Please read the posts in the Announcements section about the current status of Cumulus development now that I have retired

Please read this post before posting

Latest Cumulus release v1.9.4 (build 1099) - Nov 28 2014
Latest Cumulus MX release - v3.0.0 build 3043 Jan 20 2017. See this post for download

Cumulus startup anomalies on my suspect hardware

Discussion specific to Davis weather stations
archae86
Posts: 68
Joined: Sat May 21, 2011 4:46 pm
Weather Station: Davis Wireless Vantage Pro2
Operating System: Windows 7
Location: Albuquerque, NM, USA

Re: MX better regarding Davis reception

Postby archae86 » Mon Feb 20, 2017 5:11 pm

steve wrote: very many people are using MX for their production system. It appears to be as reliable as Cumulus 1
...
The main decision is whether you are happy with the different architecture of MX, and whether you can live without those facilities in Cumulus 1 yet to be added to MX.

That adjusts my understanding, and I am adjusting my plans. In the very short term, I'll keep running the old hardware including the old PC and C1 as my "production" environment, while using the new hardware for further testing, and preparation for cutover.

In the realm of testing, I intend over the next couple of days to do some 24-hour temperature offset comparisons, as a partial exploration of my observation that the console seems to report indoor temperature elevated above what it would otherwise be, not just as a spike when one or another of my abnormal events occurs, but also as a steady-state difference between the case that, for example C1 is running, versus nothing, or MX, or WeatherLink.

I also intend soon to do a trial transfer of the 5-year history on the old PC to new, as to which I believe there is a detailed entry in the FAQ, with test posting using C1 to a temporary trial site.

Following up on your comment here, I intend to explore the trial MX installation I made here by a rather minimalist method, seeking problems or opportunities to fix or pursue before making that my production site. My two main known issues with my trial installation are that I lack the ability to view history data graphically locally (the old Select-a-graph function), and that the trial web site I generated lacked updated graphs (the Trends page). The thread posted by mcrossley may point me a path to a wonderful solution to both problems, but may be beyond my reach. Clearly I need to spend time looking about in the abundant MX postings.
My cumulus generated weather site is at http://pastoll.info/weather

CorkHarbourWx
Posts: 6
Joined: Thu Oct 12, 2017 5:21 am
Weather Station: Davis Vantage Pro2
Operating System: Windows 10

Re: Cumulus startup anomalies on my suspect hardware

Postby CorkHarbourWx » Thu Oct 12, 2017 10:57 am

Hi,

I'm in the process of upgrading my weather setup and recently migrated my weather monitoring activities to a Windows 10 machine. I also made the decision to try Cumulus 1 (I've been using WeeWx for many years and Weather Display before that). Unfortunately, I'm experiencing significant issues with the Davis VP2 "console to R mode" on startup with Cumulus 1 as is described in this thread. I'm running the most recent C1 build with a serial (via serial to USB adapter) data logger. I've tried two separate VP2 consoles (one circa 2006 vintage, the other circa 2011) and the issue is arising with both.

Every time Cumulus starts, the console kicks into R mode and only sometimes does the setup revert to normal operation after archived data is downloaded. The archived data is downloaded OK on every occasion, however the console sometimes remains in R mode thereafter (eventually going to "L" mode). Even on those occasions when it does revert to normal "X" mode, Cumulus doesn't always start updating - sometimes it does, sometimes it doesn't.

RF reception is not a factor; the setup is in situ since 2006, has been extremely reliable, is regularly maintained, since this issue arose I've done checks to identify any interference issues (all is well) and the console RF reception diagnostics continue to indicate typically >98% successful packet reception. The issue here is one only of startup communication between the console and Cumulus 1. When the system is settled and operating, it's really reliable (which makes this startup wrinkle all the more frustrating).

I've observed that the problem is more likely to occur when there is less archived data to download from the console - on occasions when there's plenty of archived data on the console to transfer across to Cumulus, (in the hundreds of readings), then the above issue doesn't appear to arise. It happens when, for example I'm forced to do a reboot by Windows 10 for updates - only a few archived readings are stored in the console during the reboot and when Cumulus starts up again after the reboot, I invariably end up with problems. I'm not sure whether that's a red herring but I thought it worth mentioning. Other than forced reboots and other exceptional situations, I run my weather software 24x7x365 for near real time website and 3rd party weather feed updates so waiting to restart Cumulus until a sufficient number of archived readings have been collected so the startup problem is sidestepped isn't really a viable solution (though it's a workaround I've been using).

When this startup problem arises, I do eventually get the setup to settle down again but it has regularly taken me over an hour to do so each time it's happened and seems to be an entirely hit & miss affair. Sometimes restarting Cumulus will eventually fix things (often several restarts are needed, waiting >10 minutes each time to see if the setup sorts itself out i.e. the console goes from "R" to "X" and then if Cumulus starts updating), sometimes forcing the console to go through RF signal acquisition and restarting Cumulus is needed. On occasion, rebooting the Windows 10 machine has also been needed. I'm a bit stumped as to why but the reboot makes me suspect the Davis dll may be at the root of this (could it be an issue with the dll under Windows 10?).

Whatever is at the heart of it, I'd appreciate any suggestions of anything I can try as other than trying Cumulus MX (which I'll do this weekend), I'm pretty much out of ideas on this one.

One specific question to finish; would someone remind me how I check/change the VP2 data logger logging frequency. I'm pretty sure it's set to 1 minute and is therefore the same as the Cumulus update rate but I want to make absolutely sure. It's ages since I've had to tinker with that side of things so would appreciate a reminder (I've just been through the Weatherlink manual and didn't see anything relevant, apologies if I missed it).

BR
Cormac
[Cork Harbour / UTC+1] NNNN++++

User avatar
steve
Cumulus Author
Posts: 26311
Joined: Mon Jun 02, 2008 6:49 pm
Weather Station: None
Operating System: None
Location: On tour in Portugal
Contact:

Re: Cumulus startup anomalies on my suspect hardware

Postby steve » Thu Oct 12, 2017 11:23 am

I can’t be much help, other than to repeat what I said previously that the console isn’t good at doing more than one thing at a time, and logger download is one thing that occupies it in such a way that it can appear to freeze temporarily, and I guess losing contact with the ISS may be a result. This is probably more so with the way the Davis DLL which C1 uses works (it asks for one record at a time). I can’t guarantee that you wouldn’t see the same thing with MX.

You do need to check that you have the archive intervals set the same, and also that the PC and console clocks are the same. You can check the console interval (and reset if necessary) using Weatherlink. Note also that a quick restart can also cause all archive entries to be downloaded. You can check whether you are experiencing the “all archive entries downloaded” issue (for whatever reason) by looking near the start of the diags file.
Steve
-----
Hosting available for Cumulus web sites. See http://sandaysoft.com/forum/viewtopic.php?f=2&t=11876

Please read the posts in the Announcements section about the current status of Cumulus development since I have retired from my day job

archae86
Posts: 68
Joined: Sat May 21, 2011 4:46 pm
Weather Station: Davis Wireless Vantage Pro2
Operating System: Windows 7
Location: Albuquerque, NM, USA

Re: Cumulus startup anomalies on my suspect hardware

Postby archae86 » Thu Oct 12, 2017 1:06 pm

CorkHarbourWx wrote:Whatever is at the heart of it, I'd appreciate any suggestions of anything I can try as other than trying Cumulus MX
[Cork Harbour / UTC+1] NNNN++++

I shall comment here as thread starter. I found that in my case activation of the Cumulus function which adjusts console clock to agree with the PC was directly associated with undesired transition to the "R" state on startup, and also at a repeatable time in the early morning.

As the drift rate of the clock in my console was pretty slow, I decided that the added value of updating the clock so frequently was outweighed by the considerable loss of hours being spent in the R state.

The setting in question is available at Configuration|Station|Settings|Synchronise station clock

In my case I also found that two different samples of console differed appreciably in their likelihood of quickly resuming X state after being forced to R.

I'm afraid my stated intention of forging forward to MX continues to be a future aspiration.

Several important details of your experience differ from mine, so I am not confident what you would see if you disabled the station clock update, but I suspect it would help your primary problem.
My cumulus generated weather site is at http://pastoll.info/weather

CorkHarbourWx
Posts: 6
Joined: Thu Oct 12, 2017 5:21 am
Weather Station: Davis Vantage Pro2
Operating System: Windows 10

Re: Cumulus startup anomalies on my suspect hardware

Postby CorkHarbourWx » Thu Oct 12, 2017 6:48 pm

Folks,

Thanks for comments / observations / suggestions. I've confirmed the 1 minute data logger time interval via Weatherlink. In Cumulus I then made but one change; I unchecked the "Synchronise station clock" option under Configuration->Station->Settings. Since unchecking that option, I've repeatedly stopped, waited (for differing time intervals from circa 10 seconds to 5 minutes) and then restarted Cumulus. Restarting Cumulus following this one change has resulted in the "R" mode issue not occurring on start up. Suffice it to say, I'm rather pleased by this development ;)

My sincere thanks and apologies for not trying it before posting initially; I was not sufficiently thorough in my reading of earlier posts where it was highlighted as a possible fix.

...and back to the important business of monitoring the weather:

"
Weather Advisory for Ireland
Issued: Thursday 12 October 2017 14:00
A combination of a vigorous Atlantic weather system and the remnants of Hurricane Ophelia will pass close to Ireland on Monday, and has the potential to be a high-impact event in parts of the country. There is a lot of uncertainty as to the exact evolution and movement of this weather system during the coming four days, but storm-force winds, outbreaks of heavy rain, and very high seas are threatened. Met Eireann will maintain a close watch on the evolution and issue further advisories and warnings as these are warranted.
"

HickoryGroveWX
Posts: 1
Joined: Fri Sep 22, 2017 10:06 pm
Weather Station: Davis 6163 Vantage Pro2+
Operating System: Win7 64bit SP1 Home Premium
Location: Kenly, NC USA

Re: Cumulus startup anomalies on my suspect hardware

Postby HickoryGroveWX » Sun Oct 15, 2017 7:32 pm

I'm a fairly new user to Cumulus v1.9.4 ( about a Month). My Davis 6163 is just a couple months old. I have W/LUSB running in Serial Mode and set the Synchronize Station Clock in Station Configuration. My Weather Computer is a 8 year old DELL Laptop with win7 Home Premium 64 bit SP1 uploading data to WU.

I just wanted to add is I never noticed the Console go to "R" after I D/L the data into W/L until I read this thread. When I initialize Cumulus and after the Load Data is complete, I do get the "R" on the Console. I get the "X" after about 50 seconds. Steve is right about the Console can only do one thing at a time as even the clock stops about the length of time that the Console is in "R" mode. Other than that little niggle, Cumulus and WeatherLink has rock solid.


Return to “Davis VP/VP2/Vue”

Who is online

Users browsing this forum: No registered users and 4 guests