]> sourceware.org Git - lvm2.git/log
lvm2.git
12 years agoUse 64 bit integers for device size & label sector in _pv_populate_lvmcache in
Petr Rockai [Thu, 1 Mar 2012 20:04:44 +0000 (20:04 +0000)]
Use 64 bit integers for device size & label sector in _pv_populate_lvmcache in
lvmetad client code. Fixes RHBZ 798267.

12 years agoAdd a regression test for the PSize bug in pvs/lvmetad.
Petr Rockai [Thu, 1 Mar 2012 20:03:45 +0000 (20:03 +0000)]
Add a regression test for the PSize bug in pvs/lvmetad.

12 years agoUse 64 bit integers whenever extracting numbers from daemon replies.
Petr Rockai [Thu, 1 Mar 2012 19:54:53 +0000 (19:54 +0000)]
Use 64 bit integers whenever extracting numbers from daemon replies.

12 years agos/CPG_/CS_: Various CPG constants are going away, even though CPG itself stays
Jonathan Earl Brassow [Thu, 1 Mar 2012 17:41:39 +0000 (17:41 +0000)]
s/CPG_/CS_: Various CPG constants are going away, even though CPG itself stays

F17 is getting rid of OpenAIS libraries (and checkpointing).  While the
CPG stuff is staying, some if its constants are being removed.  So, we
must adjust and use the remaining constants which the CPG constants were based on.

[~]# egrep 'CPG_DISPATCH_ALL|CPG_OK' /usr/include/*/*
corosync/corotypes.h:#define CPG_DISPATCH_ALL     CS_DISPATCH_ALL
corosync/corotypes.h:#define CPG_OK               CS_OK

12 years agoMissed ;
Zdenek Kabelac [Thu, 1 Mar 2012 10:46:39 +0000 (10:46 +0000)]
Missed ;

12 years agoMore useful debug message
Zdenek Kabelac [Thu, 1 Mar 2012 10:41:48 +0000 (10:41 +0000)]
More useful debug message

12 years agoAdd _rimage as reserved suffix
Zdenek Kabelac [Thu, 1 Mar 2012 10:39:21 +0000 (10:39 +0000)]
Add _rimage as reserved suffix

12 years agoAdd cast for ptrdiff_t - cleanup gcc warning
Zdenek Kabelac [Thu, 1 Mar 2012 10:31:35 +0000 (10:31 +0000)]
Add cast for ptrdiff_t - cleanup gcc warning

12 years agoImprove error logging
Zdenek Kabelac [Thu, 1 Mar 2012 10:30:48 +0000 (10:30 +0000)]
Improve error logging

Log errors instead of plain return 0.
Check for f->private strdup result.

12 years agoAdd stack trace for lv_reduce
Zdenek Kabelac [Thu, 1 Mar 2012 10:09:36 +0000 (10:09 +0000)]
Add stack trace for lv_reduce

Use common code call with stack trace.

TODO: maybe the release_lv_segment_area()
should be actually able to return error code to upper level.

12 years agoCheck for version string buffer
Zdenek Kabelac [Thu, 1 Mar 2012 10:07:38 +0000 (10:07 +0000)]
Check for version string buffer

Since lvm seems to call driver_version(NULL, 0)  this would lead
to crash. Though the combination of the code is probably very hard to hit.
If the user doesn't supply version buffer, just skip printing to buffer.

12 years agoCheck for allocation error
Zdenek Kabelac [Thu, 1 Mar 2012 09:54:23 +0000 (09:54 +0000)]
Check for allocation error

return ENOMEM when malloc fails.

12 years agoLog fail of pthread_join
Zdenek Kabelac [Thu, 1 Mar 2012 09:50:04 +0000 (09:50 +0000)]
Log fail of pthread_join

Syslog error outside of mutex section.

12 years agoFix leak of FID structure
Zdenek Kabelac [Thu, 1 Mar 2012 09:46:38 +0000 (09:46 +0000)]
Fix leak of FID structure

12 years agoWipe initial 4KiB for non-zeroed thin volumes
Zdenek Kabelac [Wed, 29 Feb 2012 22:08:57 +0000 (22:08 +0000)]
Wipe initial 4KiB for non-zeroed thin volumes

If the thin pool has disabled zeroing  (created with -Zn), we at least
clear initial 4KiB of such thin volume (provisions 1st block).

If lvcreate is executed with '-an' command will abort (same way like we for
normal LV - however for normal LV option -Zn may skip clearing completely,
for thin volumes this option is not supported (applies only for pools).

12 years agoAllow cluster mirrors to handle the absence of the checkpoint lib (libSaCkpt).
Jonathan Earl Brassow [Wed, 29 Feb 2012 21:15:34 +0000 (21:15 +0000)]
Allow cluster mirrors to handle the absence of the checkpoint lib (libSaCkpt).

The OpenAIS checkpoint library is going away; therefore, cmirrord must
operate without it.  The algorithms the handle the timing of when to send
a checkpoint, the determination of what to send, and which ongoing cluster
requests are relevent with respect to the checkpoints are unaffected.  We
need only replace the functions that actually perform the storing/transmitting
and retrieving/receiving of the checkpoint data.  Rather than store the
checkpoint data in an OpenAIS checkpoint file, we simply transmit it along
with the message that notifies the incoming node that the checkpoint is
ready.

12 years agoPass 'single_device' parameter down to suppress 'Can't find uuid' messages
Alasdair Kergon [Wed, 29 Feb 2012 02:35:35 +0000 (02:35 +0000)]
Pass 'single_device' parameter down to suppress 'Can't find uuid' messages
when reading VG text metadate and called from pvscan --lvmetad.

(Longer-term, that check needs moving outside of that code.)

12 years agoFix a whitespace bug in the last checkin.
Petr Rockai [Wed, 29 Feb 2012 00:19:14 +0000 (00:19 +0000)]
Fix a whitespace bug in the last checkin.

12 years agoAttempt a fix for lvm shell accumulating copies of orphan PVs with each "pvs"
Petr Rockai [Wed, 29 Feb 2012 00:18:27 +0000 (00:18 +0000)]
Attempt a fix for lvm shell accumulating copies of orphan PVs with each "pvs"
invocation.

12 years agoexpose new include files to build
Alasdair Kergon [Tue, 28 Feb 2012 18:35:32 +0000 (18:35 +0000)]
expose new include files to build

12 years agoReflect new file locations, include file updates etc.
Alasdair Kergon [Tue, 28 Feb 2012 18:35:04 +0000 (18:35 +0000)]
Reflect new file locations, include file updates etc.

12 years agoremove old makefile
Alasdair Kergon [Tue, 28 Feb 2012 18:31:28 +0000 (18:31 +0000)]
remove old makefile

12 years agomove daemon files
Alasdair Kergon [Tue, 28 Feb 2012 18:30:39 +0000 (18:30 +0000)]
move daemon files

12 years agonew makefiles
Alasdair Kergon [Tue, 28 Feb 2012 18:26:40 +0000 (18:26 +0000)]
new makefiles

12 years agoSeries of commits to move daemon code into separate top-level dir and
Alasdair Kergon [Tue, 28 Feb 2012 18:25:48 +0000 (18:25 +0000)]
Series of commits to move daemon code into separate top-level dir and
fix build deps etc.

12 years agoInvert some return codes (which used to get ignored anyway) in cases when
Alasdair Kergon [Tue, 28 Feb 2012 18:23:56 +0000 (18:23 +0000)]
Invert some return codes (which used to get ignored anyway) in cases when
lvmetad is not being used.

12 years agoBypass lvmetad fns when not configured to use lvmetad.
Alasdair Kergon [Tue, 28 Feb 2012 18:22:52 +0000 (18:22 +0000)]
Bypass lvmetad fns when not configured to use lvmetad.

12 years agoAdd LVMETAD_SUPPORT
Alasdair Kergon [Tue, 28 Feb 2012 18:20:58 +0000 (18:20 +0000)]
Add LVMETAD_SUPPORT

12 years agoIf lvmetad fails, don't set _has_scanned
Alasdair Kergon [Tue, 28 Feb 2012 18:20:17 +0000 (18:20 +0000)]
If lvmetad fails, don't set _has_scanned

12 years agoDefine LVMETAD_SUPPORT in source code
Alasdair Kergon [Tue, 28 Feb 2012 18:18:48 +0000 (18:18 +0000)]
Define LVMETAD_SUPPORT in source code

12 years agospacing
Alasdair Kergon [Tue, 28 Feb 2012 18:17:37 +0000 (18:17 +0000)]
spacing

12 years agoI don't think -u is meant to be used with lvmetad
Alasdair Kergon [Tue, 28 Feb 2012 18:16:23 +0000 (18:16 +0000)]
I don't think -u is meant to be used with lvmetad

12 years agoCheck return values after calling new lvmetad fns
Alasdair Kergon [Tue, 28 Feb 2012 18:08:08 +0000 (18:08 +0000)]
Check return values after calling new lvmetad fns
(Haven't checked error path handling though)

12 years agoFix empty string warning logic in _find_config_str. (1.02.68)
Alasdair Kergon [Tue, 28 Feb 2012 17:46:47 +0000 (17:46 +0000)]
Fix empty string warning logic in _find_config_str. (1.02.68)

pvcreate gives
WARNING: Ignoring unsupported value for metadata/pvmetadataignore.

It was warning if there is no config file entry instead of only if the node
exists but is empty.

12 years agoCheck for failing dm_asprintf
Zdenek Kabelac [Tue, 28 Feb 2012 14:25:37 +0000 (14:25 +0000)]
Check for failing dm_asprintf

12 years agoUsing enum types for enums
Zdenek Kabelac [Tue, 28 Feb 2012 14:24:57 +0000 (14:24 +0000)]
Using enum types for enums

alloc_policy_t, dm_string_mangling_t, percent_range_t, sign_t

12 years agoRevert free of allocated segtype
Zdenek Kabelac [Tue, 28 Feb 2012 14:23:41 +0000 (14:23 +0000)]
Revert free of allocated segtype

lvm_register_segtype takes ownership of segtype and call destructor
for it in error path.

12 years agoLVMetaD socket operates in non-blocking mode.
Peter Rajnoha [Tue, 28 Feb 2012 13:07:03 +0000 (13:07 +0000)]
LVMetaD socket operates in non-blocking mode.

12 years agoAdd support for systemd socket handover for common daemon-server code and also add...
Peter Rajnoha [Tue, 28 Feb 2012 13:05:21 +0000 (13:05 +0000)]
Add support for systemd socket handover for common daemon-server code and also add support for new OOM killer adjustment interface.

This code is already a part of dmeventd, but it's modified slightly to check
sockets instead of FIFOs.

12 years agoTest dm_hash_insert() failures mem failures
Zdenek Kabelac [Tue, 28 Feb 2012 11:12:58 +0000 (11:12 +0000)]
Test dm_hash_insert() failures mem failures

12 years agoCheck error from _lv_each_dependency
Zdenek Kabelac [Tue, 28 Feb 2012 11:10:45 +0000 (11:10 +0000)]
Check error from _lv_each_dependency

_lv_mark_if_partial_collect cannot fail, however it's good to keep
checking here as we do in all other cases.

12 years agoCheck result of export_vg_to_buffer
Zdenek Kabelac [Tue, 28 Feb 2012 11:09:06 +0000 (11:09 +0000)]
Check result of export_vg_to_buffer

and avoid usage of failed buffer.

12 years agoAdd stacktrace for init_lvmcache_orphans fail path
Zdenek Kabelac [Tue, 28 Feb 2012 11:08:04 +0000 (11:08 +0000)]
Add stacktrace for init_lvmcache_orphans fail path

12 years agoEnsure clvmd message is always \0 terminated
Zdenek Kabelac [Tue, 28 Feb 2012 11:06:56 +0000 (11:06 +0000)]
Ensure clvmd message is always \0 terminated

Drop whole buffer clearing (most messages at <100 bytes).
Just make sure we have always \0 terminated string for strlen() operations.
(before for PIPE_BUF sized messages this was not set).

12 years agoBetter detection of missing dmeventd fifo connection
Zdenek Kabelac [Tue, 28 Feb 2012 11:03:24 +0000 (11:03 +0000)]
Better detection of missing dmeventd fifo connection

12 years agoMissed to properly merge patch for definition of MAX_MISSING_LEN
Zdenek Kabelac [Tue, 28 Feb 2012 10:42:20 +0000 (10:42 +0000)]
Missed to properly merge patch for definition of MAX_MISSING_LEN

(fix previous commit)

12 years agoAdd log_sys_error for close
Zdenek Kabelac [Tue, 28 Feb 2012 10:14:06 +0000 (10:14 +0000)]
Add log_sys_error for close

12 years agoAdd some stack traces for dev_close error paths
Zdenek Kabelac [Tue, 28 Feb 2012 10:11:35 +0000 (10:11 +0000)]
Add some stack traces for dev_close error paths

12 years agoExplicitely ignore errors in label_read iteration.
Zdenek Kabelac [Tue, 28 Feb 2012 10:10:07 +0000 (10:10 +0000)]
Explicitely ignore errors in label_read iteration.

We don't want stack traces here, so mark as void.

12 years agoExplicitely check list size of segments
Zdenek Kabelac [Tue, 28 Feb 2012 10:08:20 +0000 (10:08 +0000)]
Explicitely check list size of segments

instead of checking for NULL from last_seg and first_seg.

12 years agoDuplicate standard in/out descriptors for daemon
Zdenek Kabelac [Tue, 28 Feb 2012 10:06:53 +0000 (10:06 +0000)]
Duplicate standard in/out descriptors for daemon

Addressing somewhat tricky bug here.
Since stdin,stdout,stderr were closed it's been occasionally possible to
see some unexpected messages to be flowing into a clvmd and generating some
randomly sized allocation of many megabytes. Since the message was not
being generated by standard send_message() construction, after some more
testing it apperead to be a debug log message - thus something has flown
to local socket opened on strandard out descriptor.

To fix the issue - use standard file descriptor duplication code for daemons.

For making easier debugging of polling daemon - developer might want to recompile
without modifition of standard file descriptors.

12 years agoLimit max size of clvmd message
Zdenek Kabelac [Tue, 28 Feb 2012 09:58:19 +0000 (09:58 +0000)]
Limit max size of clvmd message

This could be seen as some sort of simple validation - it's not easy to
recognize a valid message for now - but we definitely do not want to
allocate a lot of megabytes in  clvmd memory locked daemon when broken
message gets in.

Size of 8000 is just selected for now - possibly there could be much
lower value put in.

12 years agoDo not send uninitilised bytes
Zdenek Kabelac [Tue, 28 Feb 2012 09:53:55 +0000 (09:53 +0000)]
Do not send uninitilised bytes

Use struct initalizers to fill struct members and at the same time have
all unspecified members set to 0.

12 years agoFix dm_task_set_name to properly resolve path to dm name.
Peter Rajnoha [Tue, 28 Feb 2012 08:36:51 +0000 (08:36 +0000)]
Fix dm_task_set_name to properly resolve path to dm name.

Wrong variable was checked in _dm_task_set_name_from_path.

12 years agolvmetad.socket
Peter Rajnoha [Mon, 27 Feb 2012 15:45:49 +0000 (15:45 +0000)]
lvmetad.socket

12 years agoDo not hide deallocation of buffer
Zdenek Kabelac [Mon, 27 Feb 2012 11:49:16 +0000 (11:49 +0000)]
Do not hide deallocation of buffer

As API is passing structures by value, do not leave
the function which created buffer and keeps valid pointer
look like it would be some memory leak and move
free of buffer from inner function - makes more obvious,
how is the memory management handled.

12 years agoAdd FIXME for misuse in case of dm_event_mask enum.
Zdenek Kabelac [Mon, 27 Feb 2012 11:46:25 +0000 (11:46 +0000)]
Add FIXME for misuse in case of dm_event_mask enum.

Using enum for bitmask is not proper solution here.

12 years agoUse unsigned type for bitmask
Zdenek Kabelac [Mon, 27 Feb 2012 11:45:05 +0000 (11:45 +0000)]
Use unsigned type for bitmask

Using report_type_t for bitmask is not correct, since we have not defined types
for all  bit combinations - so switching to unsigned type, since values of
report_type_t enum are unsigned.

12 years agoJust code move of hash initialization in front of function
Zdenek Kabelac [Mon, 27 Feb 2012 11:40:58 +0000 (11:40 +0000)]
Just code move of hash initialization in front of function

Make sure both hash tables are initialized before _read_sections() call.
Presents no functional change (since PV scan phase was not adding LV hashes),
but makes the code easier to handle mem failing case, and static analyzer is
hapier as well.

12 years agoAdd stack traces for lock_vol failures
Zdenek Kabelac [Mon, 27 Feb 2012 11:35:59 +0000 (11:35 +0000)]
Add stack traces for lock_vol failures

Adding at least stack traces with some FIXMEs for cases,
where we might want to do something cleaver - maybe fail command
or give user hints something is not going well ?

For remote_backup is stack probably 'good' enough for now.

12 years agoCheck id_write_format result
Zdenek Kabelac [Mon, 27 Feb 2012 11:32:48 +0000 (11:32 +0000)]
Check id_write_format result

Currently we never fail with 64byte uuid buffer,
but just stay consitent with rest of the code and check for result.

12 years agoExplicitely ignore fail error on hash_insert
Zdenek Kabelac [Mon, 27 Feb 2012 11:31:30 +0000 (11:31 +0000)]
Explicitely ignore fail error on hash_insert

We cannot do anything better here anyway - we are already in logging function,
so just ignore this issue here - it will most likely stop application later.

12 years agoAdd some log_sys_errors to close() call
Zdenek Kabelac [Mon, 27 Feb 2012 11:28:47 +0000 (11:28 +0000)]
Add some log_sys_errors to close() call

12 years agoNicer cleanup of excl_uuid hash
Zdenek Kabelac [Mon, 27 Feb 2012 11:26:25 +0000 (11:26 +0000)]
Nicer cleanup of excl_uuid hash

Since it on exit path, it's not a big difference,
but makes less noise in analyzer and valgrind.

12 years agoCheck for vg_name existance
Zdenek Kabelac [Mon, 27 Feb 2012 11:23:15 +0000 (11:23 +0000)]
Check for vg_name existance

Since vg_read() mda ops could be called with NULL vg_name,
check it before derefence also for pool and format1.

12 years agoSystemd is linux-specific - move the supporting code under the 'ifdef linux'.
Peter Rajnoha [Mon, 27 Feb 2012 11:13:51 +0000 (11:13 +0000)]
Systemd is linux-specific - move the supporting code under the 'ifdef linux'.

Some 'defines' used in this specific code were already under an ifdef so this
patch just completes it.

12 years agoFix missing break
Zdenek Kabelac [Mon, 27 Feb 2012 11:13:48 +0000 (11:13 +0000)]
Fix missing break

Bug introduced with addition of internal error default case.
Seem like this code is not used.
TODO: add coverage test.

12 years agoAdd assert for oldname
Zdenek Kabelac [Mon, 27 Feb 2012 10:19:00 +0000 (10:19 +0000)]
Add assert for oldname

Code cannot proceed if oldname would be NULL.
Since lvmetad currently doesn't use logging mechanism of lvm to report
internal errors - stay with current code style of lvmetad which uses
plain asserts for cases like this.

12 years agoA bit more safe version of sprintf
Zdenek Kabelac [Mon, 27 Feb 2012 10:17:06 +0000 (10:17 +0000)]
A bit more safe version of sprintf

Use just buffer size limit (it's used for debug only)

12 years agoTest seg pointer for non-null
Zdenek Kabelac [Mon, 27 Feb 2012 10:15:08 +0000 (10:15 +0000)]
Test seg pointer for non-null

As the function accepts NULL for 'seg' parameter,
check for it before dereference.

12 years agoMove allocation after check for vgid
Zdenek Kabelac [Mon, 27 Feb 2012 10:10:43 +0000 (10:10 +0000)]
Move allocation after check for vgid

so there is no mem leak on this error path.
Also actually check if the hash exists.

12 years agoMake force_t enum usage obvious
Zdenek Kabelac [Mon, 27 Feb 2012 10:06:58 +0000 (10:06 +0000)]
Make force_t enum usage obvious

12 years agoTest result of _init_tags.
Zdenek Kabelac [Mon, 27 Feb 2012 10:05:35 +0000 (10:05 +0000)]
Test result of _init_tags.

12 years agoExplicitely ignore result from sync_unlock()
Zdenek Kabelac [Mon, 27 Feb 2012 10:02:17 +0000 (10:02 +0000)]
Explicitely ignore result from sync_unlock()

Make it obvious to analyzer - we can't do anything better here anyway.

12 years agoAlways check result of _set_vg_name()
Zdenek Kabelac [Mon, 27 Feb 2012 10:00:23 +0000 (10:00 +0000)]
Always check result of _set_vg_name()

12 years agoDrop uname call, it's not used from gulm era.
Zdenek Kabelac [Mon, 27 Feb 2012 09:58:18 +0000 (09:58 +0000)]
Drop uname call, it's not used from gulm era.

12 years agoCheck allocation result
Zdenek Kabelac [Mon, 27 Feb 2012 09:56:27 +0000 (09:56 +0000)]
Check allocation result

12 years agoRemove test for pvid
Zdenek Kabelac [Mon, 27 Feb 2012 09:54:25 +0000 (09:54 +0000)]
Remove test for pvid

Since pvid is char buffer[] and not pointer,
there is no point to check it for NULL.

12 years agoIgnore result of unlock explicitely
Zdenek Kabelac [Mon, 27 Feb 2012 09:52:37 +0000 (09:52 +0000)]
Ignore result of unlock explicitely

12 years agoJust reindent with tabs
Zdenek Kabelac [Mon, 27 Feb 2012 09:51:31 +0000 (09:51 +0000)]
Just reindent with tabs

12 years agoUse 'killproc' fn in lvmetad init script (fn provided by 'functions' init script...
Peter Rajnoha [Mon, 27 Feb 2012 08:56:46 +0000 (08:56 +0000)]
Use 'killproc' fn in lvmetad init script (fn provided by 'functions' init script library).

12 years agoFix a possible SEGV in lvmetad client code, and a minor leak in pvscan
Petr Rockai [Sun, 26 Feb 2012 13:42:50 +0000 (13:42 +0000)]
Fix a possible SEGV in lvmetad client code, and a minor leak in pvscan
--lvmetad.

12 years agoA bit of documentation for lvmetad_vg_lookup.
Petr Rockai [Sun, 26 Feb 2012 08:50:13 +0000 (08:50 +0000)]
A bit of documentation for lvmetad_vg_lookup.

12 years ago- Improve error reporting on lvmetad connection failure.
Petr Rockai [Sun, 26 Feb 2012 08:49:40 +0000 (08:49 +0000)]
- Improve error reporting on lvmetad connection failure.
- Fix a couple of memory leaks in the lvmetad client code.
- Avoid an error in lvmetad_pv_gone when we aren't using lvmetad.

12 years agoImprove error handling & reporting in common daemon code.
Petr Rockai [Sun, 26 Feb 2012 08:46:28 +0000 (08:46 +0000)]
Improve error handling & reporting in common daemon code.

12 years agoAdd tentative documentation of pvscan --lvmetad to the pvscan manpage.
Petr Rockai [Sun, 26 Feb 2012 08:45:07 +0000 (08:45 +0000)]
Add tentative documentation of pvscan --lvmetad to the pvscan manpage.

12 years agoAlso remove intermediate lvmetad init script file.
Peter Rajnoha [Fri, 24 Feb 2012 15:07:00 +0000 (15:07 +0000)]
Also remove intermediate lvmetad init script file.

12 years agoUse '-p PID_FILE' for the status call since the pidfile is in its own lvm subdir...
Peter Rajnoha [Fri, 24 Feb 2012 13:45:37 +0000 (13:45 +0000)]
Use '-p PID_FILE' for the status call since the pidfile is in its own lvm subdir and return proper status code.

12 years agoThe daemon as well as the first scan must be both executed properly, otherwise it...
Peter Rajnoha [Fri, 24 Feb 2012 13:11:57 +0000 (13:11 +0000)]
The daemon as well as the first scan must be both executed properly, otherwise it's an error!

12 years agoAdd LVMetaD init script.
Peter Rajnoha [Fri, 24 Feb 2012 13:03:50 +0000 (13:03 +0000)]
Add LVMetaD init script.

12 years agoAdd skeleton for lvmetad udev rules - 69-dm-lvm-metad.rules.
Peter Rajnoha [Fri, 24 Feb 2012 09:53:12 +0000 (09:53 +0000)]
Add skeleton for lvmetad udev rules - 69-dm-lvm-metad.rules.

Why using the order 69:

  - Storage processing in general happens in 60-persistent-storage.rules,
    including the blkid call that adds some usable information we can use
    for filtering and speedup (these rules are part of upstream udev and
    the order is preserved on most distros)

  - There's still some other storage-related processing done after
    60-persistent-storage.rules in general. These might add some detailed
    storage-related information we might use to filter devices effectively
    (e.g. MD udev rules, ...).

  - We need lvmetad rules to be processed before any consumers can use the
    output - so the metadata cache is ready soon enough (e.g. udisks rules).

  - There's no official (upstream udev) document about assigning the order,
    so this number is chosen in best belief it will suit all scenarios.

12 years agoFix server-side leaks in lvmetad.
Petr Rockai [Fri, 24 Feb 2012 00:24:37 +0000 (00:24 +0000)]
Fix server-side leaks in lvmetad.

12 years agoClean up the lvmetad state more thoroughly upon shutdown.
Petr Rockai [Fri, 24 Feb 2012 00:11:59 +0000 (00:11 +0000)]
Clean up the lvmetad state more thoroughly upon shutdown.

12 years agoFix an error path in daemon_open.
Petr Rockai [Fri, 24 Feb 2012 00:02:54 +0000 (00:02 +0000)]
Fix an error path in daemon_open.

12 years agoMake the calls to pvscan --lvmetad in the testsuite slightly more civilised.
Petr Rockai [Thu, 23 Feb 2012 23:58:42 +0000 (23:58 +0000)]
Make the calls to pvscan --lvmetad in the testsuite slightly more civilised.

12 years agoCouple of improvements in the daemon (common + lvmetad) code:
Petr Rockai [Thu, 23 Feb 2012 23:52:11 +0000 (23:52 +0000)]
Couple of improvements in the daemon (common + lvmetad) code:
- some client-side memory leak fixes
- announce and check protocols and protocol versions

12 years agoLimit sscanf params with size
Zdenek Kabelac [Thu, 23 Feb 2012 22:50:50 +0000 (22:50 +0000)]
Limit sscanf params with size

Make sure parsed string fits given char buffer.

12 years agoUnit testing for some string libdm functions
Zdenek Kabelac [Thu, 23 Feb 2012 22:47:17 +0000 (22:47 +0000)]
Unit testing for some string libdm functions

TODO: more functions need to be unit tested.

12 years agoIntroduce dm_strncpy
Zdenek Kabelac [Thu, 23 Feb 2012 22:45:43 +0000 (22:45 +0000)]
Introduce dm_strncpy

Should be faster then strncpy - since we could avoid clearing 4KB pages
with each strncpy(...,PATH_MAX).
Also it's easy to check whether string fit - and eventually avoid
to continue working we incomplete string.

12 years agoUse const for lv
Zdenek Kabelac [Thu, 23 Feb 2012 22:41:57 +0000 (22:41 +0000)]
Use const for lv

lv_is_active doesn't needs modifiable LV struct so keep it const.

Remove lv_send_message() left bits from code -
they were never released in 2.02.89.

This page took 0.064918 seconds and 5 git commands to generate.