portal entry

select a category, or use search below
(searches all categories and all time range)
Title:

Running the CF 2018 PMT? Have you manually applied the recent update to it?

| View in Portal
March 08, 2019 06:42:35 AM GMT
9 Comments
<p>If you have not done the needed manual update to the PMT, you are missing out on important bug fixes and feature enhancements. </p>
<p>The post <a rel="nofollow" href="https://coldfusion.adobe.com/2019/03/running-cf-2018-pmt-manually-applied-recent-update/">Running the CF 2018 PMT? Have you manually applied the recent update to it?</a> appeared first on <a rel="nofollow" href="https://coldfusion.adobe.com">ColdFusion</a>.</p>
Labels: Blog, CF2018 Updates, ColdFusion 2018, blog, cf2018 updates, ColdFusion, coldfusion 2018

Comments:

Thank you for the detailed post Charlie! I just applied the update.
Comment by Giancarlo Gomez
1904 | March 09, 2019 01:18:40 PM GMT
The Performance Monitoring Toolset has a refreshed installer available here from February 2019: <a href="https://www.adobe.com/support/coldfusion/downloads.html" rel="nofollow">https://www.adobe.com/support/coldfusion/downloads.html</a> I don't know if this includes PMT Update 1 or not. Note that if you go to the main download link for CF 2018 (http://www.adobe.com/go/trycoldfusion) it still offers the old PMT installer.
Comment by Legorol San
1906 | March 09, 2019 02:52:29 PM GMT
<a href="https://coldfusion.adobe.com/profile/Giancarlo Gomez">Giancarlo Gomez</a>  Hi JC. Charlie, I searching to see if there are any know issues with Windows 2016 and came across your post here. On a fresh AWS ec2 instance, Windows 2016, CF 2018 Update 3 applied, CF 2018 Lockdown applied. installed the toolset and the services refused to start. The toolset logs seem  to indicate a connection issue with the datastore and I thought it might be a timing issue as well. But I applied the PMT update and the issue remains, services will not start regardless of the order I start them in. I did check the "known issues" for this release which does not apply to my situation. Thanks, Dan
Comment by Dan OKeefe
2068 | May 27, 2019 04:35:16 AM GMT
Dan, did you assess what I discuss in the section here, "Always check the update logs"? And did you see the first bullet in the known issues section? Finally, if those don't get you going, see the help resources I offer at the bottom of that section on checking the logs.
Comment by Charlie Arehart
2069 | May 27, 2019 05:45:40 AM GMT
A little bit old on the thread, but has anyone using PM experiencing issues with the datastore (Elastic-search) process consuming most CPU cycles?  Is there a way to tame that process to lower burden on CPU?
Comment by obxBillSurf
2150 | July 09, 2019 08:56:08 PM GMT
Dan?
Comment by Charlie Arehart
2153 | July 10, 2019 10:32:57 PM GMT
<p>Legorol, yes, the newer installer is indeed newer and larger than the original, and includes the update. Thanks for pointing that out. I'll update the post to clarify that this applies to the original installer. </p><p>One can tell which is which in Windows at least by right-clicking the installer exe, choosing properties, and viewing the “file version” field on the “details” tab. It will report 2018.0.0 for the original installer, and 2018.0.2 for the refreshed one with the update from Feb.</p>
Comment by Charlie Arehart
2154 | July 10, 2019 10:42:05 PM GMT
<p>There can be many reasons for that. I’d start with whether the logs (in the elasticsearch folder) tell you anything. Then, is the box on which it’s running perhaps under-powered? Is it running on the same box as the PMT? or worse, the same box as CF? I can say that I have seen boxes buckle under the weight of running all 3.</p><p>And even if it’s broken off on its own server, it could still be underpowered. Then it’s just a process of managing/troubleshooting it like any elasticsearch process.</p><p>For instance, it could be that the heap settings for the PMT or datastore are set to use too much heap (for the box they are running on), or too little (for the work they are being asked to do). There are config files for setting the XMX for both, but it's beyond the cope of this post to get into those details. </p><p>Or it could be that someone has configured features in the PMT that are causing a lot of data to be tracked. You may want to assess those configurable features. (I’m not sure of a way to observe those via any config file.)</p><p>Finally, it could just be that your CF site (or sites or servers) being monitored by the PMT just have high volume of traffic on the site(s), which leads to lots of data being tracked in the datastore.</p><p>Just some thoughts to start you on the process.</p><p>I will add, FWIW, that you can monitor both the PMT service itself and Elasticsearch with FusionReactor. While most know it as a monitor for CF, it is in fact a monitor for any Java app, and the PMT and elastic both are. It’s not a built-in option within FR (takes some manual configuration) but should be doable, if you may want to dig in that way. (Let me know if you may want to pursue that under the guise of consulting, at carehart.org/consulting.)</p><p>But if something else above helps, great.</p>
Comment by Charlie Arehart
2155 | July 10, 2019 11:29:52 PM GMT
I've learned today that there is yet another bug in the PMT causing a negative impact on CF (indeed cfquery) processing. See the thread on the problem, and the reply from Priyank of Adobe pointing out that the problem seems due to the PMT, at <a href="https://forums.adobe.com/message/11219355#11219355" rel="nofollow">https://forums.adobe.com/message/11219355#11219355</a>. As of this writing, there is not a PMT update available to fix this, but we can expect there will be one made available eventually.
Comment by Charlie Arehart
2260 | August 20, 2019 01:57:11 AM GMT