even in an exclusively
activated manner. That leaves "tagging". HA-LVM uses tagging - coupled
with 'volume_list' - to ensure that only one machine can have an LV active
at a time. If updates are not allowed when a PV is missing, it is
impossible to add or remove tags to allow for activation. This removes
one of the most basic functionalities of HA-LVM - site redundancy. If
mirroring or RAID is used to replicate the storage in two data centers
and one of them goes down, a server and a storage device are lost. When
the service fails-over to the alternate site, the VG will be "partial".
Unable to add a tag to the VG/LV, the RAID device will be unable to
activate.
The solution is to allow vgchange and lvchange to alter the LVM metadata
for a limited set of options - --[add|del]tag included. The set of
allowable options are ones that do not cause changes to the DM kernel
target (like --resync would) or could alter the structure of the LV
(like allocation or conversion).
2012-10-10 20:33:10 +04:00
if ( ! update | | ! update_partial_unsafe )
cmd - > handles_missing_pvs = 1 ;
2010-10-30 01:15:23 +04:00
return process_each_vg ( cmd , argc , argv , update ? READ_FOR_UPDATE : 0 ,
NULL , & vgchange_single ) ;
2002-11-18 17:04:08 +03:00
}