MQ and me

wikipedia.

Read some (v6) Fundamentals.

Best url

Hardware requirements of MQ Server on Windows

• Any IBM PC machine (or compatible), that is certified as Windows 2000, Windows XP, or Windows 2003 compatible.
• Any communications hardware supporting SNA LU 6.2, TCP/IP, NetBIOS, or SPX is required for communication with other machines .
• The following disk space (for a typical installation):
  • 360 MB for product code and data (not including the Information Center)
  • 20 MB minimum working space
  • 30 MB of temporary space for the installation process

Hardware requirements of MQ Explorer

• 512 MB RAM
• 1 GHz processor
• 1.5 GB for WebSphere Eclipse Platform code and data
• A suitable monitor for the operating system with a screen size of at least 1024x768

MQ for Windows v 6.0 Quick Beginnings, amqtac05.pdf

Software requirements of MQ Server on Windows

• windows opsys :
  • XP Professional with SP2
  • XP Professional x64
  • Windows Server 2008 R2 (Standard or Enterprise 64-bit Edition) : MQ V7.0.1 or later is required
• Eclipse
• MQ FP 6.0.2.1 (that installs Java 2 v 1.4.2)

Hardware requirements of MQ server on Solaris

MQ for Sun Solaris, Version 6.0 runs on all Sun SPARC and Sun UltraSPARC desktop and server systems, supported by the appropriate release of the Solaris operating environment.

Hardware
MQ for Linux, Version 6.0 (x86 platform) runs on any machine that supports the x86 machine architecture.
MQ for Linux, Version 6.0 (x86-64 platform) runs on any machine that supports the x86-64 (AMD64 or Intel© EM64T) machine architecture.
MQ for Linux, Version 6.0 (POWER platform) runs on any machine that supports the POWER machine architecture.
MQ for Linux, Version 6.0 (zSeries platform) runs on any machine that supports the zSeries machine architecture.
MQ for Linux, Version 6.0 (zSeries s390x platform) runs on any machine that supports the zSeries s390x machine architecture.

Operating System
MQ for Linux, Version 6.0 (x86 platform) has been tested with the following distributions:

• Red Hat Enterprise Linux V3.0 plus Update 2
• Red Hat Enterprise Linux V4.0
• SuSE Linux Enterprise Server (SLES) V8 plus Service Pack 3
  Note: The MQ Explorer graphical interface is not supported on this operating system.
• SuSE Linux Enterprise Server (SLES) V9

Top

1. a la maquina remota, definim una cua per rebre missatges, QL2
2. a la maquina local, definim una cua remota que apunti a una cua local del gestor remot, QR1
3. mitjançant RFHUTIL, posem missatges a QR1 i apareixen a QL2
4. inhibim la cua QL2 per posar-hi missatges - els missatges apareixen a la DLQ del gestor remot, QM2
5. inhibim la cua DLQ del gestor remot QM2 - el missatge desapareix !

El resum de les comandes es :

Si el msg es no persistent, es perd i al fitxer AMQERR01.LOG trobem :

Si el msg es persistent, no es perd doncs s'atura el canal a la banda de recepcio i al fitxer AMQERR01.LOG trobem :

Top

url, o chapter 17, "Control Commands", llibre MQ v 6.0 System Admin Guide, amqzag08.pdf, SC34-6584-01.

AIX specific commands.

Top

STRMQTRC / ENDMQTRC / DSPMQTRC - start / stop / display MQ trace.

strmqtrc [-t TraceType] [-x TraceType] [-l MaxSize] [-d MaxUserDataSize]

Trace files ubication. During the installation process, you can choose the drive on which trace files are to be located. The trace files are always placed in the directory \<mqmwork>\errors (mq v 5.3) or \<mqmwork>\trace (mq v 6), where <mqmwork> is the directory selected when MQ was installed to hold MQ data files. Trace-file names have the following format: AMQppppp.TRC

System Administration Guide SC34-6068-02, page 228 [252/567]

UNIX special feature :
Start trace for all MQ operations occurring on a system by issuing the following command:

To format all trace files in the current directory use the following command:

Windows :

MQ for Windows trace does not require any formatting. Browse *.TRC files in "c:\MQ\trace\" directory.

Filter RC's by :

Top

Dumping the contents of the log using the dmpmqlog command

Use the dmpmqlog command to dump the contents of the queue manager log. By default all active log records are dumped, that is, the command starts dumping from the head of the log (usually the start of the last completed checkpoint). The log can usually be dumped only when the queue manager is not running. Because the queue manager takes a checkpoint during shutdown, the active portion of the log usually contains a small number of log records. However, you can use the dmpmqlog command to dump more log records using one of the following options to change the start position of the dump:

• Start dumping from the base of the log. The base of the log is the first log record in the log file that contains the head of the log. The amount of additional data dumped in this case depends on where the head of the log is positioned in the log file. If it is near the start of the log file, only a small amount of additional data is dumped. If the head is near the end of the log file, significantly more data is dumped.
• Specify the start position of the dump as an individual log record. Each log record is identified by a unique log sequence number (LSN). In the case of circular logging, this starting log record cannot be before the base of the log; this restriction does not apply to linear logs. You might need to reinstate inactive log files before running the command. You must specify a valid LSN, taken from previous dmpmqlog output, as the start position.

Top

Administrar SERVICIOS de MQ ...

Compte amb els canvis (listener) amb la versió 6

Top

Compte : moving to API module

Sample / file name	Purpose
amqsaicq.c	Create a local queue using the MQ Administration Interface (MQAI)
amqsaiem.c	Event monitor using the MQ Admin Interface (MQAI)
amqsailq.c	Inquire the current depth of all local queues using MQAI
amqsaxe0.c	ApiExit which traces MQAPI calls
amqsbcg0.c	Browse messages, text and header (use amqsgbr for text only)
amqsblst.c	Bandwidth and connectivity tester
amqscnxc.c	Client connection information on MQCONNX
amqsecha.c	(triggered program) MQPUT1() a msg to ReplyToQueue (sent by amqsreq maybe) - see QM_resolution_sample or AMQSECHA_modificat o QNR.C
amqsgbr0.c	Browse messages, text only (use amqsbcg for text and header)
amqsget0.c	MQGET() sample - Get a message
amqsgrma.c	Get reference messages from a queue. Checks that the objects, identified in the messages, exist
amqsinqa.c	(triggered program) - MQINQ() sample - inquiry Queue Depth, Get Inhibited and Input Count.
amqsldpc.c	Look up the MQ host name, channel name and queue name in an LDAP directory + MQPUT
amqsmon0.c	Get an statistics or accounting message
amqsprma.c	Create a reference message
amqsptl0.c	Put messages to a list of queues
amqspuba.c	Publish messages to a topic {subscribed by amqssuba}
amqsput0.c	MQPUT() sample - Put a message
amqsqrma.c	Channel message exit program to process reference messages
amqsreq0.c	Put request messages to a queue, and show the reply message (sent by amqsecha, maybe)
amqsseta.c	(triggered program) - MQSET() sample
amqsspin.c	SSPI Channel Exit routines (Security)
amqsstop.c	Inquire on all the connections for a selected process and request them to be stopped
amqssuba.c	Subscribe and get messages from a topic using a managed destination queue {published by amqspuba}
amqstrg0.c	Trigger monitor - read an initiation queue, and start the program associated with trigger message
amqstxgx.c	Get transaction for Tuxedo
amqstxpx.c	Put transaction for Tuxedo
amqstxsx.c	Server for Tuxedo
amqsvfc0.c	Skeleton of a Data Conversion exit
amqswlm0.c	CLWL exit that chooses a destination QMgr
amqsxae0.c	Sample Encina program
amqsxrma.c	Channel message exit to process reference messages
amquregn.c	Dump MQ values from Registry (NT)
amqzscgn.c	CICS Global User Exit (GLUE) program for the CICS Task termination user exit (UE014015)
amqzscin.c	XA switch program for CICS XA Initialisation

Top

(1) dis qstatus(SYSTEM.CHANNEL.INITQ) type(handle) all => runmqchi
(2) dis qstatus(SYSTEM.ADMIN.COMMAND.QUEUE) type(handle) all => amqpcsea
(3) dis qstatus(SYSTEM.DEFAULT.INITIATION.QUEUE) type(handle) all =>

Queue config

Queue status

Use DEFINE QREMOTE to define a new local definition of a remote queue, a queue manager alias, or a reply-to queue alias, and to set its parameters.

Page 178, MQ v 6.0, "MQSC Command Reference", SC34-6597-00

The remote queue definition supplies the name of the destination queue, the destination queue manager, and optionally, a transmission queue.

"Using LDAP with WebSphere MQ", page 374, Application Programing Guide, MQ v 6.0, SC34-6595-01.

If an application opens the local definition of a remote queue, RQMNAME must not be blank or the name of the local queue manager. When the open occurs, if XMITQ is blank there must be a local queue of this name, which is to be used as the transmission queue.

Page 196, MQ v 6.0, "MQSC Command Reference", SC34-6597-00

When you use remote queue definitions to specify a transmission queue, your applications must not specify the location (that is, the destination queue manager) themselves. If they do, the queue manager will not make use of the remote queue definitions. Remote queue definitions make the location of queues and the transmission queue transparent to applications. Applications can put messages to a logical queue without knowing where the queue is located and you can alter the physical queue without having to change your applications.

Page 27/525196, MQ v 7, "Intercommunication", SC34-6931-00

A model queue is a template of a queue definition used only when you want to create a dynamic local queue. You can create a local queue dynamically from a MQ program, naming the model queue you wish to use as the template for the queue attributes. You may now, if you wish, change some attributes of the new queue. However, you cannot change the DefinitionType (see bellow).

MQ 5.3, "Application Programming Guide", SC34-6064-02, page 38 [60/643]

MQopen() + ObjDesc : If the ObjectName field in the ObjDesc parameter is the name of a model queue, a dynamic local queue is created with the attributes of the model queue.

MQ v 6.0, Application Programming Reference, SC34-6596-00, page 385 [409/665]

DEFTYPE : queue definition type (MQLONG) - indicates how the queue was defined

Page 445, MQ v 6.0, "MQSC Command Reference", SC34-6597-00

Mind MaxInstances - maximum number of simultaneous instances of an individual server-connection channel that can be started.

Also "MaxChannels=1000" in qm.ini

How can I display the actual value in use ?

MQ v 6.0, Application Programming Reference, SC34-6596-00, page 334 [359/665]

Top

Queue name resolution occurs at every queue manager each time a queue is opened. Its purpose is to identify the target queue, the target queue manager (which may be local), and the route to that queue manager (which may be null). The resolved name has three parts: the queue manager name, the queue name, and, if the queue manager is remote, the transmission queue.

When a remote queue definition exists, no alias definitions are referenced. The queue name supplied by the application is resolved to the name of the destination queue, the remote queue manager, and the transmission queue specified in the remote queue definition.

If there is no remote queue definition and a queue manager name is specified, or resolved by the name service, the queue manager looks to see if there is a queue manager alias definition that matches the supplied queue manager name. If there is, the information in it is used to resolve the queue manager name to the name of the destination queue manager. The queue manager alias definition can also be used to determine the transmission queue to the destination queue manager.

MQ v 7 Intercommunication [34/525]

MQ v 7 Intercommunication [290/525]

Queue name resolution is vital to DQM.

The sequence of events is as follows:

1. The putting application issues MQOPEN and MQPUT calls to put messages to the target queue.
2. During the MQOPEN call, the name resolution function detects that the target queue is not local, and decides which transmission queue is appropriate. Thereafter, on the MQPUT calls associated with the MQOPEN call, all messages are placed on this transmission queue.
3. The sending MCA gets the messages from the transmission queue and passes them to the receiving MCA at the remote computer.
4. The receiving MCA puts the messages on the target queue, or queues.
5. The getting application issues MQOPEN and MQGET calls to get the messages from the target queue.

Chapter 7, MQ v 7 Intercommunication [499/525]

Name resolution
When you open a WebSphere MQ queue, the MQOPEN call performs a name resolution function on the queue name that you specify. This determines on which queue the queue manager performs subsequent operations. This means that when you specify the name of an alias queue or a remote queue in your object descriptor (MQOD), the call resolves the name either to a local queue or to a transmission queue.

MQ v6.0 Application Programming Guide, page 113/601.

Working directory = \\MQ\Eines\QM_resolution_sample

En definitiva, al gestor QMM1 es fa un MQ_Put1() a la cua i gestor indicats al missatge. Evidentment, no existeixen a QMM1.

\\MQ\Eines\AMQSPUT_modificat\$_ru_qnr.cmd - modificar parametre NOMCUAPOSAR !

Els missatges apareixen al gestor CNT2 :

Top

• install MQ on each computer in the cluster.

  Queue managers themselves need to exist only on the computer on which you create them. In the event of failover, the MSCS initiates the queue managers on the other computer. The queue managers, however, must have their log and data files on a cluster shared disk, and not on a local drive.

• set Queue Manager to have MANUAL startup.
• set IBM MQ Service to have MANUAL startup.
• crear un queue manager amb cues PERSISTENTS
• move the queue manager :
  hamvmqm /m qmname /dd "f:\WebSphere MQ" /ld "f:\WebSphere MQ\log" aca : hamvmqm /m PRIMEQM /dd "f:\mqm" /ld "f:\mqm\log"
• registrar un tipus de recurs pel MSCS :
  haregtyp /r
• crear un recurs de tipus MQ
  nom := MQ PRIMEQM
• verificar que funciona : "bring online" + put/get messages
• FAILOVER !
• put a msg into a queue
• failover (back)
• veure missatge !!!

System Admin Guide, chapter 13 - Supporting MSCS

Top

Dead Letter Queue must exist if you dont want to have problems with the channels and/or the messages.

You can define it to the qmgr by using the DEADQ attribute on the ALTER QMGR command (see sample bellow), if you did not do it when qmgr was created.

DLQ must be persistent (DEFPSIST(YES)) and put-enabled !

DLQ monitor : MS71 !

If a message arrives at a queue manager but there is no queue there to receive it, the message is put on the dead-letter queue as usual. If there is no dead-letter queue, the channel fails and retries, as described in the MQ Intercommunication book.

"Clustering", v 5.3, SC34-6061-02, page 8.

Again : For example, if an application tries to put a message on a queue on another queue manager, but gives the wrong queue name, the channel is stopped and the message remains on the transmission queue. Other applications cannot then use this channel for their messages.

System Administration Guide, v 5.3, SC34-6068-02, page 26.

Again : If you do not, and the MCA is unable to put a message, it is left on the transmission queue and the channel is stopped.

Intercommunication, v 5.3, SC34-6059-02, page 13.

Again : If all the retries are unsuccessful, the message is written to the dead-letter queue. If there is no dead-letter queue available, the channel stops.

MQ v6 Intercommunication, SC34-6587, page 429 [451/573].

Top

Method #1 :

Define a remote queue pointing to an existing queue manager, but refering to a non-existing remote queue.

Find the message in the remote queue manager's DLQ.

Similar but faster : wrong "target" qmgr name for "open()"

Method #2 :

put a message into a XMIT queue, and get RC = d'0271 = MQFB_XMIT_Q_MSG_ERROR
Feedback indicating that a message channel agent has found that a message on the transmission queue is not in the correct format. The message channel agent puts the message on the dead-letter queue using this feedback code.

Requires a Running channel !

Find the message in the local queue manager's DLQ.

Method #3 :

Define a cluster of 3 queue managers : TQM1, TQM2 and TQM3.
TQM1 and TQM2 share the queue WLMQ1 on the cluster. This is, TQM3 sees two "cluster" queues.
On TQM3, define an ALIAS for WLMQ1, named WLMAQ.
If an application writes - connecting to TQM3 - into WLMAQ, then the messages are split between WLMQ1 on both TQM1 and TQM2.

Now, define a queue manager external to the cluster, named TQM4.
Define a REMOTE QUEUE on it, called RMQ99, pointing to WLMAQ on TQM3. And a CHANNEL also ... and activate it !
If an application sends a message connecting to TQM4 and using this Remote Queue RMQ99, it is placed in the TQM3's Dead Letter Queue with Reason Code d'2082 = MQRC_UNKNOWN_ALIAS_BASE_Q - Read about Dead-Letter Header.

Method #4 :

Attempt to receive a persistent message on a temporary dynamic queue. The message will fail and be routed to the DLQ with reason code: wrong persistence attribute.

Method #5 :

Send a message to Command Server, but not a Request Type :
md.MsgType = MQMT_REQUEST ; \\MQ\Eines\PCF_Inquiry_qmgr_Queues\codi.c

Method #6 :

Write to a cluster queue. Messages are always accepted. If a queue gets filled up, messages go to DLQ, Reason d'2053, x'0805

Method #7 :

Compile AMQSECHA using GCC or CL (v12) compiler. Set it into a triggered process using APPLICID. As RUNMQTRM wont be able to start the EXE, then the trigger message shall go to DLQ, and DLQH indicates 00.00.01.09

• DLQ handler {easy !}

• RFHutil

  RFHUtil + Read Queue + "DLQ" + uncheck "Include DLQH" on last Tab + Write Queue + "Q1" + and msg is on Q1, without DLQH

• MA0T ?

• If you want to move one single message: MQMON (SupportPac MO71)

El camp Reason ens diu el motiu per el que s'ha posat el missatge a la DLQ. Ha de ser del tipus MQFB_* o bé MQRC_* : mirar cmqc.h

Que fem si el MB posa un msg a la DLQ amb Dead Letter Header i amb Reason = MQFB_APPL_FIRST ? Mirar Event Log per problemes de Broker !

MQ v 6.0 System Admin Guide, chapter 12 : the MQ DLQ handler

Hubert : if you want to move all messages, specify a rule file (let's say rules.dlq) which contains the following lines

Then run the following command (Unix sample)

Using DLQ handler, developerWorks

Sending messages to a distribution list
In MQ on all platforms except z/OS, an application can send a message to several destinations with a single MQPUT call. This applies in both a distributed-queuing environment and a clustering environment. You have to define the destinations in a distribution list, as described in the MQ Application Programming Guide.

Top

A channel is a communication link used by distributed queue managers. MCA max speed = 35 MB/sec.

Do not confuse message channels with MQI channels. There are two categories of channel in MQ:

• Message channels, which are unidirectional, and transfer messages from one queue manager to another.
• MQI channels, which are bidirectional, and transfer MQI calls from a MQ client to a queue manager, and responses from a queue manager to a MQ client.

  There are two types of MQI channel : server-connection and client-connection.

MQ 6.0, "Application Programming Guide", SC34-6595-01, page 45 [65/601]

The definition of each end of a message channel can be one of the following types:

• Sender
• Receiver
• Server
• Requester
• Cluster sender
• Cluster receiver

A message channel is defined using one of these types defined at one end, and a compatible type at the other end. Possible combinations are:

• Sender - Receiver
  A sender in one system starts the channel so that it can send messages to the other system. The sender requests the receiver at the other end of the channel to start.
• Requester - Server
  A requester in one system starts the channel so that it can receive messages from the other system. The requester requests the server at the other end of the channel to start.
• Requester - Sender (callback)
  The requester starts the channel and the sender terminates the call. The sender then restarts the communication according to information in its channel definition
• Server - Receiver (server is used as a sender)
  This is similar to sender-receiver but applies only to fully qualified servers, that is server channels that have the connection name of the partner specified in the channel definition. Channel startup must be initiated at the server end of the link.
• Client-connection with Server-connection
• Cluster sender-cluster receiver

MQ v 6.0, "Intercommunication", SC34-6587, page 8 [30/573].

Supose we have this environment, where firewall prevents QM2 to start a normal Sender-Receiver channel from QM2.TO.QM1, but not TCP connections from QM1 to QM2.

We shall use a SERVER channel at QM2 and a REQUESTER channel at QM1. In this way, the data can flow from QM2 to QM1, and the channel is started from QM1.

A server connection channel object defines the name of a channel that a client can use to connect to a queue manager and the attributes of the MCA that hosts that connection.

This is different to all other channel types, because it is never used by the queue manager itself. Instead, an entry is added to a client channel definition table (CCDT) file, which can be distributed to other machines and used by client applications to configure their MCAs.

A channel can be caused to start transmitting messages in one of four ways. It can be:

• Started by an operator (not receiver, cluster-receiver or server-connection channels).
• Triggered from the transmission queue (sender, and fully-qualified server channels only). You will need to prepare the necessary objects for triggering channels.
• Started from an application program (not receiver, cluster-receiver or server-connection channels).
• Started remotely from the network by a sender, cluster-sender, requester, server, or client-connection channel. Receiver, cluster-receiver and possibly server and requester channel transmissions, are started this way; so are server-connection channels. The channels themselves must already be started (that is, enabled).

In Windows systems, start a listener as a background process at the receiver end of each channel.

• On the source queue manager, type:
  runmqlsr -t TCP -m source.queue.manager
• On the target queue manager, type:
  runmqlsr -t TCP -m target.queue.manager

Then start the channels, again as background processes:

• On the source queue manager, type:
  runmqchl -c source.to.target -m source.queue.manager
• On the target queue manager, type:
  runmqchl -c target.to.source -m target.queue.manager

System Administration, page 63 [87 of 567]

If you want the first message put in the queue DESA4 to start the associated transmit channel DESA3.DESA4, then define the queue this way :

The TrigData attribute must contain the name of the channel to be triggered.
The InitQ must be exactly SYSTEM.CHANNEL.INITQ

Application Programming, page 197 [219 of 643]

Transmission queue definition (for channel "QM2.QM1")

When the first message is put on this transmission queue, a trigger message is sent to the initiation queue, SYSTEM.CHANNEL.INITQ. The channel initiator gets the message from the initiation queue and starts the channel identified in the trigger data (QM2.TO.QM1). The channel initiator can only get trigger messages from SYSTEM.CHANNEL.INITQ so you should not use any other queue as the initiation queue.

Intercommunication, page 424 [448 of 815]

Try using message segmentation. That will allow MQ to split the message up into segments (allowing a file bigger than the max message length to be successfully transmitted). Make sure that the program receiving the file either reads the segments (using MQGMO_LOGICAL_ORDER) and reassembles the file itself, or ask MQ to do it (using MQGMO_COMPLETE_MSG). Segmentation sample, both JMS and J2EE.

Increase max msg size at :

• dead letter queue, all channels, SVRCONN channel used by client(s), SYSTEM.CLUSTER.TRANSMIT.QUEUE, ...
• qmgr(MAXMSGLENGTH), queue(), channel().

Top

• www.MQseries.net (u/k). Has a nice software repository ... ( \\MQ\Java\Repository )

• IBM'rs Developer Works, Blog , from HursleyOnWMQ.

• Open Source mailing list knowledge

• WebSphere MQ Forums, 2008 & WS

• MainframeForum (u/k)

• Sun Developer (u/k)

• Information about MQ : http://www.mywiseowl.com/

• Red Hat Cluster Suite + HW reqs + white paper. 500 USD !

• Linux High Availability : HeartBeat [free]

Si, al instalar un FP de MQ, ens diu AMQ4757 error, "MQ files in use", hem de fer :

• set PATH to include d:\Program Files\IBM\WebSphere MQ\bin
• reboot
• double chech the /var/mqm/qmgrs directory for any stray files in there.
• stop amq*, as amqmtbrn.exe (W2003S) at Task Manager.
• C:\Temp> WebSphereMQMDV6.0.2.0EnUs.exe -a mqpinuseok=1
• c:\zMQiMB\MQv602forWinIA32-FP0005-EnUs> WebSphereMQMDV6.0.2.5EnUs.exe -a mqpinuseok=1
• U200242A.exe MQINUSEOK=1
• start install but select "no elimine los archivos extraidos" (2-a opcion). Ir a esa carpeta y lanzar c:\temp\mq\amqicsdn -a MQPINUSEOK=1
  If the path is not set, an AMQ4756 is received !
• see amqicsdn.txt file (in TEMP directory, as d:\temp\2\ ...)
• start WebSphereMQMDV6.0.1.1EnUs.exe
• Si hay algún problema, la instalación mediante launchpad deja un "log" de eventos en el directorio temporal del usuario. El fichero se llama MQv6_Install_YYYY-MM-DDTHH-MM-SS.log y podemos ir al directorio temporal usando el comando cd %TEMP%
• Los niveles de mantenimiento están en una sequencia incorrecta. "El último mantenimiento aplicado es 0,0.(AMQ4758)". "CSD's out of sequence. Latest CSD applied is CSD06(AMQ4758)".
• use ListDLLs or Process Explorer

Top

When running a multi-threaded process such as the agent process, you might reach the soft limit for file descriptors. This gives you the WebSphere MQ reason code MQRC_UNEXPECTED_ERROR (2195) and, if there are enough file descriptors, a MQ FFST file.
To avoid this problem, you can increase the process limit for the number of file descriptors. To do this, alter the nofiles attribute in /etc/security/limits to 10,000 for the mqm user id or in the default stanza.

MQ v 6.0 for AIX quick beginnings, page 5 [19/72], GC34-6478-01.

Top

When you put a message on a queue, you can select to receive:

• An exception report message. This is sent in response to a message with the exceptions flag set. It is generated by the message channel agent (MCA) or the application.
• An expiry report message. This indicates that an application attempted to retrieve a message that had reached its expiry threshold; the message is marked to be discarded. This type of report is generated by the queue manager.
• A confirmation of arrival (COA) report message. This indicates that the message has reached its target queue. It is generated by the queue manager.
• A confirmation of delivery (COD) report message. This indicates that the message has been retrieved by a receiving application. It is generated by the queue manager.

If you want to receive a report message, specify the name of your reply-to queue in the ReplyToQ field; otherwise the MQPUT or MQPUT1 of your original message fails with MQRC_MISSING_REPLY_TO_Q.

Application Programming Guide, SC34-6595-01, page 22, [42/601]

PAN (Positive Action Notification) or a NAN (Negative Action Notification) event ... < Pending

Top

This indicates whether the message survives system failures and restarts of the queue manager.

Message persistence
Persistent messages are written to logs and queue data files. If a queue manager is restarted after a failure, it recovers these persistent messages as necessary from the logged data.

MQ 6.0 Programming Guide, pg 30 [50/601], SC34-6595-01

This means that the message survives system failures and restarts of the queue manager. Once the message has been put, and the putter's unit of work committed (if the message is put as part of a unit of work), the message is preserved on auxiliary storage. It remains there until the message is removed from the queue, and the getter's unit of work committed (if the message is retrieved as part of a unit of work).

MQ 5.3 Application Programming Guide, pg 171 [197/781], SC34-6062-02

Data for persistent messages is written to log buffers. These buffers are written to the log data sets when:

• a commit occurs
• a message is got or put out of syncpoint
• WRTHRSH buffers are filled

MQ z/OS Problem Determination Guide, GC34-6600-00

La persistencia de un missatge es determina :

• per l'aplicació
• per la cua

When you create a message, if you initialize the message descriptor (MQMD) using the defaults, the persistence for the message is taken from the DefPersistence attribute of the queue specified in the MQOPEN command. Alternatively, you can set the persistence of the message using the Persistence field of the MQMD structure to define the message as persistent or nonpersistent.

url

Persistence is really a message attribute and not a queue attribute.

Taylor @ developerWorks

No such thing as a persistent queue! ... but ... "Except for temporary dynamic queues", he told me with a grin, "they’re pretty much ALL persistent. Once you define 'em, they never go away." T. Rob

The messages may have come through an alias or a QRemote where DEFPSIST(NO) was set.

Default:

Problema : si els missatges son no persistents, i el log creix 7 GB cada dia, com podem investigar el seu contingut ? Create a new qmgr, start it, create a queue, put a persistent message, end the qmgr, print the log.

Imagine this setting :

1. we have a cluster queue - DEFBIND(NOTFIXED)
2. we stop the 2 qmgr instances containing it
3. we send 2000 messages to the queue
4. 1998 messages stay in SYSTEM.CLUSTER.TRANSMIT.QUEUE
5. those 2 remaining messages are not lost when qmgrs are started only if they are persistent

Top

Per cada "Data Source"

Ben Ritchie at DeveloperWorks article

A guindous tenim:

Top

Are those in which other resources, such as tables in a relational database, are also updated. When more than one resource manager is involved, there is a need for transaction manager software that uses a two-phase commit process to coordinate the global unit of work. Use global units of work when you also need to include updates to relational database manager software, such as DB2, Oracle, and Sybase.
We define two scenarios for global units of work:

• In the first, the queue manager itself acts as the transaction manager. In this scenario, MQI verbs control the global units of work; they are started in applications using the MQBEGIN verb and then committed using MQCMIT or backed out using MQBACK.

• In the second, the transaction manager role is performed by other software, such as TXSeries, Encina©, or Tuxedo. In this scenario, an API provided by the transaction manager software is used to control the unit of work (for example, EXEC CICS SYNCPOINT for TXSeries).

MQ 5.3 System Admin Guide - SC34-6068-02, chapter 11 : Transactional support.

Using CICS with WebSphere MQ & setting up the TXSeries interface to WebSphere MQ :

• CICS Administration Guide for Open Systems.
• TxSeries library [***]
• Using CICS with WebSphere MQ
• CICS Application Programming Guide

Top

Has a default capacity of 256 KB (2 MB after v7.1) Can be changed in the QMErrorLog stanza in the qm.ini file.
The log file sizes can be tuned using the MAXMQERRORLOGSIZE environment variable, the qm.ini file and/or the QMgr attributes in the WebSphere MQ Explorer Extended Attributes panel. url

Also can be acessed under Websphere MQ Explorer -> <qmgr_name> -> Properties -> Extended.

On UNIX and Linux systems, if you do not want certain error messages to be written to a queue manager error log, you can specify the error codes that are to be ignored using the QMErrorLog stanza.

Read here how to do it on Win systems :

Solution: use the environment variable MQS_FORCE_NTLANGID. Thanks, Jason ! MQS_FORCE_NTLANGID=3081 (0x0C09) - Spanish = 0x2C0A ;

Top

Para que podamos decir que una instalación de MQ está plenamente funcional, debe cumplir con todos los requisitos siguientes.

Tambien podemos decir que son una serie de pasos a verificar cuando se encuentra algún problema con el MQ, es decir, un botiquin de primeros auxilios.

• dspmq
  Debe mostrar todos los gestores de colas en estado "running". Si no lo está, daremos el comando strmqm <nombre gestor> tras determinar la causa de la parada (AMQER01.LOG)

• dspmqcsv <nom-qm>
  El estado del Servidor de Comandos debe ser "running"

• ps -ef | grep runmqlsr {linux} o tasklist /svc | find "runmqlsr" {win}
  El Listener de cada Gestor de Colas debe estar activo.

• runmqsc + display chstatus(*)
  Los canales de transmisión y recepción deben estar activos.

• runmqsc + display ql(*) curdepth // the current depth includes uncommitted messages and expired messages
  Las colas deben estar vacias o casi.

• runmqsc + display ql('QL.DLQ') curdepth
  La cola "papelera" debe estar vacia o indicará un problema

• ps -ef | grep amq
  Los procesos de MQ deben ejecutarse bajo un usuario del grupo "mqm" {linux} o MUSR_MQADMIN {win}.

• no hay entradas nuevas en los ficheros AMQERR01.LOG

• no se generan ficheros *.FDC

• MB, objectes : mqsilist + ps -ef | grep objecte
• Oracle, listener : ps -ef | grep lsn - (1)
• Oracle : ps -ef | grep ora_pmon - (2)
• Listeners MQ : ps -ef | grep lsr - (3)

(1)

(2)

(3)

Top

• look for probe point
• ubicació
  • Win : c:\P F\ibm\mq\errors\
  • AIX : /var/mqm/errors/

• contingut:
  -----} recv rc=lrcE_PERSISTENT_NOT_ALLOWED : | Major Errorcode :- zrcX_TASK_STARTED | lab2 1x FDC / min : 1.LOG owner | Major Errorcode :- xecS_E_BLOCK_NOT_FREE | {t400, nov'2011} --{ xihCSThreadClear ---{ xcsFFST

• resolució:

  Start a MQ trace :

  12:14:25.529793 10597.6 : Opening log file '/var/mqm/qmgrs/P9111/errors/AMQERR01.LOG' with flags 0x2, permissions 0x1B0 12:14:25.529816 10597.6 : ---}! xeeOpenFiles rc=xecP_E_ACCESS_DENIED

• com crear un FDC ?
  posar AMQERR01.LOG as "read only" or "owner" != mqm' !

Top

• when installing MQ Server, do install MQ Client also. It is very useful to connect to remote sites. Install also Samples and SDK (headers & libs).
• use ReplyToQ/ReplyToQMgr fields ; dont calculate response queue name
• use REPLY msg type in response message ; use DTG/Request msg type in request message
• if possible, use triggering to wake up receiving application(s). Use TRIGGER FIRST always, TRIGGER EVERY only if needed, and avoid TRIGGER DEPTH.
• copy Message_Id into Correl_Id field to correlate response to request
• request a Report if your message is rejected or expires
• know the default queue manager and verify it exists - used by runmqsc -w
• MQ & clustering :
  • define a Qmgr to be exclusively FR1 and another Qmgr to be FR2
  • use 2 (and no more) Full Repositories
  • make sure both FRs have a Cluster Sender Channel to the other FR
  • nice doc : best practices & monitoring
• MQ under HACMP :
  • use virtual IP in "IPADDR( )" at listener definition
  • use separated disks for Data & Log - better performance
  • if cluster is used, place FR's on HACMP machines
• MQ for MB :
  • increase LOG size, as default MQ log for Windows is not enough to deploy a 500 KB bar file
  • increase CM 2 BK channel message length; otherwise flows wont be deployed. Easiest way is: immediately after creating the queue manager, ALTER the MAXMSGL attribute of SYSTEM.DEFAULT.LOCAL.QUEUE, SYSTEM.DEF.SENDER, SYSTEM.DEF.RECEIVER, SYSTEM.DEF.CLUSSDR, and SYSTEM.DEF.CLUSRCVR. Even better: have a script to create queue managers for MB that does log size and MAXMSGL automatically.
• use scripts for administration, not GUI
• administer local, using telnet password
• if you need to use a remote MQ_Explorer, make it read-only
• always assign a dead letter queue (DLQ) to the queue manager, make it persistent and monitor it
• provide "alias" queue names to applications, that shall not know real name of local queues
• set message expiration, MD.Expiry, so messages wont live forever

Best practices experts chat.

More best practices:

• have and use a naming convention (MD00, GG24-4469, and MD04). See Intercommunication manual for suggested naming conventions

  Mind

  • queue name max length = 48 chars (all MQ objects, except ...)
  • channel name max length = 20 chars

  I like :

  • use full UPPERCASE for all objects (portability)
  • do not use lowercase letters in names. Object names in WebSphere MQ are case sensitive. We recommend you use uppercase letters only for naming.
  • do not include types in names. Using such names would reveal more internal details to client applications than you have to. Remote queues might soon become local or vice versa.
  • remote_queue_name := local_queue_name.QR ;
  • transmission queues shall have exactly the same name as the queue manager they are pointing to
  • QMGR name shall reflect the location, function and environment
  • CHANNEL names should reflect function and origin/destination qmgr. Common best practice : SOURCE_QMN.DESTINATION_QMN, without "TO." - it doesn't add meaning.
  • queues should not have the word QUEUE in their name, neither topology (local, remote, alias). Queue names shouldn't encode environment (Dev, Test, Production), so that apps and MB flows don't change when promoted.
  • applications should have a high-level qualifier, so that the names sort easily and the purpose of the queue is obvious, as APPXYZ.SERVICE1.REQUEST
  • do not use SPACES, neither forward slash, no percent sign, even allowed

  Samples :

  • QL.GRP1.JAVA - local queue, with permissions of group "one", used by Java programs
  • QL.INITQ.NAMERES - init queue used for name resolution sample
  • CHL.SVRCONN.JAVA - channel of type SVRCONN, used by all Java programs

  Reason for convention:

  • runmqsc - display with wild card char, as display ql(QL.*)
  • setmqaut - assign permissions using wild card char

• use standards (as JMS) whenever possible [?]
• build with performance in mind
  • message size affect performance - send only essential data
  • do not use persistence if possible
  • get first message - dont search for specific one
  • do not use MQPUT1()
  • do COMMIT() often
  • use threads but not too many
• avoid location assumptions and fixed queue names in programs
• when feasible, cluster your MQ servers
• design infrastructures with fewer queue managers with more queues
• capture completion and reason codes in all application MQI calls
• code applications to continually process messages
• close and disconnect connections properly
• be aware of the different features of various MQ clients
• use security best practices
  • ALTER CHANNEL(SYSTEM.DEF.SVRCONN) MCAUSER('nouser')
  • ALTER CHANNEL(SYSTEM.AUTO.SVRCONN) MCAUSER('nouser')
  • ALTER CHANNEL(SYSTEM.ADMIN.SVRCONN) MCAUSER('nouser')
• restrict remote administrative authority using SYSTEM.ADMIN.SVRCONN channels on production-level (restricted-access) boxes
• do not use the sample get/put utilities for production purposes
• use runmqlsr in lieu of inetd listener
• use SupportPacs to extend MQ functionality
• when not using third-party configuration tools, automate all configuration and changes via MQSC scripts
• use the appropriate logging mechanism
• configure automated maintenance of queue managers - save object configuration regularly, as well as security configuration
• schedule and apply fix packs on a regular basis

Few come from The top 15 WebSphere MQ best practices - gracias, Francisco !

More best practices : shared queues and application programs.

Mind there are MB Best Practices also, and MQ @ HACMP Best Practices too.

Top

Specifies whether the message-channel-agent program on an outbound MCA channel should run as a thread or a process. {Jeff improved}
PROCESS - The message channel agent runs as a separate process
THREAD - The message channel agent runs as a separate thread

In situations where a threaded listener is required to service a large number of incoming requests, resources can become strained. In this case, you should use multiple listener processes and target incoming requests at specific listeners though the port number specified on the listener.

MQSC Command Reference, SC34-6598-00, page 135 [155/501]

Channel type : Sender (SDR), Receiver (RCVR), Server (SVR), Requester (RQSTR), Cluster-Sender (CLUSSDR), Cluster-Receiver (CLUSRCVR).

Have to choose :

On the Sender channel MCA, define the MCATYPE as Thread, instead of Process, as performance is better.
Read "Running channels as processes or threads", page 168, (5.3) Intercommunication.

If you specify "process" on the channel definition, a RUNMQCHL process is started.
If you specify "thread", the MCA runs on a thread of the AMQRMPPA process, or of the RUNMQCHI process if MQNOREMPOOL is specified.
On the machine that receives the inbound allocates, the MCA runs as a thread or process depending on whether you use RUNMQLSR or inetd respectively.

Top

runmqlsr listens for new connections, then hands them off to amqrmppa. If you stop a listener, it has no impact on channels already running, but new incoming connections obviously will fail.

I prefer 'Admin' channels to have their own listener.

I have multiple Listeners on my Edge Queue Manager. Each incoming connection from another company has its own RCVR channel, and each company is told to use a specific port into my QM. Firewall rules prevent them from trying others.

Compte : "endmqlsr" ends all qmgr listeners - use GUI to stop a specific one

The backlog number would only apply for a MQ listener. After a listener program (like the MQ listener runmqlsr) has done a bind to a port like 1414, it will then call the listen() function to convert the socket into a passive or listening socket. One of the arguments that the program gives to the listen function is a backlog count. This number is used by the TCP implementation to create a unique backlog queue for this specific listening socket. So if you have 3 MQ listeners on say port 1414, 1415, 1416, each listener has its own backlog queue. This backlog queue is used to store partially established connections or established connections that have not been accepted by the MQ listener program. So this backlog count is just specific to an MQ listener and is used to adjust how many TCP connections that the listener has not accepted can queue up until they are discarded.

So you would probably want to consider adjusting the backlog number on a MQ listener if you feel that your listeners are getting overwhelmed with connection requests and you are finding evidence that some of these TCP connection requests to your listener are being dropped.

The runmqlsr has a much smaller overhead of connecting to and disconnecting from the queue manager because it only uses a single thread per connection rather than an entire process. A inetd listener has a significantly smaller capacity because of the need to create a new process for every client.
Channels can be started by using the INETD or the RUNMQLSR listener. INETD initiated channels use between 5 and 10 times more memory than RUNMQLSR channels.

MQ for AIX v6.0 - performance evaluations

You can now start and stop listeners and services using the START and STOP MQSC commands. However, the old commands strmqcsv, endmqcsv, runmqchi, runmqlsr, endmqlsr, and runmqtrm are still supported.

MQ v6.0 - migration information

If you specify "process" on the channel definition, a RUNMQCHL process is started. If you specify "thread", the MCA runs on a thread of the AMQRMPPA process, or of the RUNMQCHI process if MQNOREMPOOL is specified. On the machine that receives the inbound allocates, the MCA runs as a thread or process depending on whether you use RUNMQLSR or inetd respectively.

MQ v6.0 - intercommunication

Guess it has to be done using MQ_CONNX ! Structure has to be similar to Delphi multi-threading, ain't I right ? We can use WireShark to learn the details (and post them here) {Server2Server or Client2Server}

Top

Overview

Top

Since MQ v 6.0, we have Client File transfer & Server File transfer, but only under windows and linux x86 platform servers

A \\mq\bin we have :

• mqftrcv.exe - receive a file
• mqftsnd.exe - send a file
• mqftrcvc.exe - rcv a file, client version
• mqftsndc.exe - send a file, client version

To see the complete syntax enter the command without any parameters

url, url, mqftsnd

• -a : list all files and messages {default}
• -l : list all complete files
• -i : list all incomplete files
• -o : list all other messages
• -d : delete the specified file
• -g : receive a complete file
• -e : extract a complete or incomplete file
• -y : replace an existing file
• -r : asign a new name
• -f : select all files matching FileName

RCV params

Uses "Format : MQHRF2 " and "<FILENAME>nom_fitxer.txt</FILENAME>"

Can be used within your own script/batch files to (1) send, (2) list or (3) receive files:

• mqftsnd -q MY.QUEUE -f myfile.doc
• mqftrcv -q MY.QUEUE
• mqftrcv -q MY.QUEUE -f myfile.doc

Managed File Transfer is MQFTE !

File Extender, para manejar archivos con Message Broker.

File processing options with MB File Extender

PHP accesing FTP

Top

Think of PHP as "web version of Perl" ...

I know that it is easy to execute MQ command line calls from within PHP (or any other scripting language)

Linking WebSphere MQ to PHP : url - use our existing web service to take the requests from external vendors and push their XML data back into the queue inside our firewall - mqseries client library for PHP : url.

CPAN MQ Engineering group.

Philippe Tjon-A-Hen (Joomla user) has a very good PHP page, with lots of good PHP samples

On getting started he says: to compile the extension in your PHP, do as follows

• install the package(s)
  Extract MQ software and install the "MQSeriesSDK" to get the right libraries in place to compile the PHP extension.
• build the mqseries extension
  Go into the mqseries directory and run "phpize", "./configure" and "make" to create the .so file. The process should drop it into the default extensions directory.
  either load the extension dynamically into your scripts with dl() or add an extension= entry for it in your php.ini
• if needed, move it
  Be sure that the shared module for the extension is in the right directory for the PHP install to find it. (You can make a phpinfo() page if you're not sure where that is.)
• update your php.ini
  Add in a line to include the extension in your current setup. Remember, after any changes to the php.ini, you need to restart the web server.

See PHP install and PFP for MQ install.

Here are the samples (and API documentation):

To start with we define an array with connect options.

Once that is setup we can call the mqseries_connx method.

Next two calls are to illustrate how different mqseries_open methods work. The first opens an object of type queue manager for inquiry. The returned $obj handle can then be use as an input parameter when calling the inq method.

The second open simply opens a queue residing on the connected queue manager. This can either be a local queue, remote queue or even a queue alias.

Important here is the $obj and $obj_snd parameters. These are "call by reference" passed to the open methods. Afther successful completion of the method these hold the MQSeries object handle, which should be used when calling methods that work on mqseries objects.

In the next part an inquiry is made on the Queue Manager object and on the Queue object.

Note the 48 as the size of the expected character attribute. This is not the size of the &char_attr itself as in PHP we cannot specify a buffer. The parameter is needed to allocate enough memory that the method can succeed.

Finally we close the opened objects. Garbage collection is also done by the extention not closing object or connection will be closed by the extention.

And finally we close the connection.

mqseries Functions

PHP at RH v6.1 and MQ des PHP at RH v6.1

Top

Investigar callback : MQCB() i MQCTL() - see sg24-7583, v7 enhancements

CICS samples, from CICS Integration with MQ

Sample \\mq\tools\c\samples\amqscbf0.c code ...

MQ Solutions in a Microsoft .NET Environment - 7012 redbook.

Code Project : client/server sample, synchronization,

.Net support pack : MA7P

Top

• AMQAPI.exe {\\MQ\bin\} - simple API tester tool to perform basic operations as Connect(), Open(), Put(), Get() ... using GUI

• RFHutil - Support Pack IH03 - best tool to put()/get() messages withh ALL options/flags

  In case of problems, use verbose trace:

  SET RFHUTIL_VERBOSE_TRACE=1 SET RFHUTIL_TRACE_FILE=C:\RFHUtilTrace_local.txt

• MQMONNTP - WS MQ Administrator - Support Pack MO71
  Get latest version here. Requires IMQS23VN.DLL ... from MQ installation.

• SaveQMgr - Support Pack MS03.
  Since v7.1, use
  c:\> dmpmqcfg -h Usage: dmpmqcfg [-m QMgrName] [-n ObjName] [-t ObjType] [-x ExportType] [-o Format] [-a] [-z] [-s MsgSeqNo] [-q RplQName] [-r RmtQMgrName] [-c default|-c DEFINE CHANNEL..] -m Queue manager name. -n Object name or a generic object name. -t Object type: all, authinfo, channel, clntconn, comminfo, listener, namelist, process, queue, qmgr, service, topic. -x Export type: all, object, authrec, chlauth, sub. -o Format: mqsc, 1line, setmqaut, grtmqmaut. -a Dump all attributes. -z Suppress warnings. -s Reset channel message sequence number -q Reply to queue name, default SYSTEM.DEFAULT.MODEL.QUEUE -r Remote queue manager name (queued mode) -c Client connection: default, DEFINE CHANNEL(chlname) CHLTYPE(CLNTCONN) ...

• HealthCheck - Support Pack MH01. Verifies MQ recomendations (DLQ, TxQ, etc)

• Support Pack IS02 : adds MB Explorer to MQ Eclipse.

• (pend) save queue to file & viceversa. - Support Pack "MO03, WebSphere MQ Queue Load / Unload Utility"
  qload -m FEM49P01 -i Q.FEM49.NAR_FAZ_REPLY.1 -f /faz1/scripts/descarga.bin /usr/lib/dld.sl: Can't open shared library: /opt/mqm/lib/libmqm.sl /usr/lib/dld.sl: No such file or directory ARIES32: Core file for PA32 application saved to /faz1/scripts/core.qload Abort(coredump)

• cua circular : Q (MA01) + perl, by T-Rob !

• MQ Visual Edit - export from a queue to PDF !

• {delphi} issue a MQPUT1() with all params - uses MQCONNX() - mqrc = 2460 0x0000099c MQRC_HMSG_ERROR
  \\Delphi\MQ\put1\Put1.pas
• {delphi, multithread} establish as many Client Connections as system resources allow - uses MQCONNX()
  \\MQ\Eines\AMQSCNXC_proves_Conexio_Client\IMI_Buscar_Max_Conexiones
• {delphi, multithread} empty a queue using mqget() and count messages (bytes) per second - MQconn()
  \\Delphi\MQ\Buida_Cua\Buidar_Cua.pas
• {delphi/java, pend} wqm connection/security exerciser - TMM33 :
  • queue manager
  • svrconn channel
  • host + port
  • usr + pwd

Clients:

• 6q (7.0.1 @ w7)
• patan (7.5 @ xp)
• rac (7.5 @ xp)
• t400 (7.0.1.@ xp)

Servers:

• p7029 (7.0.1 @ SLES 10)
• p9111 (7.0.1 @ SLES 10)
• rh64 (7.5 @ rh)

Results :

• a decent MQ internal trace
• browse MQ objects from a browser
• customize MQExplorer to display custom-built scenario(s)
• cluster : ha de ser fàcil especificar quins gestors poden entrar al cluster i quins no poden
• gestió de seguretat - visualitzacio dels drets dels usuaris vs objectes
• scan IP range for MQ ports - use MQCONNX()
• inquiry queue / channel throughput : inquiry_queue (put count / get count) ; inquiry_channel (bytes received / bytes sent) - "inquiry" pot demanar "listener status" ?

Top

Llista de certificacions, per número.

Tinc :

• Certified Solution Developer -- WebSphere MQ V5.3 [2006-05-03, Las Vegas 2006]
• Certified System Administrator -- WebSphere MQ V6.0 [2006-12-21, Madrid 2007]
• MQ 6.0 Solution Designer [Barcelona, 2008]

url, url.

2012 :

• MQ v7.0 System Administrator - test 000-374 : 59% of 56 questions in 105 minutes.
• MQ v7.0 Solution Designer - test 000-376 : 57% of 49 questions in 90 minutes.

996, objectives, DevWorks.

998 objectives, assessment tests, training resources

Top

MQ Advanced Message Security v 7.0.1 documentation

Top

• cuál es el problema actual ?
• revisión de seguridad (SVRCONN)
• que cambios en el futuro próximo hay previstos ?
• análisis de la infraestructura actual
• posibles mejoras (migración a ultima versión)
• buenas practicas, formación, FixPacks, SupportPacks,

• configuració MQ :
  • determinar a quin filesystem hi ha el codi del producte - /opt/mqm/
  • determinar on es el filesystem amb les dades - /var/mqm/
  • determinar quina versio de MQ estem fent servir
  • on es MQS.INI
  • on son QM.INI
  • on es AMQERR01.LOG general (i dels clients)
  • on son els *.FDC generals - /var/mqm/errors/
  • determinar quants gestors hi ha
  • quins ports/listeners fa servir cada gestor
  • quin tipus de log fa servir cada gestor
  • revisar errors particulars : AMQ*.LOG & *.FDC's
  • determinar si es fa servir clustering i/o HACMP
  • determinar si es fa servir SSL
  • determinar si es fan servir User-Exit's
• arquitectura dels missatges
  • req/rsp
  • client connections
  • persistencia
  • cues estatiques/dinamiques o sense defincio
• quins missatges es fan servir
  • dimensionar cues i canals : tamany màxim
  • dimensionar log : persistència = tamany mitjà * num missatges al dia * dies a recuperar
• configuracio gestors : existeix DLQ, és persistent i és gran (eg 500.000) i és buida
• configuracio gestors : tipus de LOG, tamany
• tamany dels LOG vs espai al disc
• definició de cues - les cues de transmissió han de ser grans (eg 50.000)
• definició dels canals - autoTrigger
• definició dels listeners - MAXINST & BACKLOG. Engegar / aturar amb el gestor
• definició dels clients - fer servir CCDT / AMQCLCHL.TAB - tenen MCAuser
• backup : període SAVEQMGR - guarda configuració objectes MQ
• eina de MQ Explorer : MQ Explorer + boto dret + "Pruebas" : "39 pruebas"
• seguretat : quins usuaris hi ha al grup MQM - see /etc/groups & /etc/password
• seguretat : tots els usuaris podem accedir a totes les cues ?
• seguretat : usuari obligatori "mqm_cli_user" a SYSTEM.DEF.SVRCONN + SYSTEM.AUTO.SVRCONN + SYSTEM.ADMIN.SVRCONN

• separar filesystem de les dades (/var/mqm) del filesystem dels LOG (/var/mqm/log) i dels errors (/var/mqm/errors)
• fer servir la trassa de MQ en cas de problemes - details
• fer servir les proves incloses al MQ Explorer - details, ex-MH01.

• AMQP -
• a queue manager alias is a remote queue definition without an RNAME field.
• managed subscriptions :
  if an application has no specific need to use a particular queue as the destination for those publications that match its subscription, it can make use of the managed subscription feature. If an application elects to use a managed subscription, the queue manager informs the subscriber about the destination where published messages will be sent, by providing an object handle as an output from the MQSUB call.

Top

• SYSTEM.ADMIN.STATISTICS.QUEUE

• es pot conectar un JMS using MQ Connection Factory a un canal SVRCONN, es a dir, fer servir MQ Server API ?

• AMS

• monitoritzar el nombre de missatges que passen per un canal ...

• MQ LLM (Haifa), Low Latency Messaging - InfoCenter - molts misstg/seg.
  Can deliver over 1 million 120 byte messages per second with less than 80 microsecond latency, or 8 million smaller messages per second on common x86 servers.

• seguretat :
  • OAM = object authority manager (amqoamd)
  • MQ-ESE - Extended Security Edition.

• DMZ - acces from Internet

• DMZ - as Renó / $intax {\\dep\*} - url about MQIPT

• MQ/FTE - File Transfer Edition. Information Center Home page SAG's

• mq trace - read it when MQRC=2042 (indy)

• use STRINGS at Solaris to display queue data !
  root@sercom1 # strings /var/mqm/qmgrs/sebas/queues/c1/q

• send a msg to command server to create a queue - display format here (amqsaicq.c)
  The mqExecute (MQAI) call ... ... creates the PCF structure required, ... sends it to the command server, ... and receives the reply from the command server into the response bag. */

  Stop the command server and hex-dump the message here ...

  What format is required by command server ? MQAI ??

• activar una alarma per "cua molt plena" amb un llindar, i veure el missatge que es genera (i on es posa)

• SSH en Cluster.
• XA manager

Top

• Quick beginnings for AIX
• Quick beginnings for HP-UX
• Quick beginnings for iSeries
• Quick beginnings for Linux
• Quick beginnings for Solaris
• Quick beginnings for Windows
• Application Programming Guide

V 6.0 Information Center

Top

Sonic MQ [\\MQ\Competencia\Sonic MQ]

mqjeff : JMS used with a SonicMQ JMSProvider would feed the JMSInput node. At that point, there would be no more Sonic code at all. Everything is now Broker. This also requires no user code of any kind that knows anything about SonicMQ.
Broker can easily, using a JMS->MQ Transformation node, convert the JMS headers into a set of MQ headers, which could then be output using an MQOutput node.
Three Broker nodes, all fully supported, no custom user code at all....

• sonic esb to wmb
• sonic mq to wmq

Amazon Machine Images (AMIs) are available for the following IBM products: url - MQ no hi es (Abril 2010).

Top

MSDN

Top

See also amqp products

Apache ActiveMQ is an open source (Apache 2.0 licensed) message broker which fully implements the Java Message Service 1.1 (JMS). wiki. Mine.

MOM : RabbitMQ (implementa AMQP), StormMQ, 0MQ ...

Tuxedo (Transactions for Unix, Extended for Distributed Operations) is a middleware platform used to manage distributed transaction processing in distributed computing environments. Tuxedo is a transaction processing system or transaction oriented middleware, or enterprise application server

Tuxedo

Habari client libraries ... provide access to standards-based, enterprise quality solutions for distributed messaging.

Top

mq-light @ developer.ibm.com

You can start working with the MQ Light Node.js API by installing it to your working directory using

MQ web homepage

Make it possible to administrate your Websphere MQ environment with your favorite browser or programming language

mq web app is the official web application for MQWeb.

Top

• video
• product website
• KC, knowledge center
• community, wiki
• vmWare - "2.240.742.536 MQ_APPL_M2000_DC_FOR_IBM_INT_U_V1.zip"

  Commands

  • clock - set date or time
  • configure terminal - enter Global Configuration Mode
  • diagnostics - provide diagnostic info
  • disable - enter User Mode
  • disconnect - terminate a user session
  • echo - echo text to the console
  • exec - execute a configuration script
  • exit - terminate the CLI connection
  • help
  • login - login as specific user
  • mqcli - enter MQ Configuration Mode
    • help mq - general MQ administration commands
      • crtmqm - create a queue manager
      • dltmqm - delete a queue manager
      • dspmq - display info about queue managers
      • dspmqini - display queue manager initialization parameters
      • dspmqtrn - display in-doubt and heuristically completed transactions
      • dspmqvar - display global and queue manager environment variables
      • endmqm - stop a queue manager
      • rsvmqtrn - resolve in-doubt and heuristically completed transactions
      • runmqsc - run MQSC commands
      • runswchl - switch transmission queue for cluster channel
      • setmqini - configure queue manager initialization parameters
      • setmqvar - configure global and queue manager environment variables
      • status - display status information
      • strmqm - start a queue manager
    • help cert - channel security certificate administration commands
    • help diag - MQ problem diagnostics commands
      • dmpmqcfg - dump the configuration of a queue manager
      • dspmqerr - display error logs and FDC files
      • dspmqrte - determine the route that a message has taken through a queue manager network
      • dspmqver - display MQ version and build information
      • endmqtrc - end trace
      • mqrc - display info about return codes
      • runmqras - gather MQ diagnostic info (to submit to support)
      • strmqtrc - enable trace (or report the level of tracing in effect)
    • help ha - high availability administration commands
    • help user - messaging user and group administration
  • ntp - identify network time server
  • ping - send an ICMP ECHO_REQUEST
  • show - display statistical or configuration data
    • show ip - display all IPs
  • shutdown - restart or shutdown the appliance
    • shutdown reload - specify an immediate restart
    • shutdown reboot 10 - specify a reboot in 10 seconds
    • shutdown halt 60 - specify a shutdown in 1 minute
  • test tcp-connections - test TCP reachability
  • top - return to initial login mode
  • traceroute - discover the route to a host

  Configuration

  • u/p = admin/admin
  • SSH at port 22
  • WebGUI - MQ console

Top

• write some MQ header field as ReplyToQueue into JMS message properties
• implement Java agent to read q depth and set CLWLPRTY properly ( see here ) {\\T430\MQ\Java\PCF} [lab]
• define a threshold and display the generated alarm when the queue depth reaches this value {\\T430\MQ\Events\queue_full}
• [lab] what is the throughput of a global cache ?
• [lab] how fast does CLWLPRTY change propagate ? Why its range of values ?

Links, urls :

• handling PCF messages, from "Using Java"
• how to issue MQSC Commands using the Java API -> com.ibm.mq.pcf {websphere mq classes for java} & PCF and AI
• inquire Log
• Samples
  MS0B -> \\T430\MQ\MQ_SupportPacks\MS0B-MQ_Java_classes_for_PCF\ms0b\com.ibm.mq.pcf\examples (+) \\T430\MQ\tools\pcf\samples\*.java = \\T430\MQ\Java\PCF
  \\T430\MQ\Eines\PCF_Inquiry_qmgr_Queues
• to get the queue depth, we shall have to use MQIA_CURRENT_Q_DEPTH I guess
  Perfect sample : PCFMessageListQueueDepth.java
  Java code
• to set { put(disabled) or CLWLPRTY(n) } we shall use MQCMD_CHANGE_Q
• the Change Queue (MQCMD_CHANGE_Q) command changes the specified attributes of an existing WebSphere MQ queue.
  Required parameters (Change and Create Queue) - QName (MQCFST) : Queue name (parameter identifier: MQCA_Q_NAME)
  InhibitPut (MQCFIN) - whether put operations are allowed (parameter identifier: MQIA_INHIBIT_PUT). Specifies whether messages can be put on the queue. The value can be: MQQA_PUT_ALLOWED = Put operations are allowed. MQQA_PUT_INHIBITED = Put operations are inhibited.
• MQ_v7_PCF_and_AI_csqzac06.pdf
  Response data : CurrentQDepth (MQCFIN) Current queue depth (parameter identifier: MQIA_CURRENT_Q_DEPTH).
  Optional parameters (Inquire Queue Status) : QStatusAttrs (MQCFIL) Queue status attributes (parameter identifier: MQIACF_Q_STATUS_ATTRS). MQIA_CURRENT_Q_DEPTH The current number of messages on the queue.
  Response data if StatusType is MQIACF_Q_STATUS : CurrentQDepth (MQCFIN) Current queue depth (parameter identifier: MQIA_CURRENT_Q_DEPTH).

{Peter & Saper & Luisa} Cluster Queue Monitoring sample program (AMQSCLM) : 7.1 7.5
The cluster queue monitoring sample program uses the IPPROCS (open for input process count) value of a queue to determine whether that queue has any consumers. A value greater than 0 indicates that the queue has at least one consuming application attached. Such queues are active. A value of 0 indicates that the queue has no attached consuming programs. Such queues are inactive. The cluster queue monitoring sample program activates a cluster queue by setting the local CLWLPRTY value to 1. The program deactivates a cluster queue by setting its CLWLPRTY value to 0

CLWLPRTY(integer)
Specifies the priority of the queue for the purposes of cluster workload distribution. This parameter is valid only for local, remote, and alias queues. The value must be in the range zero through 9 where zero is the lowest priority and 9 is the highest.

See mq_admin

Top

Mira que m'envia en Albert :

MQ $ in UK : contracts, jobs.

Top

• Michael Dag, email, working at http://www.mqsystems.com/, and their product MQ Architect