[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference pamsrc::decmessageq

Title:NAS Message Queuing Bus
Notice:KITS/DOC, see 4.*; Entering QARs, see 9.1; Register in 10
Moderator:PAMSRC::MARCUSEN
Created:Wed Feb 27 1991
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2898
Total number of notes:12363

2816.0. "How can node down timeout be affected with OpenVMS DmQ V3.2B" by SEFI04::VOLKMAN () Thu Mar 20 1997 12:23

	We are seeing a delay of 3-4 minutes before we recieve a group
	down message when a node crashs ungracefully.  We need to have
	notification within 30 seconds if possible. I've seen mention
	of some parameters 'DMQLD_PING_INTERVAL' and 'HEARTBEAT'
	but I can't seem to find those symbols in the DMQ*.EXE files.
	Is there a way to control this with DECmessageQ V3.2B on OpenVMS
	ALPHA?
Thanks,
Bill V.
T.RTitleUserPersonal
Name		DateLines
2816.1XHOST::SJZKick Butt In Your Face Messaging !Thu Mar 20 1997 13:275
    
    The DMQLD_PING_INTERVAL and DMQLD_PING_TIMEOUT are specific
    to the non-VMS platforms.
    
    _sjz.
2816.2DALTEL::STEPHENSThu Mar 20 1997 17:3018
Hi Bill,

I'm going to assume you are using TCP and not DECnet as a transport.
Further, I'm going to assume you are using UCX, and on out on the limb
to say you are using the latest version of UCX :-)  If all of this is
true, then there are a couple of knobs you can turn to lower this 
notification event, keep in mind however, that these knobs are global 
for the TCP community on this node.  They are the probe and drop timers.

$ucx show proto tcp /par - lets you see the current values

To lower the values use something like:

$ucx set proto tcp/probe=10
$ucx set proto tcp/drop=30

Hope this helps,
Bruce