Project

General

Profile

cluster problem(Feature not supported yet)

dawei liu
Added over 4 years ago

The single machine environment Ping returned to normal, as follows

05:05:40 下午 SENT (1151020120):

<iq id="0oUA4-12" to="pubtest.sgc.com" from="test@pubtest.sgc.com/hiss" type="get">
    <ping xmlns="urn:xmpp:ping">
        <username>test</username>
    </ping>
</iq>

05:05:40 下午 RCV (1151020120):

<iq type="result" id="0oUA4-12" from="pubtest.sgc.com" to="test@pubtest.sgc.com/hiss"/>

Ping returns the error cluster environment,as follows

下午 SENT (1151020120):

<iq id="5r12S-317" to="mam.sgc.com" from="7eddcb37-5325-37a5-a57a-88c5d613445b@mam.sgc.com/hiss" type="get">
    <ping xmlns="urn:xmpp:ping">
        <username>7eddcb37-5325-37a5-a57a-88c5d613445b</username>
    </ping>
</iq>

05:05:40 下午 RCV (1151020120):

<iq id="5r12S-317" from="mam.sgc.com" to="7eddcb37-5325-37a5-a57a-88c5d613445b@mam.sgc.com/hiss" type="error" xmlns="jabber:client">
    <ping xmlns="urn:xmpp:ping">
        <username>7eddcb37-5325-37a5-a57a-88c5d613445b</username>
    </ping>
    <error code="501" type="cancel">
        <feature-not-implemented xmlns="urn:ietf:params:xml:ns:xmpp-stanzas"/>
        <text xmlns="urn:ietf:params:xml:ns:xmpp-stanzas" xml:lang="en">Feature not supported yet.</text>
    </error>
</iq>

Cluster environment configuration file as follows

--cluster-mode = true
config-type = --gen-config-def
--debug = server.push
#domain name,according to the actual change.
--admins = admin@mam.sgc.com
--virt-hosts = mam.sgc.com
--user-db=oracle
#database url ,according to the actual change.
--user-db-uri=jdbc:nds://127.0.0.1:8102/v_ora11d01?appname=app_ora11d01
--user-db-username=nari
--user-db-passward=888888
--comp-name-1 = push
--comp-class-1 = tigase.server.push.PushComponent
c2s/max-inactivity-time[L]=45
#memcached ip and port ,according to the actual change.
--memcached-url=127.0.0.1:8103
# ios push url ,according to the actual change.
--iospush-url=http://127.0.0.1:8101/netplatform-web/service/user/iOSPushMessage.do
--schema=wx.
--pushthreadnum=4
--isoswitch=on
--pagenum=10000
--monitoring=jmx:9050

the vesion is 5.2.0 and the Table CLUSTER_NODES already contains the node configuration information

hostname password last_update    port  cpu_usage  mem_usage
push2   e128fef01.. 2014-11-01.. 5277  11         7
push1   ec5e53ec..  2014-11-01.. 5277  12         16

What is the problem in which? Cluster configuration is wrong?


Replies (1)

Added by Wojciech Kapcia TigaseTeam over 4 years ago

Clustering should not matter as ping requests are handled by local SessionManager of the node to which user is connected. Please enable following debug:

--debug = server,xmpp.impl

check resulting logs and look for the place where error originates.

    (1-1/1)