Title: ALTRAN OUEST
I have the answer (ouff),
In fact my error was to restart the stopped controller too early. It
seems that sequoia waits for 2 minutes before discarding a lost
controller from its group.
If you start the controller before the end of this interval, the new
controller is seen but not added to the group (I suppose, but that's
not really clear).
For your case, you should add the "local_address" appia parameter to be
sure that the virtual IP is not used
Gérard
Gérard BUNEL a écrit :
Hi <NoNameInYourMail>,
Not sure it's really the same problem.
In our case, when the virtual IP address is switched from first node
the second one, all seems OK as there is still synchronization betwwend
the 2 controllers.
But if I stop 1 controller and restart it they can't see each other and
each VDB is considered as enable on both side.
What is strange is that when I type the command "show controllers" on
the first node (the one which has not been stopped), it displays:
MATISSEDB is hosted by 1 controller(s):
192.168.4.200:1090
But in the controller logs I can see:
09:49:11,773 INFO controller.virtualdatabase.MATISSEDB
Member(address=/192.168.4.200:27752, uid=192.168.4.200:27752) see
members:[Member(address=/192.168.4.200:27752, uid=192.168.4.200:27752),
Member(address=/192.168.4.201:27752, uid=192.168.4.201:27752)] and has
mapping:{Member(address=/192.168.4.200:27752,
uid=192.168.4.200:27752)=192.168.4.200:1090}
Where it seems to have member 192.168.4.201 still registered:
Member(address=/192.168.4.201:27752, uid=192.168.4.201:27752)
If I stop the two controllers and restart them, then it's OK
[EMAIL PROTECTED] a
écrit :
Hi Gerard,
I'm facing the same problem and it blocks my production release, see my
mail:
https://forge.continuent.org/pipermail/sequoia/2007-January/004457.html
May be you can vote in JIRA for resolving this issue. It is at all a
problem with detecting
the primary IP address of the underlying OS.
Regards,
)ngo
[EMAIL PROTECTED] schrieb am 12.02.2007 15:02:53:
Hello,
I'm currently facing some strange situation when testing
simultaneously Sequoia and Heartbeat.
In our case Heartbeat is used to manage Failover for Web applications.
So an Virtual IP adresse is transferred from the active host to the
passive host when failover is detected.
These hosts also supports Sequoia controllers.
When this happen, everything seem to work fine. both controllers
remain active and up in the group.
But now, if I stop a controller and start it again, each controller
can't see the other.
I need to stop both controllers and restart them to have a stable
situation (each controller contained in the group).
My appia configuration is the following:
<channel name="UDP p2p SEQ Channel" template="udp_p2p_sequencer"
initialized="yes">
<memorymanagement size="40000000" up_threshold="15000000"
down_threshold="7000000" />
<chsession name="hederalayer">
<parameter name="gossip_address">224.1.1.5:45566</parameter>
<parameter name="local_address">192.168.4.200:22000</parameter>
</chsession>
<chsession name="UdpSimpleSession">
<parameter name="max_udp_message_size">32000</parameter>
</chsession>
</channel>
Any advice appreciated,
Gerard
--
Gérard BUNEL
Chef de Projet
____________________________________________________________________
[Bild entfernt]
Technopôle Brest Iroise
Site du Vernis – CS 23866
29238 Brest Cedex 3
Tél : + 33 2 98 05 43 21
Fax : + 33 2 98 05 20 34
www.altran.com
+----------------------------------------------------------------------+
| Z1 SecureMail Gateway Info - http://www.zertificon.com |
+----------------------------------------------------------------------+
| - Die Nachricht war weder verschluesselt noch digital unterschrieben |
+----------------------------------------------------------------------+
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
--
Gérard
BUNEL
Chef
de Projet
____________________________________________________________________
Technopôle
Brest Iroise
Site du Vernis – CS 23866
29238
Brest Cedex 3
Tél : + 33 2 98 05 43 21
Fax : + 33 2 98 05 20 34
www.altran.com
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
--
ALTRAN OUEST
Gérard
BUNEL
Chef
de Projet
____________________________________________________________________
Technopôle
Brest Iroise
Site du Vernis – CS 23866
29238
Brest Cedex 3
Tél : + 33 2 98 05 43 21
Fax : + 33 2 98 05 20 34
www.altran.com
|
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
