| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Big problem - please help! hearbeatINT and disconnectINT |
« View previous topic :: View next topic » |
| Author |
Message
|
| ping master |
 Posted: Fri Nov 22, 2002 7:42 am Post subject: Big problem - please help! hearbeatINT and disconnectINT |
 |
|
Centurion
Joined: 20 Sep 2002
Posts: 116
Location: USA
|
I have a workstation running XP with 5.3 server connecting to a UNIX box running mq 5.2.
when I disconnect the dongle from the laptop, I expect the the sender channel(USER.to.X14) to go inactive when it recognizes the LAN connection is gone. Problem is, the OS sees it and says it's disconnected, BUT the channel remains in a runnng status, and never goes to inactive..
Ive read a couple posts that seem relative to this, but I am kinda confused on the solution..I would like MQ to recognize the connection is gone as soon as possible- so should I set the heartbeat interval to a very low number and set a disconnect interval for a very low number on the workstation, then go to the UNIX box and alter the same channel to change it's heartbeat int to the same low number?? I noticed currently on the workstation the heartbeat int is 300, disconnect int is 0, and on the UNIX box the heartbeat int is set to 0.
thanks alot |
|
| Back to top |
|
 |
| cronydude |
| Posted: Sun Nov 24, 2002 8:06 pm Post subject: |
|
|
Voyager
Joined: 11 Nov 2001
Posts: 85
Location: US
|
Note that a value of 0 to DISCINT will mean to never disconnect....
_________________
Regs,
crony
IBM Certified Specialist - MQSeries |
|
| Back to top |
|
|
| pgorak |
| Posted: Sun Nov 24, 2002 10:49 pm Post subject: |
|
|
Disciple
Joined: 15 Jul 2002
Posts: 158
Location: Cracow, Poland
|
After the heartbeat interval expires, the sending side of the channel will go into RETRYING state. The larger value of the two HBINT attributes (at sending and receiving side) is used, so in your case you have to wait 300 seconds for the sender/server channel to go into RETRYING state (it will not go into INACTIVE state).
Piotr |
|
| Back to top |
|
|
| ping master |
| Posted: Mon Nov 25, 2002 1:58 pm Post subject: |
|
|
Centurion
Joined: 20 Sep 2002
Posts: 116
Location: USA
|
So in other words, I need to change my HBINT on the workstation to 60 and the discInt to 90? does that sound correct? so If my connected failed 1 1/2 minutes is the longest is would take to go inactive?
thanks for the resposes |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Nov 25, 2002 6:02 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
With a DISCINT of 90 and a heartbeat of 60, the channel will disconnect after 90 seconds. I believe the SNDR channel will go inactive, not retrying. If the SNDR can disconnect cleanly, it sends a message to the RCVR for it to go inactive as well.
The SNDR channel would go to retrying (among other reasons) if after it went inactive it tried to reestablish connection with the RCVR channel and it was still running.
The RCVR might be still running with an inactive SNDR because of a network outage. The network goes out, the sender eventually disconnects on its own, but can't send the disconnect message to the RCVR, so it stays running. This is where Heartbeat comes in. After the HBInterval passes, the RCVR sees that it never got the HB back from the SNDR, so it assumes the network is down and should stop waiting for the SNDR to send messages, and now goes Inactive.
What if the SNDR tries to reconnect after a network outage but before HB tells the sender to go inactive? Then AdoptMCA comes in and lets the SNDR establish connection with the RCVR, even if it is still running, thinking it has a connection with a long gone SNDR.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
|
|
