> If you are talking about the problem that you reported in February, I
> thought now it was a different issue :)
i only wanted to write a review ;) but this is the same issue :(
> If you mean that at the beginning everything is normal, but when you put
> load they start to suspect each other, this is a known problem that we
> need to fix asap. It could be related with flow control between the
> elements of the group. I just did not had the time for it :(
Yes, and the weird thing is, if i activate the debug-mode everything
works fine even there is heavy load on the controller. So i fixed this
problem for the moment :P
for your overview i've tihs log4j config for hedera
# Hedera channels test #
log4j.logger.test.org.continuent.hedera.channel=DEBUG, Console, Filetrace
log4j.additivity.test.org.continuent.hedera.channel=true
# Hedera adapters #
log4j.logger.org.continuent.hedera.adapters=INFO, Console, Filetrace
log4j.additivity.org.continuent.hedera.adapters=true
# Hedera factories #
log4j.logger.org.continuent.hedera.factory=INFO, Console, Filetrace
log4j.additivity.org.continuent.hedera.factory=true
# Hedera channels #
log4j.logger.org.continuent.hedera.channel=DEBUG, Console, Filetrace
log4j.additivity.org.continuent.hedera.channel=true
# Hedera Group Membership Service #
log4j.logger.org.continuent.hedera.gms=DEBUG, Console, Filetrace
log4j.additivity.org.continuent.hedera.gms=true
Greetings
Christopher
Nuno Carvalho schrieb:
Hi Christopher,
On Apr 24, 2007, at 1:06 , Christopher Hartung wrote:
Hi Nuno,
yes, they see each other and sends message to the group (both are
still in group). But they don't see each other...i told you: weird ;)
As configuration we use this:
<channel name="UDP multicast SEQ Channel"
template="udp_multicast_sequencer" initialized="yes">
<memorymanagement size="40000000" up_threshold="15000000"
down_threshold="7000000" />
<chsession name="hederalayer">
<parameter name="gossip_address">224.1.1.7</parameter>
<parameter name="multicast">224.2.2.7:11000</parameter>
</chsession>
<chsession name="UdpSimpleSession">
<parameter name="max_udp_message_size">32000</parameter>
</chsession>
</channel>
your statement sounded like "you've never heard about it" :)
If you are talking about the problem that you reported in February, I
thought now it was a different issue :)
If you mean that at the beginning everything is normal, but when you put
load they start to suspect each other, this is a known problem that we
need to fix asap. It could be related with flow control between the
elements of the group. I just did not had the time for it :(
But, by your messages, I did not realize that the
problem disappears when you put log4j in DEBUG mode. But that should
also be related with the same problem. Maybe the process gets slower
with all the messages on the screen and Appia doesn't need the flow
control any more because there are less messages to dispatch... it
happened to me before.
Cheers,
--
Nuno Carvalho
University of Lisbon, Portugal
http://dialnp.di.fc.ul.pt
------------------------------------------------------------------------
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
