Hi there,
I am currently trying to configure Pacemaker/Corosync. I managed to install
the required packages for the cluster configuration, however I could not
start the cluster service. Based on the log file, there was an issue with
the directory /var/lib/pacemaker/.
I have tried some suggestion
Ajit,
Ajit Singh1 via Openais napsal(a):
Hi All,
I am facing a issue in my application, It is unable to wake up due to ais
faliure.
I am getting following error in my OS(system log).
==
*ais.service: main process exited, code=exited, status=254*
*Unit ais.service entered failed state*
The
Michael,
Michael Weiner napsal(a):
Hello all,
Just to start we are running a custom script at boot to change our
corosync.conf configuration files on boot based on a IP ping of the
gateway.
The corosync script being run after boot should be run simultaneously if
configurations are changing, an
What version of libqb are you using? There is a bug with
libqb-0.17.1-1.el7_1.1
(https://bugzilla.redhat.com/show_bug.cgi?id=1211375), maybe you are
hitting it.
Regards,
Honza
Денис _ napsal(a):
Coredump for problem from post
http://lists.linuxfoundation.org/pipermail/openais/2015-March/01
Денис _ napsal(a):
> Hi. We have Pacemaker+Corosync cluster with 5 nodes (Centos 7). Everything
> works but
> 1. sometimes (1 time per month) Corosync segfaults
>
> Feb 27 17:34:34 cloudnode-2-2 kernel: corosync[3293]: segfault at 0 ip
> (null) sp 7fff2ac84fb8 error 14 in
> qb-cpg-event-
This is really question for pacemaker list, so CCing.
Regards,
Honza
Redeye napsal(a):
> I am not certain where I should post this, hopefully someone will point me in
> the right direction.
>
> I have a two node cluster on Ubuntu 12.04, corosync, pacemaker, and squid.
> Squid is not startin
This is really question for pacemaker list, so CCing.
Regards,
Honza
Vladimir Berezovski (vberezov) napsal(a):
Hi ,
I added a new resourse like
crm(live)configure# primitive p_drbd_ora ocf:linbit:drbd params
drbd_resource="clusterdb_res_ora" op monitor interval="60s"
but its status is F
Daniel,
glad you found out solution and thanks for sharing.
Regards,
Honza
> Daniel Dehennin writes:
>
>> The routing is local to the VM using corosync to be sure packets
>> are outputed by eth1.
>>
>>> - Most routers/clever switches blocks multicast/don't allow
>>> multicast routing. I don'
Daniel,
Hello,
I'm trying to setup a pacemaker/corosync on Ubuntu Trusty to access a
SAN to use with OpenNebula[1]:
- pacemaker .1.10+git20130802-1ubuntu2.1
- corosync 2.3.3-1ubuntu1
I have a dedicated VLAN for cluster communications.
Each bare metal node have a dedicated interface eth0 on
Stefan,
sending to Pacemaker list because your question seems to be not Corosync
related.
Regards,
Honza
Senftleben, Stefan (itsc) napsal(a):
Hello,
I set the cluster in a maintainance mode with: crm configure property
maintenance-mode=true .
Afterwards I did stop one resource manually, b
Per,
it looks like none of your question is really corosync related (so I'm
CC'ing linux clustering (this is really
better list) but I will try to answer at least some of your questions.
> Hi all
>
> I have redhat clustering running on a 3 VMware vm's 2 nodes and 1
> management server I can join
Stefan,
>
> On 16 May 2014, at 11:11 pm, Senftleben, Stefan (itsc)
> wrote:
>
>> Hello,
>>
>> I hope that someone can help me… I have a two node pacemaker
>> cluster, with to corosync rings. Ubuntu 10.04, 64 bit. Pacemaker
>> 1.0.8+hg15494-2ubuntu2, corosync 1.2.0-0ubuntu1.
>
> It _could_ be a
b. stack HA is re-started on all nodes
>
> 3/ if in case 2/ it switches back on first ring, let's say we re-start
> HA stack on only one node of the HA cluster, does it also switch back to
> first ring , or does it remain on the ring
> currently used by the other(s) node(
Alain,
passive mode is much better tested. Another big plus of passive is, that
if one network becomes faulty, passive makes progress (one packet is
send thru active device, another via faulty - this is not delivered but
resend via active device, ...). Active RRP waits until enough failures
and the
or you:
1.2.3 - Just don't use it. If you are using it, you may experience
various problems. If you really need to use it, please consider bonding.
1.4.1 - Only passive mode is fully supported.
Honza
> Alain
>
> Le 19/11/2013 09:54, Jan Friesse a écrit :
>> Alain,
>>
>&g
to not use RRP and consider bonding.
Regards,
Honza
> Thanks a lot for your help.
> Alain Moullé
>
>
> Le 18/11/2013 15:50, Jan Friesse a écrit :
>> Moullé Alain napsal(a):
>>> Hi,
>>>
>>> with corosync.1.2.3-36 (with Pacemaker) on a 4 nodes HA clu
Moullé Alain napsal(a):
> Hi,
>
> with corosync.1.2.3-36 (with Pacemaker) on a 4 nodes HA cluster, we got
1.2.3-36 is problem. This was last release WITHOUT official support for RRP.
> a strange and random problem :
>
> For some reason that we can't identify in the syslog, one node (let's
> say
__do_global_ctors_aux ()
> #4 0x00403c0b in _init ()
> #5 0x7fffe7a8 in ?? ()
> #6 0x00410d55 in __libc_csu_init ()
> #7 0x76dfcc70 in __libc_start_main () from /lib64/libc.so.6
> #8 0x00404539 in _start ()
>
>
> Thanks,
> Aarti Sawant
> N
Aarti,
can you please try to install debug informations and include backtrace
from coredump?
Regards,
Honza
Aarti Sawant napsal(a):
> hello,
>
> I am trying to setup HA on centos6.4 lxc container.
> i have install pacemaker and corosync on this container.
> my settings of /etc/corosync/corosyn
g them. Actually, if you are compiling from
source and don't pass configure parameter --enable-rdma, corosync is
built without RDMA code and without dependencies to IB libraries.
Honza
> Thanks again
> Alain
>
>
> Le 29/05/2013 15:53, Jan Friesse a écrit :
>> Alain,
>
Alain,
Moullé Alain napsal(a):
> Hi Jan,
>
> Just for my information, I would like only to know the dependancies
> between corosync and both IB libs libibverbs-devel librdmacm-devel
:
Yes
> in which configuration corosync needs functions from these both IB libs ?
> only if there is a heartb
ast thing ;-) :
> which is the last official release available of the corosync rpm ?
> and where can I get this last release ?
> (can't find it on clusterlabs site)
>
> Thanks
> Alain
>
> Le 17/05/2013 07:55, Jan Friesse a écrit :
>> Moullé Alain napsal(a):
>>
han two nodes, or may it happen also in two nodes HA
> clusters ?
>
Also with two nodes.
> 3/ Which first release of corosync includes the patch given in 854216 ?
>
Upstream 1.4.5 (or 2.3.0). If you are asking about RHEL, it's not in
6.4, but hopefully will be in 6.5.
> Thanks a
there was only one heartbeat network.
>>>So my question, does the risk of rrp mode not working correctly (if
>>> mcastaddr and mcastport are the same
>>>for both rings), depends on IF used ?
>> Not from corosync side, but routing of packets may do a really ba
vailable setup, use two nics per machine connected to two independent
switches.
Honza
> And is the risk null when using two std eth IF ?
>
> Thanks for all information.
> Alain
>
> Le 25/04/2013 17:33, Jan Friesse a écrit :
>> Moullé Alain napsal(a):
>>> Hi,
>>>
Moullé Alain napsal(a):
> Hi,
>
> "you can choose" ... meaning that it is not mandatory ? and my
> configuration is correct anyway ?
No, your configuration is not correct. "You can choose..." means binary
OR. So (table)
same_mcast_addr | same_port +- 1 | works
---
Moullé Alain napsal(a):
> Hi,
>
> corosync-1.4.1-7
>
> with two rings in corosync.conf , and rrp mode active, is it recommended
> to have two distinct mcastaddr ?
You can choose to have ether two distinct mcastaddr(eses) or distinct
ports (don't use port +- 1).
> (and if so, where can I find th
side, there is now way how to help.
Regards,
Honza
eXeC001er napsal(a):
> 2013/3/28 Jan Friesse
>
>> Hi,
>> corosync functions never returns ERRNO errors. They are returning errors
>> like CS_ERR_*. What is return value you've got from cmap_initialize?
>&
Hi,
corosync functions never returns ERRNO errors. They are returning errors
like CS_ERR_*. What is return value you've got from cmap_initialize?
Regards,
Honza
eXeC001er napsal(a):
> Hello.
>
> I tried to create an application that uses corosync via its libraries, but
> it seems something wro
Great news!
Regards,
Honza
Charles Williams napsal(a):
> Hey all,
>
> I recently got a chance to finally build Debian packages for the
> 0.5.2 version of ClusterLabs Hawk GUI. These are Squeeze packages
> ATM (Wheezy to come next week dependent upon testing of the
> current packages) and I am
Rosser,
actually, problem is much simpler to solve then you would expect.
If you look closely to your config there is:
> member {
> mamberaddr: 10.198.156.47
^
> }
> member {
> memberaddr: 10.198.156.48
> }
you have mAmberaddr instead of mEmberad
Reviewed-by: Jan Friesse
Russell Bryant napsal(a):
> This patch resolves a deadlock between the serialize lock (in
> exec/main.c) and the timer lock (in exec/timer.c). I observed this
> deadlock happening fairly quickly on a cluster using the EVT service
> from OpenAIS. (O
RRP threshold can be lower value then 5.
Signed-off-by: Jan Friesse
---
exec/totemconfig.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/exec/totemconfig.c b/exec/totemconfig.c
index f767f69..a475bb3 100644
--- a/exec/totemconfig.c
+++ b/exec/totemconfig.c
@@ -82,7
Reviewed-by: Jan Friesse
Steven Dake napsal(a):
> a memb_join operation that occurs during flushing can result in an
> entry into the GATHER state from the RECOVERY state. This results in the
> regular sort queue being used instead of the recovery sort queue, resulting
>
Included is API proposal for replacement of objdb/confdb API. It should
keep all good things there (triggers, ...), remove hard to use bits
(like whole object idea) and improve existing things (like typing)
Even I wrote it before, also configuration file will need change.
Proposed change is
ht_k
Backport of corosync patch
a memb_join operation that occurs during flushing can result in an
entry into the GATHER state from the RECOVERY state. This results in the
regular sort queue being used instead of the recovery sort queue, resulting
in segfault.
Signed-off-by: Jan Friesse
Backport of corosync patch
a memb_join operation that occurs during flushing can result in an
entry into the GATHER state from the RECOVERY state. This results in the
regular sort queue being used instead of the recovery sort queue, resulting
in segfault.
Signed-off-by: Jan Friesse
. Variable is
unused in active mode.
Signed-off-by: Jan Friesse
---
exec/totemconfig.c | 11 +++
exec/totemrrp.c|6 --
exec/totemsrp.c|3 +++
include/corosync/totem/totem.h |2 ++
man/corosync.conf.5|8
If all interfaces were faulty, passive_mcast_flush_send and related
functions ended in endless loop. This is now handled and if there is no
live interface, message is dropped.
Signed-off-by: Jan Friesse
---
exec/totemrrp.c | 29 -
1 files changed, 20 insertions
Fabio M. Di Nitto napsal(a):
> On 08/25/2011 06:31 AM, Angus Salkeld wrote:
>> On Thu, Aug 25, 2011 at 05:16:20AM +0200, Fabio M. Di Nitto wrote:
>>> On 08/25/2011 04:56 AM, Angus Salkeld wrote:
>>>
Possible Solutions
==
1] API
We really just want to get/set
Two interfaces (in RRP mode) shouldn't have equal unicast or
multicast addresses.
Signed-off-by: Jan Friesse
---
exec/totemconfig.c | 24 +++-
1 files changed, 23 insertions(+), 1 deletions(-)
diff --git a/exec/totemconfig.c b/exec/totemconfig.c
index 80ca182..86
last messages should be delivered after a transitional
configuration to maintain SAFE agreement. We don't have support for
SAFE atm, so it is probably safe just to throw these messages away. Without
this change, the new message queue on a new configuraton change is out of sync.
Signed-off-by
EINTR is now correctly handled in msg_send function by retry of
send operation.
Signed-off-by: Jan Friesse
---
branches/whitetank/exec/ipc.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/branches/whitetank/exec/ipc.c b/branches/whitetank/exec/ipc.c
index 17520c7
function ipc_disconnect
is converted to call of new function by using locked parameter
set to 0.
Signed-off-by: Jan Friesse
---
branches/whitetank/exec/ipc.c | 20 +---
1 files changed, 17 insertions(+), 3 deletions(-)
diff --git a/branches/whitetank/exec/ipc.c b/branches/whitetank
Angus Salkeld napsal(a):
> On Wed, Aug 17, 2011 at 01:19:53PM +1200, Tim Beale wrote:
>> Hi,
>>
>> I'm resending this patch in a separate thread because I think this part of
>> the
>> cluster formation problems I'm seeing has been overlooked. The patch attached
>> is one way of addressing the prob
Michael,
I've added few more distributions (like Gentoo, ...)
Regards,
Honza
Fletcher, Michael wrote:
> Sirs:
>
> Please note that Ubuntu should be added to the list of Linux distributions
> providing the Corosync packages. Corosync is in the Ubuntu repositories. As
> such, it's probably i
corosync_timer_handle_t is know conditionally defined to prevent double
definition causing compile fault on RHEL 6 systems.
Signed-off-by: Jan Friesse
---
exec/timer.h |3 +++
include/corosync/engine/coroapi.h |4
2 files changed, 7 insertions(+), 0 deletions
Current objdb/confdb is really hard to use, because of all iterationing,
... It would be nice to replace it by hash table and thus for simple get
item or set item, no iteration is needed. But iteration functionality
should still somehow be there to allow user select for example all
totem.* item
that,
Reviewed-by: Jan Friesse
(on all 3 of them)
Florian Haas napsal(a):
> It's nice to say people should read the man page. It's also naive to
> assume that they always do. Include comments in the example config
> file itself.
>
> Signed-off-by: Florian Haas
totem_mcast function can return -1 if corosync is overloaded. Sadly in
many calls of this functions was error code ether not handled at all, or
handled by assert.
Commit changes behaviour to ether return CS_ERR_TRY_AGAIN or put error
code to later layers to handle it.
Signed-off-by: Jan Friesse
totem_mcast function can return -1 if corosync is overloaded. Sadly in
many calls of this functions was error code ether not handled at all, or
handled by assert.
Commit changes behaviour to ether return CS_ERR_TRY_AGAIN or put error
code to later layers to handle it.
Signed-off-by: Jan Friesse
totem_mcast function can return -1 if corosync is overloaded. Sadly
in many calls of this functions was error code ether not handled at
all, or handled by assert.
Commit changes behaviour to ether return CS_ERR_TRY_AGAIN or put
error code to later layers to handle it.
Signed-off-by: Jan Friesse
cla...@gmail.com>
Signed-off-by: Jan Friesse
---
lib/coroipcc.c | 67 +--
1 files changed, 40 insertions(+), 27 deletions(-)
diff --git a/lib/coroipcc.c b/lib/coroipcc.c
index 14860e2..54d9aa7 100644
--- a/lib/coroipcc.c
+++ b/lib/coroipcc.c
@@
This reverts commit 2167
Reversion is needed to remove overflow of receive buffers and dropping
messages.
Signed-off-by: Jan Friesse
---
branches/whitetank/exec/totemnet.c | 45 -
branches/whitetank/exec/totemnet.h |2 +
branches/whitetank/exec/totemrrp.c | 65
Probably this is because one node uses secauth (one with messages
"invalid digest ...") and second node doesn't (one with Type of
received message is wrong).
Proskurin Kirill wrote:
> Hello all.
>
> Just catch fully reproducible crash of corosync 1.4.1
>
> OS: Centos 5.3 i386
> RPMS:
> Corosyn
I am pleased to announce the latest maintenance release of Corosync
1.3.3 and 1.4.1 available immediately from our website at
http://www.corosync.org.
This release fixes mainly problem with Retransmit list errors even if
network is perfectly OK. Bug appears only on high cpu load/weak cpu and
lo
Signed-off-by: Jan Friesse
---
exec/main.c | 24 +++-
1 files changed, 15 insertions(+), 9 deletions(-)
diff --git a/exec/main.c b/exec/main.c
index be9e118..1c4fb37 100644
--- a/exec/main.c
+++ b/exec/main.c
@@ -184,6 +184,8 @@ static int32_t corosync_not_enough_fds_left
Steven Dake wrote:
> On 07/19/2011 08:01 AM, Jan Friesse wrote:
>> Signed-off-by: Jan Friesse
>> ---
>> corosync.spec.in |2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/corosync.spec.in b/corosync.spec.in
&g
Signed-off-by: Jan Friesse
---
corosync.spec.in |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/corosync.spec.in b/corosync.spec.in
index e1dcf19..37e53ed 100644
--- a/corosync.spec.in
+++ b/corosync.spec.in
@@ -18,8 +18,8 @@ Version: @version@
Release:
1%{?numcomm
Signed-off-by: Jan Friesse
---
corosync.spec.in |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/corosync.spec.in b/corosync.spec.in
index 37e53ed..823ad3d 100644
--- a/corosync.spec.in
+++ b/corosync.spec.in
@@ -138,7 +138,7 @@ fi
%{_sysconfdir}/dbus-1/system.d
Signed-off-by: Jan Friesse
---
corosync.spec.in |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/corosync.spec.in b/corosync.spec.in
index 823ad3d..74ab851 100644
--- a/corosync.spec.in
+++ b/corosync.spec.in
@@ -92,6 +92,11 @@ rm -rf %{buildroot}
make install
Proskurin Kirill napsal(a):
> On 07/18/2011 06:37 PM, Jan Friesse wrote:
>> Corosync 1.4.0 is available for immediate download from our website.
>
> Great news!
>
> Should we consider it stable, ready for production use and easy to
> update? :-)
>
Yep, definitivel
Corosync 1.4.0 is available for immediate download from our website.
This version brings many enhancements to the software but most visible
change is redundant ring auto recovery functionality.
Please retrieve the latest sources from our website:
http://www.corosync.org
Regards
Honza
__
Signed-off-by: Jan Friesse
---
exec/totemrrp.c | 24 +++-
1 files changed, 23 insertions(+), 1 deletions(-)
diff --git a/exec/totemrrp.c b/exec/totemrrp.c
index 6fb5772..eb9b788 100644
--- a/exec/totemrrp.c
+++ b/exec/totemrrp.c
@@ -468,6 +468,22 @@ static void
code.
Signed-off-by: Jan Friesse
---
exec/totemconfig.c |2 +-
man/corosync.conf.5 |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/exec/totemconfig.c b/exec/totemconfig.c
index 5135672..80ca182 100644
--- a/exec/totemconfig.c
+++ b/exec/totemconfig.c
@@ -73,7
Reviewed-by: Jan Friesse
Steven Dake napsal(a):
> Fixes a problem where there are gaps in the recovery queue. Example my_aru =
> 5,
> but there are messages at 7,8. 8 = my_high_seq_received which results
> in data slots taken up in new message queue. What should really happen
>
code.
Signed-off-by: Jan Friesse
---
exec/totemconfig.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/exec/totemconfig.c b/exec/totemconfig.c
index 5135672..80ca182 100644
--- a/exec/totemconfig.c
+++ b/exec/totemconfig.c
@@ -73,7 +73,7 @@
#define JOIN_TI
Signed-off-by: Jan Friesse
---
exec/totemrrp.c | 31 ++-
1 files changed, 30 insertions(+), 1 deletions(-)
diff --git a/exec/totemrrp.c b/exec/totemrrp.c
index 6fb5772..0445be2 100644
--- a/exec/totemrrp.c
+++ b/exec/totemrrp.c
@@ -468,6 +468,22 @@ static void
Signed-off-by: Jan Friesse
---
exec/totemrrp.c | 175 +++
1 files changed, 112 insertions(+), 63 deletions(-)
diff --git a/exec/totemrrp.c b/exec/totemrrp.c
index 0445be2..6bfacd9 100644
--- a/exec/totemrrp.c
+++ b/exec/totemrrp.c
@@ -335,6
Hi Xin,
UDPU means UDP unicast. In other words, if you have 6 nodes, each
message must be send 5 times (to each other node), instead of 1 time as
in mcast/bcast (switch/hub will take care to deliver message to all
other 5 nodes). So UDPU is definitively less effective then mcast/bcast.
I'm not sur
Proskurin Kirill napsal(a):
> On 07/07/2011 07:13 PM, Jan Friesse wrote:
>> I am pleased to announce the latest maintenance release of Corosync
>> 1.3.2 available immediately from our website at http://www.corosync.org.
>
> Could you tell - if 1.3.x is stable and recommend
Reviewed-by: Jan Friesse
Steven Dake napsal(a):
> Signed-off-by: Steven Dake
> ---
> exec/apidef.c |4 +++-
> include/corosync/engine/coroapi.h | 17 -
> services/confdb.c |4 ++--
> 3 files changed, 21 insertio
c16981e02d3612f521bfbdb)
2011-07-04 Jan Friesse
coroipcc: Test _SC_PAGESIZE result
Reviewed-by: Steven Dake
(cherry picked from commit
94d934e0e0fa55027a974eb709a488802ee6134e)
Remove spinlocks
Spinlocks are now removed, because even spinloc
Signed-off-by: Jan Friesse
---
exec/totemiba.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/exec/totemiba.c b/exec/totemiba.c
index ec4ccfc..0b2d2ca 100644
--- a/exec/totemiba.c
+++ b/exec/totemiba.c
@@ -271,6 +271,7 @@ static inline struct send_buf
Signed-off-by: Jan Friesse
---
exec/totemudpu.c|1 -
test/testcpg.c |3 ++-
tools/corosync-objctl.c |1 -
3 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/exec/totemudpu.c b/exec/totemudpu.c
index 9281057..a036b38 100644
--- a/exec/totemudpu.c
+++ b
Andrew,
you are right in the fact that coroipcs uses functions defined in
utils.c which are not linked to coroipcs itself. On the other hand,
libcoroipcs is always used only with corosync and corosync executable
contains this symbols, so there shouldn't be any problem.
What compiler + OS + conf
Tested-by: Jan Friesse
I've tested 3 vms with both active and passive rr, iptables on each VM
and also on bridge (each vm had 2 nics, vm nic was connected to one bridge).
One thing which I found little "confusing" was message "... FAULTY -
administrative intervention re
Hi Juaju,
patch looks good so
Reviewed-by: Jan Friesse
but I would like to have also Steve reviewed-by.
So Steve, can you please review patch so ether I or you can commit that?
Regards,
Honza
Jiaju Zhang napsal(a):
> This is the V4 version of the implementation of automatic redund
Signed-off-by: Jan Friesse
---
exec/logsys.c | 17 ++---
1 files changed, 14 insertions(+), 3 deletions(-)
diff --git a/exec/logsys.c b/exec/logsys.c
index cef745c..1ba2c11 100644
--- a/exec/logsys.c
+++ b/exec/logsys.c
@@ -1032,6 +1032,7 @@ int _logsys_wthread_create (void)
int
Coverity doesn't complain even with testing to -1, so better patch included.
Regards,
Honza
Jan Friesse wrote:
> Steven Dake wrote:
>> On 06/01/2011 09:14 AM, Jan Friesse wrote:
>>> Signed-off-by: Jan Friesse
>
>> Prefer a direct comparison for the erro
Jiaju Zhang wrote:
> On Mon, Jun 27, 2011 at 03:33:00PM +0200, Jan Friesse wrote:
>> Jiaju,
>> thanks for your work. Patch looks pretty good and seems to resolve all
>> issues Steve had. Few notes:
>> - There are trailing whitespaces, please remove them (git diff with
Jiaju,
thanks for your work. Patch looks pretty good and seems to resolve all
issues Steve had. Few notes:
- There are trailing whitespaces, please remove them (git diff with
color output helps you)
- at least encapsulated field in message_header is unitialized. Please
initialize it (even by mem
gt; to turn that functionality on.
>
> Regards
> -steve
>
>
> On 06/21/2011 03:55 AM, Jan Friesse wrote:
>> Spinlocks are now by default disabled, because even spinlock can improve
>> speed is some special cases, in most cases it makes corosync CPU usage
>> much
Spinlocks are now by default disabled, because even spinlock can improve
speed is some special cases, in most cases it makes corosync CPU usage
much more intensive and less responsive then if only mutexes are used.
Signed-off-by: Jan Friesse
---
configure.ac | 13 ++---
1 files
Signed-off-by: Jan Friesse
---
configure.ac |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/configure.ac b/configure.ac
index fee8629..41dfeaf 100644
--- a/configure.ac
+++ b/configure.ac
@@ -264,7 +264,7 @@ AM_CONDITIONAL(BUILD_MONITORING, test x$enable_monitoring
Signed-off-by: Jan Friesse
---
configure.ac | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/configure.ac b/configure.ac
index a01c90c..76afc9b 100644
--- a/configure.ac
+++ b/configure.ac
@@ -229,7 +229,7 @@ AC_ARG_ENABLE([debug],
[ default=&qu
Spinlocks are now by default disabled, because even spinlock can improve
speed is some special cases, in most cases it makes corosync CPU usage
much more intensive and less responsive then if only mutexes are used.
Signed-off-by: Jan Friesse
---
configure.ac | 13 ++---
1 files
Steven Dake napsal(a):
Looks really good. one point below about error handling when writing to
full pipe.
Regards
-steve
On 06/16/2011 08:18 AM, Jan Friesse wrote:
Following situation could happen:
- one thread is waiting for finish write operation (line 853), objdb is
locked
- flush
Test is disabled by default because it depends on SMP and about 2GB RAM.
It's also testing race, so test is unreliable.
Signed-off-by: Jan Friesse
---
cts/agents/Makefile.am |4 +-
cts/agents/confdb-dispatch-deadlock.sh | 57
appears.
Commit solves this by:
- setting pipe to non-blocking mode
- pipe is used only as trigger for coropoll
- dispatch messages are stored in list
- main thread is processing messages from list
Signed-off-by: Jan Friesse
---
services/confdb.c | 143
ault
Now object_find_create creates array of matching object handlers and
object_find_next uses that array together with check for name. This
prevents situation where between steps 2 and 3 new object is created
with different name but sadly with same handler.
Signed-off-by: Jan Friesse
---
exec/objdb
According specification recvmsg can return 0, which means that
connection is closed. We had this check, but limited only for systems
other then Linux. recvmsg can return 0 even on Linux, so check is now
applied on all systems.
Signed-off-by: Jan Friesse
---
lib/coroipcc.c |6 ++
1 files
in confdb_object_iter result of object_find_create is now properly
checked. object_find_create can return -1 if object doesn't exists.
Without this check, incorrect handle (memory garbage) was directly
passed to object_find_next.
Signed-off-by: Jan Friesse
---
services/confdb.c |
Signed-off-by: Jan Friesse
---
exec/coroipcs.c |7 +++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a/exec/coroipcs.c b/exec/coroipcs.c
index e06acb5..ff78a92 100644
--- a/exec/coroipcs.c
+++ b/exec/coroipcs.c
@@ -1564,6 +1564,7 @@ static void coroipcs_init_conn_stats
Better patch included
Honza
Steven Dake napsal(a):
On 06/02/2011 12:19 AM, Jan Friesse wrote:
Steven Dake wrote:
On 06/01/2011 09:14 AM, Jan Friesse wrote:
Signed-off-by: Jan Friesse
---
exec/mainconfig.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/exec
Reviewed-By: Jan Friesse
Steve,
also note that same problem may be with circular_memory_map.
Steven Dake napsal(a):
> Previously in the non-circular memory map call, a private mapping was created
> and then this private mapping was then remapped as a shared mapping. This
> doesn&
Steven Dake wrote:
> On 06/01/2011 09:14 AM, Jan Friesse wrote:
>> Signed-off-by: Jan Friesse
>> ---
>> test/testcpgzc.c |2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/test/testcpgzc.c b/test/testcpgzc.c
>>
Steven Dake wrote:
> On 06/01/2011 09:14 AM, Jan Friesse wrote:
>> Signed-off-by: Jan Friesse
>> ---
>> exec/mainconfig.c |4 +++-
>> 1 files changed, 3 insertions(+), 1 deletions(-)
>>
>> diff --git a/exec/mainconfig.c b/exec/mainconfig.c
>>
Steven Dake wrote:
> On 06/01/2011 09:14 AM, Jan Friesse wrote:
>> Signed-off-by: Jan Friesse
>> ---
>> tools/corosync-notifyd.c |7 ++-
>> 1 files changed, 6 insertions(+), 1 deletions(-)
>>
>> diff --git a/tools/corosync-notifyd.c b/tools/coros
Steven Dake wrote:
> On 06/01/2011 09:14 AM, Jan Friesse wrote:
>> Signed-off-by: Jan Friesse
>> ---
>> lib/coroipcc.c |8
>> 1 files changed, 8 insertions(+), 0 deletions(-)
>>
>> diff --git a/lib/coroipcc.c b/lib/coroipcc.c
>> index 1a
1 - 100 of 445 matches
Mail list logo
