qpid-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From wei6rong <wei6r...@hotmail.com>
Subject RE: qpid-cluster broker recovery failed
Date Fri, 21 Sep 2012 01:33:30 GMT

hi, 
no, in my default --module-dir 
[root@localhost daemon]# pwd
/usr/local/lib/qpid/daemon
have following lib
[root@localhost daemon]# ls
acl.la  cluster.la  ha.la  msgstore.so     replicating_listener.la  replication_exchange.la
 ssl.la  watchdog.la  xml.la
acl.so  cluster.so  ha.so  msgstore.so_16  replicating_listener.so  replication_exchange.so
 ssl.so  watchdog.so  xml.so
[root@localhost daemon]#

and you can see, the fail message is "Unsupported transport type: tcp", not ssl or rdma, and
i have been using gdb to restart my cluster_two,  when run into Broker.cpp:1004,
boost::shared_ptr<ProtocolFactory> pf = getProtocolFactory(transport);
i print transport, it is tcp, and pf is 0x0, so NoSuchTransportException is throw. 

following is snippet,when NoSuchTransportException  is throw.
void Broker::connect(
    const std::string& host, const std::string& port, const std::string& transport,
    boost::function2<void, int, std::string> failed,
    sys::ConnectionCodec::Factory* f)
{
    boost::shared_ptr<ProtocolFactory> pf = getProtocolFactory(transport);
    if (pf) pf->connect(poller, host, port, f ? f : factory.get(), failed);
    else throw NoSuchTransportException(QPID_MSG("Unsupported transport type: " << transport));
}

thanks
rong


Date: Thu, 20 Sep 2012 12:06:48 -0700
From: ml-node+s2158936n7582439h58@n2.nabble.com
To: wei6rong@hotmail.com
Subject: Re: qpid-cluster broker recovery failed



	On 09/20/2012 10:02 AM, wei6rong wrote:

> Hi guys,

>

> I have the following issue:

> I have two node 172.26.184.35 and 172.26.184.45

> in 172.26.184.35

> cd /root/qpid-0.18/cpp/src/.libs

> ./lt-qpidd -d --auth 0 -p 5682 --cluster-name cluster_one --load-module

> ./cluster.so --data-dir=/tmp/5682

> ./lt-qpidd -d --auth 0 -p 5683 --cluster-name cluster_two --load-module

> ./cluster.so --data-dir=/tmp/5683

>

> in 172.26.184.45

> /root/qpid-0.18/cpp/src/.libs

> ./lt-qpidd -d --auth 0 -p 5682 --cluster-name cluster_two --load-module

> ./cluster.so --data-dir=/tmp/5682

> ./lt-qpidd -d --auth 0 -p 5683 --cluster-name cluster_one --load-module

> ./cluster.so --data-dir=/tmp/5683

>

> 35-->45

> 172.26.184.35

> /root/qpid-0.18/cpp/src/.libs

> qpid-config -b 172.26.184.35:5682 --durable add queue queue-test

> --limit-policy=flow-to-disk --flow-stop-size=0 --flow-resume-size=0

> --file-size=2 --file-count=4 --max-queue-size=10000

> qpid-config -b 172.26.184.45:5682 --durable add queue queue-test

> --limit-policy=flow-to-disk --flow-stop-size=0 --flow-resume-size=0

> --file-size=2 --file-count=4 --max-queue-size=10000

> qpid-config -b 172.26.184.45:5682 --durable add exchange direct ex-test

> qpid-config -b 172.26.184.45:5682 --durable bind ex-test queue-test

> queue-test

> qpid-route queue add 172.26.184.45:5682 172.26.184.35:5682 ex-test

> queue-test --durable

>

> after then, i using

> qpid-cluster -C 172.26.184.35:5683 -k

> kill cluster_two

>

> but when i restart cluster_two, they would throw NoSuchTransportException in

> Broker.cpp:1006

> follow is output log

> [root@localhost .libs]# ./lt-qpidd --auth 0 -p 5682 --cluster-name

> cluster_two --load-module ./cluster.so --data-dir=/tmp/5682

> 2012-09-20 16:44:37 [Unspecified] notice Journal "TplStore": Created

> 2012-09-20 16:44:37 [Unspecified] notice Store module initialized;

> store-dir=/tmp/5682

> 2012-09-20 16:44:37 [HA] notice Initializing CPG

> 2012-09-20 16:44:37 [HA] notice Cluster store state: clean

> cluster-id=2ee15447-470a-4ef8-bf96-46bd29551bf7

> shutdown-id=7d851924-0082-44ab-9f2f-e43bd21eaece

> 2012-09-20 16:44:37 [HA] notice cluster(172.26.184.45:4822 PRE_INIT)

> configuration change: 172.26.184.45:4822

> 2012-09-20 16:44:37 [HA] notice cluster(172.26.184.45:4822 PRE_INIT) Members

> joined: 172.26.184.45:4822

> 2012-09-20 16:44:37 [Unspecified] notice Journal "queue-test": Created

> 2012-09-20 16:44:37 [Unspecified] warning Journal "queue-test": Recovery

> found 4 files (different from --num-jfiles value of 8).

> 2012-09-20 16:44:37 [Unspecified] warning Journal "queue-test": Recovery

> found file size = 2 (different from --jfile-size-pgs value of 24).

> 2012-09-20 16:44:37 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

> 2012-09-20 16:44:39 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

> 2012-09-20 16:44:43 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

> 2012-09-20 16:44:51 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

> 2012-09-20 16:45:07 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

> 2012-09-20 16:45:39 [Broker] error Link connection to 172.26.184.35:5682

> failed: Unsupported transport type: tcp (qpid/broker/Broker.cpp:1006)

>

>  From what I understood, in

> qpid/broker/Broker.cpp:338

> would call Plugin::initializeAll(*this);

> but when my restart my cluster_two, this is link info in my rhm store,

> the recovery code is in qpid/broker/Broker.cpp:287

> store->recover(recoverer); which is before Plugin::initializeAll(*this);

>

> so when recover there would throw NoSuchTransportException because at this

> time some plugin is not initialized,

>

> Is it possible? I would be very appreciate of any advice you my offer.

The NoSuchTransportException would suggest that the SSL (or less likely 

the rdma) plugin was not loaded or not configured on restart but was 

somehow being requested...


Do you have SSL enabled federation links recorded in the store?



---------------------------------------------------------------------

To unsubscribe, e-mail: [hidden email]

For additional commands, e-mail: [hidden email]




	
	
	
	

	

	
	
		If you reply to this email, your message will be added to the discussion below:
		http://qpid.2158936.n2.nabble.com/qpid-cluster-broker-recovery-failed-tp7582404p7582439.html
	
	
		
		To unsubscribe from qpid-cluster broker recovery failed, click here.

		NAML
	 		 	   		  



--
View this message in context: http://qpid.2158936.n2.nabble.com/qpid-cluster-broker-recovery-failed-tp7582404p7582460.html
Sent from the Apache Qpid developers mailing list archive at Nabble.com.
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message