Hi , I have a cman-based cluster that uses pcmk-fencing. we have
configured an ipmilan fencing device and an apc fencing device with
stonith.
I set a fencing order like this:
\
target="gw2"/> \
target="gw2"/> \
This all works as intended, i.e. the apc is used as first device and
shuts d
Am 20.09.2013 17:07, schrieb Dejan Muhamedagic:
Hi,
On Wed, Sep 18, 2013 at 11:22:46AM +0200, Jakob Curdes wrote:
Hi, we would like to use the rather uncommon "rcd_serial" plugin in our
CentOS based cluster.
The CentOS stonith packages do not seem to have it included; I tried to
co
Am 19.09.2013 13:08, schrieb Andrew Beekhof:
Also it seems that at least some parameters in corosync.conf are read,
but I cannot recall which ones.
No. When cman is in use, corosync.conf is not touched.
cman make be setting some defaults to the same values though
"Good that we have talked about
Am 19.09.2013 11:49, schrieb David Lang:
On Thu, 19 Sep 2013, Jakob Curdes wrote:
That's the direction we started, but apparently the centos
pacemaker/corosync packages don't look at the corosync.conf file, they
expect to extract everything out of cluster.conf.
Ok but you have note
Am 19.09.2013 01:21, schrieb David Lang:
token_retransmits_before_loss_const="10" join="60" consensus="4800"
rrp_mode="none" transport="udpu">
ttl="1"
I have the following configuration which works:
cluster.conf (excerpt):
default-action-timeout="60s" />
c
Am 18.09.2013 11:43, schrieb Andreas Mock:
Hi Jakob,
besides of you compiling problem. Be sure that you can use
this stonith agent with CentOS. The stonith subsystem is
different. The stonith agents you find with /usr/sbin/fence*
conform to two API.
Probably anybody here can answer which stonit
ing libltdl includes etc.; when I
enable "--enable-bundled-ltdl" during configure, make tells me that
there is no rule to make all in libltdl.
Any idea out there if there is a package I can use or how to get this to
compile?
Regards, Jakob Curdes
__
addresses (i.e. the ones managed by the cluster). Anyway I
wonder whiy even with -d64 it doesn't tell me anything about what it is
doing. I think the timespan of an hour is just because we have lots of
VLAN interfaces the he wants to get a DNS name for
Regards,
Jakob C
Hi,
we have a simple 2-node cluster running CMAN and pacemaker under CentOS 6.
The problem is that upon startup the machines (even if "alone", i.e.
second machine is off), will give a cman timeout on startup saying
"Timed-out waiting for cluster".
*If I start the services manually an hour later
Am 15.08.2013 13:25, schrieb Fredrik Hudner:
So if you have any other suggestions how I can come out of this situation,
it would be much appreciated
Are you sure the machines see each other? "diskless" for the remote side
sounds not too convinving...
JC
___
nated with exit code 17
I had to use
"drbdadm --overwrite-data-of-peer primary"
on one of the systems in a similar case.
HTH, Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-
Am 03.07.2013 12:03, schrieb Florian Crouzat:
Le 02/07/2013 00:53, Chris Feist a écrit :
pcs resource create myping ocf:pacemaker:ping
host_list="www.microsoft.com" timeout=5 op monitor interval=10
You want to put the host_list and timeout before 'op' because they're
options for the ocf:pacemak
and make sure that the OCF script invokes proftpd
in the same way and with the same PID file locations as the LSB init
script in /etc/init.d .
Hopel this helps,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
Am 01.07.2013 16:17, schrieb David Vossel:
- Original Message -
From: "Jakob Curdes"
To: "General Linux-HA mailing list"
Sent: Sunday, June 30, 2013 6:04:58 AM
Subject: [Linux-HA] PCS and ping resources?
Hello, I have configured a cluster on CentOS 6.x using PCS.
to do this with PCS. I could not
find any document describing this. Did I miss something?
Regards,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org
the IP is nicely assigned to trhe correct vlan
interface.
I tried to figure out where the error occurs, but I am running out of
time so for now I will live with the workaround.
Regards,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-h
cib_create failed (-76): Name not unique on network
It seems that in the old heartbeat v1 times it was possible to specify a
VLAN interface with something like
192.168.1.1/29/eth0.30
How can I do this nowadays?
Thank you for hints,
Jakob Curdes
>Later I came to turn it back on and found I was stuck! Because it was in auto
>mode and had been shut down correctly, it was stuck turned off
>- turning on the PSU had no effect, so I had to call out an engineer to go and
>press the power button.
>The moral of the story is to leave the BIOS set
group started the python GUI
while other developers declared this development as a bad thing which in
their view did not even exist.
HA could be in much broader use if we could agree to some sort of
"mainstream" instead of switching principles every other year or
introducing new admin
lifetime pointing at a reachable server. But
perhaps a remote datacenter solution is the cheaper way... note that to
reach high availability you need to test your setup thoroughly,
otherwise you will end up with bad things like "split-brain" or unusable
services besides y
you must arrange
for mail delivered to that server to be passed on to the primary or a
separate backend server. And you need to protect it exactly as good as
your primary against virus, spam, and DOS attacks.
Best regards,
Jakob Curdes
___
Linux-HA mailing
Am 16.09.2011 16:58, schrieb Dejan Muhamedagic:
> Hi,
>
> On Thu, Sep 15, 2011 at 04:08:58PM -0500, Dimitri Maziuk wrote:
>> On 09/15/2011 03:29 PM, Jakob Curdes wrote:
>>> Yes, it is confusing. In short: Heartbeat is more or less "out of
>>> service".
Am 15.09.2011 23:32, schrieb Charles Richard:
> Hi Jakob,
>
> When you say that Heartbeat is more or less "out of service", are you saying
> that this project is not supported or being worked on anymore? Did you read
> this somewhere?
No. I did not want to infer this. But most descriptions for ins
for the HA "heartbeat",
especially if you are DRBDing something. You must not risk a split-brain
scenario in this case.
- test, test, test. pull the plugs, overload one machine, misconfigure
your application ...
HTH,
Jakob Curdes
___
Linux-HA m
Am 9.3.2011 11:59, schrieb Yan Gao:
etc.
Hello Yan,
thank you for the specs. I will try my luck, might take a couple of days.
I will get back to the list as soon as I have something.
Best regards,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA
led cluster-glue from source, not sure if I need that to compile
the GUI.
I assume the cluster-glue components are part of the pacemaker package?
If somebody can tell me how the dependencies really are, I am willing to build
an RPM for the GUI and provide it.
Best regards and thanks,
Jakob Curdes
__
artbeat_cluster'
gmake[2]: *** [libhbmgmt_la-mgmt_lib.lo] Error 1
Still another component missing or what is up here? I still use the
source from the above link on the clusterlabs website, which is
equivalent to "Pacemaker-Python-GUI-999c06bac9da".
Best regards,
Jakob Curd
ling this, even providing an rpm when done, but I have
no clue why make complains abount missing files in the mgmtd directory.
Am I using the wrong source perhaps? I used the source from
http://hg.clusterlabs.org/pacemaker/pygui/archive/tip.tar.bz2
as indicated on the clusterlabs website.
]: *** [_pymgmt_la-pymgmt_wrap.lo] Error 1
As I am not the python guy I have no clue what happens. Can anybody point me to
a current RPM for fedora or CentOS or give me a hint what I am doing wrong when
compiling?
Thanks in advance,
Jakob Curdes
___
Lin
ff1ab132449ad5b236169403c6a23cf4168b"/>
> value="Heartbeat"/>
>
>
>
> uname="apauat1b.intranet.mydomain.com" type="normal"/>
> uname="apauat1a.intranet.mydomain.com" type="normal"/>
>
>
>
> type="IPa
need to read further down that page and use the settings in
"Failover IP Service in a Group"
What you probably actually want is to have IP and service running always
on the same node.
(plus- last step - on the node with best connectivity).
HTH,
Jakob Curdes
_
Perhaps you retry it by not removing the init script but just commenting
out the line where it starts apache.
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also:
Néstor schrieb:
> We have a host HA for our web server and we are planning on putting the back
> up server
> on another building.
>
> Is it true that you can not have an HA with host on separate subnets?
>
The problem is not the subnets, the main problem is that you need *at
least* two separate
ounds reasonable, but then what I did is taken from the curent
pacemaker 0.6 docs ... I will try this later today and see what happens.
Thank you,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
is
should give me a score of 2000 only with full connectivity.
But If I pull the network cable from the first machine it still wants to
run the complete resource group there - I cannot figure out why.
Full anonymized CIB attached below.
Where is my error???
Hoping for a hint,
Ja
Bruening, Stefan wrote:
This ist the Result of grep -r MAILCMD /usr/lib/ocf/resource.d/*
/usr/lib/ocf/resource.d/heartbeat/.ocf-binaries:: ${MAILCMD:=}
/usr/lib/ocf/resource.d/heartbeat/MailTo: $MAILCMD -s "$1" "$email"
Is anything wrong?
Well,
probably there is something wrong or we
d external IP ?
Hoping for a hint,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
Probably a MTU issue, not HA related. Compare the MTUs of the network
interfaces on both machines.
You're probably right. The ls -al makes full sized packets and
that would explain the problem.
Sometimes the symptom is that you can make ls in a small directory,
e.g. "/" but ls -l in
related. Compare the MTUs of the network
interfaces on both machines.
I know this behavior from VPN connections where the clam-mss-to-mtu rule
in iptables is not active.
Yours,
Jakob Curdes
___
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http
39 matches
Mail list logo
