This is part of an occasional series of small blog posts where I (Morag) will write about some of the quirks in IBM MQ, in the hope of invoking the response, "Well, I didn't know that!" Read other posts in this series.
On both queues and channels there is a monitoring switch, MONQ and MONCHL, which both serve the same purpose. They allow you to turn on some monitoring capabilities that collect various timestamps - which might be expensive and so are not on by default. The table below summarises what these are and you can read more about them in Knowledge Center.
These attributes were introduced in WebSphere MQ V6, and in that release, both MONQ and MONCHL had HIGH, MEDIUM and LOW settings that caused a different level of data collection. The table below that you can also see in Knowledge Center that now only applies to MONCHL, applied to both in MQ V6.
|Low||Measure a small sample of the data, at regular intervals.||For objects that process a high volume of messages.||1 in 64|
|Medium||Measure a small sample of the data, at regular intervals.||For most objects.||1 in 8|
|High||Measure all data, at regular intervals.||For objects that process only a few messages per second, on which the most current information is important.||every|
By MQ V7.0.0, this table only applies to MONCHL and all three levels are the same for the data collection for MONQ.
Footnote: Checked also in a downloadable MQ V7.0.0 Information Center.
In addition, even back in MQ V6, the MONCHL attribute on z/OS was treated the same for HIGH, MEDIUM and LOW due to the implementation of all the time-stamp gathering resulting in one less timestamp than was there previously and so there was no additional cost over and above the previous release.
So although the MQ that you know now has three levels of collection, in many cases these different levels amount to exactly the same behaviour!