ASG
IBM
Zystems
Cressida
Icon
Netflexity
 
  MQSeries.net
Search  Search       Tech Exchange      Education      Certifications      Library      Info Center      SupportPacs      LinkedIn  Search  Search                                                                   FAQ  FAQ   Usergroups  Usergroups
 
Register  ::  Log in Log in to check your private messages
 
RSS Feed - WebSphere MQ Support RSS Feed - Message Broker Support

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Manual Intervention frequency

Post new topic  Reply to topic Goto page 1, 2, 3, 4  Next
 Manual Intervention frequency « View previous topic :: View next topic » 
Author Message
mattfarney
PostPosted: Tue Sep 13, 2011 1:59 pm    Post subject: Manual Intervention frequency Reply with quote

Disciple

Joined: 17 Jan 2006
Posts: 167
Location: Ohio

I have another unusual question.

Assuming that I have broker installed on several clustered windows machines and an automated way of handling the logs...

How often do you find yourself needing to do something to the environment? (Starting a channel that didn't/wouldn't start, repairing a damaged object, restarting a broker, recycling the queue manager, etc.)

I'm trying to figure out if our problem incidence rate is high or low.

-mf
Back to top
View user's profile Send private message
fjb_saper
PostPosted: Tue Sep 13, 2011 2:13 pm    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY

mattfarney wrote:
I have another unusual question.

Assuming that I have broker installed on several clustered windows machines and an automated way of handling the logs...

How often do you find yourself needing to do something to the environment? (Starting a channel that didn't/wouldn't start, repairing a damaged object, restarting a broker, recycling the queue manager, etc.)

I'm trying to figure out if our problem incidence rate is high or low.

-mf

None of the above ever, but our env is AIX.
Once a while (blue moon), we need to recycle a qmgr to allow it to roll the logs (linear logging ).
However we look every day at the flow error queues to resubmit stranded messages (1 or 2).
_________________
MQ & Broker admin
Back to top
View user's profile Send private message Send e-mail
mattfarney
PostPosted: Tue Sep 13, 2011 2:16 pm    Post subject: Re: Manual Intervention frequency Reply with quote

Disciple

Joined: 17 Jan 2006
Posts: 167
Location: Ohio

fjb_saper wrote:
None of the above ever, but our env is AIX.

I have noticed that our Windows environment is more problematic than our HP-UX environments.

-mf
Back to top
View user's profile Send private message
mqjeff
PostPosted: Wed Sep 14, 2011 3:44 am    Post subject: Reply with quote

Grand Master

Joined: 25 Jun 2008
Posts: 17447

In general, I would expect that windows machines would have a higher rate of broker/qmgr restart than non-windows machines. This is because Windows software undergoes a lot more routine maintenance than most unix systems undergo or require.

Whether or not this correlates to a higher incident of actual mq/broker failure, I can't speculate. I wouldn't expect it to directly correlate, but it's not unreasonable to expect some impact.
Back to top
View user's profile Send private message
Vitor
PostPosted: Wed Sep 14, 2011 4:16 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

mattfarney wrote:
I'm trying to figure out if our problem incidence rate is high or low.


Sounds high. Even with the high level of routine maintenance my most worthy associate mentions, you tend not to do anything rather than watch the queue managers stop and start.

This current site (rightly or wrongly) deliberately does not patch Windows servers unless there's a percieved problem with said server or it's the 5th April / 5th October. None of the MQ servers have required any attention as you describe since April; there's one unloved one who didn't get maintenance for complex reasons who's been working normally for nearly a year.

Windows 2000 / 2003 / 2008, all in a WMQ cluster, MCSC, some point to point connections.
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
lancelotlinc
PostPosted: Wed Sep 14, 2011 4:46 am    Post subject: Re: Manual Intervention frequency Reply with quote

Jedi Knight

Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA

mattfarney wrote:
several clustered windows machines


Stop right there. You're asking for trouble. Clustering and Windows are not a good combination.

If you elaborate on the business purpose of the configuration, we can provide alternatives to consider.
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER
Back to top
View user's profile Send private message Send e-mail
Vitor
PostPosted: Wed Sep 14, 2011 5:01 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

lancelotlinc wrote:
Stop right there. You're asking for trouble. Clustering and Windows are not a good combination.


You stop right there. Clustering works fine on Windows and is a perfectly valid combination. Justify that comment without simply evangelising a "better" platform, i.e explain why Windows and clusters are not good / trouble rather than explaining why <a different OS> and clutsers are good.

lancelotlinc wrote:
If you elaborate on the business purpose of the configuration, we can provide alternatives to consider.


Probably the same as the business purpose of any clustered configuration.
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
lancelotlinc
PostPosted: Wed Sep 14, 2011 5:09 am    Post subject: Re: Manual Intervention frequency Reply with quote

Jedi Knight

Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA

Vitor wrote:
lancelotlinc wrote:
Stop right there. You're asking for trouble. Clustering and Windows are not a good combination.


You stop right there. Clustering works fine on Windows and is a perfectly valid combination. Justify that comment without simply evangelising a "better" platform, i.e explain why Windows and clusters are not good / trouble rather than explaining why <a different OS> and clutsers are good.

lancelotlinc wrote:
If you elaborate on the business purpose of the configuration, we can provide alternatives to consider.


Probably the same as the business purpose of any clustered configuration.


I'm asking the OP to clarify the business purpose of such configuration. There may be more tenable solutions to offer.

Many people like Windows Clusters until they actually have to recover a production outage that should have been automatically failed over.
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER
Back to top
View user's profile Send private message Send e-mail
Vitor
PostPosted: Wed Sep 14, 2011 5:19 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

lancelotlinc wrote:
I'm asking the OP to clarify the business purpose of such configuration. There may be more tenable solutions to offer.


I'll conceed that the OP could mean clustered in the sense of MCSC or other failover tachnology, or clustered in the sense of a WMQ cluster and hence clarification is needed there.

But by the same token you should have been clearer: it's ambiguous if you're saying "using Windows for HA is asking for trouble" or "using WMQ on Windows for load balancing is asking for trouble".

lancelotlinc wrote:
Many people like Windows Clusters until they actually have to recover a production outage that should have been automatically failed over.


Many people like Windows until they try and use it in production & try and scale it, stabalise it or in other ways use it.

I'm interested to see the alternatives you present, especially those applicable to a site with a large Windows investment (including non-MSoft applications) disinclined to move platforms. Because, as I indicate above, this site uses MCSC & the automatic failover does suck.
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
lancelotlinc
PostPosted: Wed Sep 14, 2011 5:42 am    Post subject: Re: Manual Intervention frequency Reply with quote

Jedi Knight

Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA

Vitor wrote:
I'm interested to see the alternatives you present, especially those applicable to a site with a large Windows investment (including non-MSoft applications) disinclined to move platforms. Because, as I indicate above, this site uses MCSC & the automatic failover does suck.


In the 1990s, I did see alot of zealously emotional association to a particular technology by IT management. In the 21st century, cooler heads have prevailed where there is an open-mindedness to think about things from a homogenous perspective. That is, if we have a mainframe, we don't rule out using a midrange; if we have Windows, we don't rule out using a Unix solution.

Overall, I consider myself to be agnostic when it comes to which technology. I get excited when I see advancement in the genre, by leaps and bounds, as in the capabilities that Power 7 technology offers. Not because of who makes it, but because of what Power 7 is capable of.

Windows is a nice user interface. I prefer Windows user interface over OS/2. I prefer Windows user interface over Unix user interface. In fact, Windows code base started from Unix code base, so in many respects, Windows was born from Unix.

I find that Microsoft unnecessarily complicates its products. I believe the motivation for the over-complication is as a barrier to entry to competitors. Therefore, due to the overly complicated infrastructure within the Windows OS itself, it cannot operate simply. Lack of simplicity is what impedes its ability to be agile. And automatically recoverable most of the time.

In my experience, on a clustered Windows farm, 7 times out of 10, the automatic failover will not work as intended.
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER
Back to top
View user's profile Send private message Send e-mail
Vitor
PostPosted: Wed Sep 14, 2011 5:47 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

lancelotlinc wrote:
In my experience, on a clustered Windows farm, 7 times out of 10, the automatic failover will not work as intended.


So to be clear, we're taking about "cluster" in terms of HA not WMQ?

lancelotlinc wrote:
I'm asking the OP to clarify the business purpose of such configuration. There may be more tenable solutions to offer


So given the business purpose to maintain system availability, what are the more tenable alternatives? Given that even from the open minded homogenous perspective you evangelise (again) the business has Windows application necessary to their operation which need WMQ conectivity and high availability?
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
lancelotlinc
PostPosted: Wed Sep 14, 2011 5:55 am    Post subject: Re: Manual Intervention frequency Reply with quote

Jedi Knight

Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA

Vitor wrote:
lancelotlinc wrote:
In my experience, on a clustered Windows farm, 7 times out of 10, the automatic failover will not work as intended.


So to be clear, we're taking about "cluster" in terms of HA not WMQ?

lancelotlinc wrote:
I'm asking the OP to clarify the business purpose of such configuration. There may be more tenable solutions to offer


So given the business purpose to maintain system availability, what are the more tenable alternatives? Given that even from the open minded homogenous perspective you evangelise (again) the business has Windows application necessary to their operation which need WMQ conectivity and high availability?


It is a more philosophically different approach to manage uptime. The business value delivered is having a system that is up. The question becomes: what can a business do to ensure that their systems are up when they need to be?

I contend that the active-passive form of uptime insurance does not deliver the best business value. I advocate the concept that more business value is delivered for less cost following an active-active or active-active-active model. This philosophical shift is platform agnostic. It can apply equally to Windows or Unix. Not so equal to z/OS (due to the high cost of *everything* on that platform).
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER
Back to top
View user's profile Send private message Send e-mail
Vitor
PostPosted: Wed Sep 14, 2011 6:04 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

lancelotlinc wrote:
Not so equal to z/OS (due to the high cost of *everything* on that platform).


And the decreased risk of downtime on z/OS.

So you're happy WMQ clustering is not trouble on Windows? For clarity?

And your assertion that "Clustering and Windows are not a good combination" applies only to an active/passive HA cluster?

What's your recommendation for implementing active/active (or active/active/active) on Windows?
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
lancelotlinc
PostPosted: Wed Sep 14, 2011 6:19 am    Post subject: Re: Manual Intervention frequency Reply with quote

Jedi Knight

Joined: 22 Mar 2010
Posts: 4941
Location: Bloomington, IL USA

Vitor wrote:
lancelotlinc wrote:
Not so equal to z/OS (due to the high cost of *everything* on that platform).


And the decreased risk of downtime on z/OS.


z/OS has stellar uptime. The expense to the business to attain that is multiples of other platforms. In the 1970s, z/OS was the only solution many times. In the 21st century, lower cost solutions provide similar up time experience.

Vitor wrote:
So you're happy WMQ clustering is not trouble on Windows? For clarity?


WMQ product-to-product style clustering provides a better experience than clustering OS-to-OS.

Vitor wrote:
And your assertion that "Clustering and Windows are not a good combination" applies only to an active/passive HA cluster?


My comment was directed at the OS-to-OS style of clustering, and I asked the OP to clarify which one he is speaking of.

Vitor wrote:
What's your recommendation for implementing active/active (or active/active/active) on Windows?


The same as for any other platform. Business value is delivered when systems are up and functioning. How best to achieve that? Co-hosting a Unix or Windows instance is cheap, like a few dollars a day for commercial-grade co-hosting. I like putting WMB instances in Chicago, Atlanta, Croydon UK, Cambridge UK, Cebu Philippines. Use WMQ channel priority to route between the instances. solidDB backed by DB2. For the same Euro cost or better, you get value-added geographic dispersion, and fault-tolerance.
_________________
http://leanpub.com/IIB_Tips_and_Tricks
Save $20: Coupon Code: MQSERIES_READER
Back to top
View user's profile Send private message Send e-mail
Vitor
PostPosted: Wed Sep 14, 2011 6:33 am    Post subject: Re: Manual Intervention frequency Reply with quote

Grand High Poobah

Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA

lancelotlinc wrote:
Co-hosting a Unix or Windows instance is cheap, like a few dollars a day for commercial-grade co-hosting.


But when co-hosting is not acceptable to the business and/or the software doesn't run on Unix?

lancelotlinc wrote:
Use WMQ channel priority to route between the instances.


If the active/active are linked only by WMQ, how do you avoid the "stuck message" problem?
_________________
Honesty is the best policy.
Insanity is the best defence.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic  Reply to topic Goto page 1, 2, 3, 4  Next Page 1 of 4

MQSeries.net Forum Index » WebSphere Message Broker (ACE) Support » Manual Intervention frequency
Jump to:  



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
Protected by Anti-Spam ACP
 
 


Theme by Dustin Baccetti
Powered by phpBB © 2001, 2002 phpBB Group

Copyright © MQSeries.net. All rights reserved.