1
0
mirror of git://sourceware.org/git/lvm2.git synced 2025-01-03 05:18:29 +03:00
Commit Graph

445 Commits

Author SHA1 Message Date
Zdenek Kabelac
639983b6b7 clvmd: skip adding reply when finished
Prior adding new reply to the list, check
if the reply thread is not already finished.
In that case discard adding message
(which would otherwise be leaked).
2014-04-14 13:01:42 +02:00
Zdenek Kabelac
7236b92857 clvmd: improve mutex usage in request_timed_out
Use mutex to access localsock values, so check
num_replies when the thread is not yet finished.

Check for threadid prior the mutex taking
(though this check is probably not really needed)
2014-04-14 13:00:51 +02:00
Zdenek Kabelac
7075656034 clvmd: drop reply_mutex
Added complexity with extra reply mutex is not worth the troubles.
The only place which may slightly benefit from this mutex is timeout
and since this is rather error case - let's convert it to
localsock.mutex and keep it simple.
2014-04-14 12:59:07 +02:00
Zdenek Kabelac
6115c0d112 clvmd: set finished flag with mutex
Setting this variable needs to be protected with mutex.
2014-04-14 12:58:28 +02:00
Zdenek Kabelac
cc0096ebdd clvmd: move mutex init and detroy
Move the pthread mutex and condition creation and destroy
to correct place right after client memory is allocatedd
or is going to be released.

In the original place it's been in race with lvm thread
which could have still unlock mutex while it's been already
destroyed.
2014-04-14 12:57:39 +02:00
Zdenek Kabelac
91f4e09b48 clvmd: fix test mode race
When TEST_MODE flag is passed around the cluster,
it's been use in thread unprotected way, so it may have
influenced behaviour of other running parallel lvm commands
(activation/deactivation/suspend/resume).

Fix it by set/query function only under lvm mutex.
For hold_un/lock function calls check lock_flags bits directly.
2014-04-14 12:55:46 +02:00
Alasdair G Kergon
c16c1a9f70 clvmd: Update new remove_info INTERNAL_ERRORS. 2014-04-04 01:37:43 +01:00
Zdenek Kabelac
b714c7ebc6 clvmd: validate open device state
If clvmd does not hold any lock, it should also not keep any opened
device.

The reason for this patch is, that refresh_toolcontext calls
dev_cache_exit() which destroys whole device cache (even those with
opened file) - previous patch added recovery path to avoid memory
corruption, but opened files are still bugs that need to be fixed.

So this patch certainly kills many internal mirror & raid tests,
since they leak opened file descriptors (when tests are executed
with 'abort_on_error').
2014-03-25 11:22:57 +01:00
Zdenek Kabelac
b522312678 clvmd: hardening leak on exit
Operate with lvm_thread_exit while holding lvm_thread_mutex.
Don't leave unfinished work in the lvm thread queue
and always finish all queued tasks before exit,
so no cmd struct is left in the list.

(in-release fix)
2014-03-25 11:22:57 +01:00
Zdenek Kabelac
01efb20bdb cleanup: clvmd add more info debug message 2014-03-21 22:29:27 +01:00
Zdenek Kabelac
0d449fe183 cleanup: clvmd uses struct initializers
Code easier to read
2014-03-21 22:29:27 +01:00
Zdenek Kabelac
9196274c1e cleanup: clvmd zombie removal loop
Simplier code to start scan from the next node,
and remove matching pipe client.
2014-03-21 22:29:27 +01:00
Zdenek Kabelac
5740c00f3b cleanup: clvmd reindent read_from_local_sock
Shift indent of else branch to right since
error path returns in the front.
(Simplier to read)
2014-03-21 22:29:26 +01:00
Zdenek Kabelac
dd17286c90 cleanup: clvmd indent change
Plain indent changes.
2014-03-21 22:29:26 +01:00
Zdenek Kabelac
0b79979bb9 cleanup: clvmd dump_messages 2014-03-21 22:29:26 +01:00
Zdenek Kabelac
f8cd435cd8 cleanup: clvmd update log_error
Drop \n from log_error() and add '.'
2014-03-21 22:29:26 +01:00
Zdenek Kabelac
4c97ea2ce5 cleanup: clvmd drop goto 2014-03-21 22:29:25 +01:00
Zdenek Kabelac
7d49c33ffc cleanup: clvmd skip reset of null pointer 2014-03-21 22:29:25 +01:00
Zdenek Kabelac
bf19c6be2c cleanup: clvmd move destroy_lvhash into main
Keep destruction code path consistent and simple and
destroy lvhash in the place it's been created.
Also issue debug message from a single place.
2014-03-21 22:29:25 +01:00
Zdenek Kabelac
46ae028cd3 singlenode: reset pointer when hash is destroyed
Just keep pointer consistent.
2014-03-21 22:29:25 +01:00
Zdenek Kabelac
7041c8bf51 clvmd: drop dead code
Since pipe_client has been already reset to NULL,
remove this dead code.
2014-03-21 22:29:24 +01:00
Zdenek Kabelac
73978f8d7d clvmd: drop unused local_sock parameter 2014-03-21 22:29:24 +01:00
Zdenek Kabelac
37396e2fe5 clvmd: update add_reply_to_list
Take mutex lock after the allocation just before
the structure is merged into reply list.
2014-03-21 22:29:24 +01:00
Zdenek Kabelac
2847994624 clvmd: use dm_malloc
Use libdm malloc wrapper functions.
2014-03-21 22:29:24 +01:00
Zdenek Kabelac
c45cd6eb8d clvmd: move call of cluster_closedown
We have to close cluster in some predicatable way,
otherwise we may access released memory from different
threads.

So move closedown till the point we know all thread
are closed. New messages from cluster are discarded.
2014-03-21 22:29:23 +01:00
Zdenek Kabelac
fc39ad677b clvmd: move destroy_lvm into lvm thread
Since lvm was initialized in lvm thread call destroy there as well.
2014-03-21 22:29:23 +01:00
Zdenek Kabelac
8431d47b3b clvmd: add special lvm thread exit
When multiple threads act on the same 'quit' variable
the order of exit becomes unpredictable.

So let the main_loop() finish first and then clean up
all queued lvm jobs.

Do not add any new work, when lvm_thread_exit is set.
2014-03-21 22:29:23 +01:00
Zdenek Kabelac
05a532e171 clvmd: always set cleanup_needed
When thread is created, set the cleanup_needed flag
immediately so it could be used properly any time later
when cleanup_zombie() is needed.
2014-03-21 22:29:23 +01:00
Zdenek Kabelac
5bea2b5c82 clvmd: fix clean memory on exit
Properly clean 'client' structure only for LOCAL_SOCK type.
(Fixes bug from commit 460c19df62)
(in release fix)

Also cleanup-up associated pthreads by using cleanup_zombie() function.
Since this function may change the list, restart scanning always from
the list header.

Note: couple following patches are necessary to make this working properly.
2014-03-21 22:29:22 +01:00
Zdenek Kabelac
6c892be4a5 clvmd: fix the len setting in last commit
When theoretically sending 0 length buffer, ensure 0 is returned.
2014-03-10 12:35:44 +01:00
Zdenek Kabelac
341055ddb3 cleanup: clvmd use struct initializers
Simplify code, and use compiler capability to
initilize struct members on its own.
2014-03-10 12:24:10 +01:00
Zdenek Kabelac
c44ede31a3 cleanup: clvmd use else if
When string already matches, skip rest of tests.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
2cfe0840d2 cleanup: clvmd assign NULL to pointers
Use NULL when assigning to void*.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
c824369fbd cleanup: clvmd indent changes
Improve readbility of clvmd code.
Remove some unneeded braces.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
109564d6a5 cleanup: clvmd simplify loops
Rewrite write loops to be more readable.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
3d23404081 cleanup: clvmd rewrite send_message
Improve readability of code and make it clear what it tries
to achieve.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
cbca815dc4 cleanup: clvmd reindent lock_vg code
Code had wrong indent level, improve readability.
2014-03-10 12:24:09 +01:00
Zdenek Kabelac
460c19df62 clvmd: fix memleak on exit
This patch will releases allocated private resources from
startup. Needs previous dm_zalloc patch to ensure unset
private pointer is NULL.

TODO: check on real cluster.
2014-03-10 12:21:32 +01:00
Zdenek Kabelac
38ce06e448 clvmd: use dm_zalloc for socket allocation
Instead of doing individual settings for struct members,
ensure whole struct is in defined state.
2014-03-10 12:20:49 +01:00
Zdenek Kabelac
760714829b cleanup: skip double assign
Assing NULL to type only in defaut: switch.
Debug print '--'  for  unlocked and unused resource (-1)
2013-12-17 14:08:54 +01:00
Zdenek Kabelac
30a81e5989 cleanup: self compilable headers 2013-12-12 13:28:19 +01:00
Peter Rajnoha
481edce41f compile/link: use RELRO/PIE compiler/linker options for executables 2013-12-05 14:03:10 +01:00
Peter Rajnoha
a65ab773b4 daemons: use PIE and RELRO compiler/linker options
The PIE and RELRO compiler/linker options can be used to produce a code
some techniques applied that makes the code more immune to some attacks:

  - PIE (Position Independent Executable). It can make use of the ASLR
    (Address Space Layout Randomization) provided by kernel to avoid
    static locations for .text regions of executables (this is the 'pie'
    compiler and linker option)

  - RELRO (Relocation Read-Only). This prevents overwrite attacks of
    the GOT (Global Offset Table) and PLT (Procedure Lookup Table)
    used for relocations by making it read-only after all relocations
    are resolved (these are the 'relro' and 'now' linker options) -
    hence all symbols are resolved at the very start so there's no
    need for those tables to be writeable later.

These compiler/linker options are now used by default for daemons
if the compiler/linker supports it.
2013-12-04 13:30:08 +01:00
Zdenek Kabelac
50e1fad86a cleanup: use matching signed types 2013-11-28 12:47:51 +01:00
Peter Rajnoha
b6b5299d1e corosync: fix some gcc warnings
warning: function declaration isn't a prototype [-Wstrict-prototypes]
warning: old-style function definition [-Wold-style-definition]
2013-11-06 14:55:18 +01:00
David Teigland
4c0db84948 clvmd: fix verify message rejection of REMOTE flag
This fixes a bug in commit 19baf842 where verify_message
was rejecting the CLVMD_FLAG_REMOTE flag.  It was missed
since the patch was ported from an lvm version where that
flag does not exist.
2013-10-24 11:18:22 -05:00
Peter Rajnoha
039bdad732 activation: flag temporary LVs internally
Add LV_TEMPORARY flag for LVs with limited existence during command
execution. Such LVs are temporary in way that they need to be activated,
some action done and then removed immediately. Such LVs are just like
any normal LV - the only difference is that they are removed during
LVM command execution. This is also the case for LVs representing
future pool metadata spare LVs which we need to initialize by using
the usual LV before they are declared as pool metadata spare.

We can optimize some other parts like udev to do a better job if
it knows that the LV is temporary and any processing on it is just
useless.

This flag is orthogonal to LV_NOSCAN flag introduced recently
as LV_NOSCAN flag is primarily used to mark an LV for the scanning
to be avoided before the zeroing of the device happens. The LV_TEMPORARY
flag makes a difference between a full-fledged LV visible in the system
and the LV just used as a temporary overlay for some action that needs to
be done on underlying PVs.

For example: lvcreate --thinpool POOL --zero n -L 1G vg

- first, the usual LV is created to do a clean up for pool metadata
  spare. The LV is activated, zeroed, deactivated.

- between "activated" and "zeroed" stage, the LV_NOSCAN flag is used
  to avoid any scanning in udev

- betwen "zeroed" and "deactivated" stage, we need to avoid the WATCH
  udev rule, but since the LV is just a usual LV, we can't make a
  difference. The LV_TEMPORARY internal LV flag helps here. If we
  create the LV with this flag, the DM_UDEV_DISABLE_DISK_RULES
  and DM_UDEV_DISABLE_OTHER_RULES flag are set (just like as it is
  with "invisible" and non-top-level LVs) - udev is directed to
  skip WATCH rule use.

- if the LV_TEMPORARY flag was not used, there would normally be
  a WATCH event generated once the LV is closed after "zeroed"
  stage. This will make problems with immediated deactivation that
  follows.
2013-10-23 14:09:37 +02:00
Peter Rajnoha
ce7489ed22 activation: add support for flagging an LV to skip udev scanning during activation
A common scenario is during new LV creation when we need to wipe the
newly created LV and avoid any udev scanning before this stage otherwise
it could cause the device (the LV) to be claimed by some other subsystem
for which there were stale metadata within LV data.

This patch adds possibility to mark the LV we're just about to wipe with
a flag that gets passed to udev via DM_COOKIE as a subsystem specific
flag - DM_SUBSYSTEM_UDEV_FLAG0 (in this case the subsystem is "LVM")
so LVM udev rules will take care of handling that.
2013-10-08 13:43:14 +02:00
Christine Caulfield
431eda63cc clvmd: Fix node up/down handing in corosync module
The corosync cluster interface for clvmd did not correctly
deal with node up/down events so that when a node was removed
from the cluster clvmd would prevent remote operations
from happening, as it thought the node was up but not
running clvmd.

This patch fixes that code by simplifying the case to node
being  up or down - which was the original intention
and is supported by pacemaker and CPG in the higher layers.

Signed-off-by: Christine Caulfield <ccaulfie@redhat.com>
2013-09-23 13:23:00 +01:00
Petr Rockai
25bed99681 clvmd: Avoid a 3-way deadlock in dead-client cleanup. 2013-09-18 21:17:48 +02:00