Random spikes

I would guess that the spikes I keep sending out are not real, so a question to you long time rad folks. My GMC-200 unit has been outside, running for some number of years now, can the M4011 tube eventually go bad? what is the lifetime expectancy of these kinds of tubes? Other issue could be just my computer hanging up causing the spikes now and then. There are times when I go to open the radlog app full screen that it hangs open to a full white screen, then the rest of the image shows up. I do have 2 other programs running as well, retrieving data every minute, then ftp'ing it up to my page. Question still, how long do these tubes last on average? Frank

Hi Frank,
Modern Halogen-quenched tubes should last quite a long time, based on some brief googling. Some references say infinite, some guess at 10^10 events or 634 years at 30 CPM, The manufacturer Soeks claims 5-8 years for theirs, likely assuming a bit of rough handling and wanting to limit expectations. After an initial apparently leaking tube, my M4011 has been running almost 4 years now.

Looking closely at your 24 hour data, it appears that the submitted values stopped about 12 hours before the spike appeared. That points to a possible stability glitch. I was about to suggest looking at the Radlog 'Log and Analysis' tab, at the Message log latest (bottom) entries for a clue, but I see the last few days have many submission errors in addition to the normal stuff, so it may be hard to find a unique error message. If the PC is otherwise running OK, I wouldn't worry about multiple programs.

My graphs look fine so this is not related to Frank's issue. The timeout errors in the log started on Jan 6. Here is my log for the last 7 hours or so. The burst of entries at 12:30 AM looks like a normal nightly cleanup to me.

10/01/2019 7:40:43 PM : Successfully updated shared data viewer
10/01/2019 7:44:27 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 7:44:33 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 7:50:28 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 8:11:33 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 8:28:44 PM : Successfully updated shared data viewer
10/01/2019 8:37:28 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 8:40:43 PM : Successfully updated shared data viewer
10/01/2019 9:05:29 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 9:17:34 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 9:25:34 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 9:28:44 PM : Successfully updated shared data viewer
10/01/2019 9:31:34 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 9:40:44 PM : Successfully updated shared data viewer
10/01/2019 10:28:44 PM : Successfully updated shared data viewer
10/01/2019 10:40:44 PM : Successfully updated shared data viewer
10/01/2019 10:44:35 PM : Error submitting CPM value to radmon.org : The operation has timed out
10/01/2019 11:28:44 PM : Successfully updated shared data viewer
10/01/2019 11:40:44 PM : Successfully updated shared data viewer
11/01/2019 12:09:36 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 12:25:36 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 12:28:44 AM : Successfully updated shared data viewer
11/01/2019 12:30:08 AM : Creating new log datapoint
11/01/2019 12:30:08 AM : Creating new log datapoint \RadLog 20190111 003008 :
11/01/2019 12:30:08 AM : Created new datapoint directory.
11/01/2019 12:30:08 AM : Saved graph data.
11/01/2019 12:30:08 AM : Saved chart image.
11/01/2019 12:30:08 AM : Exported interval log.
11/01/2019 12:30:08 AM : Saved distribution plot data.
11/01/2019 12:30:09 AM : Updated long term trend graph
11/01/2019 12:30:09 AM : Saved trend report
11/01/2019 12:30:09 AM : New log datapoint complete.
11/01/2019 12:30:09 AM : Attempting (S)FTP upload of historical log data...
11/01/2019 12:30:09 AM : Uploading shared datapoint...
11/01/2019 12:30:09 AM : Error in FTP transfer: Invalid URI: The hostname could not be parsed.
11/01/2019 12:30:09 AM : Error in FTP transfer: Invalid URI: The hostname could not be parsed.
11/01/2019 12:30:09 AM : Error in FTP transfer: Invalid URI: The hostname could not be parsed.
11/01/2019 12:30:09 AM : Error in FTP transfer: Invalid URI: The hostname could not be parsed.
11/01/2019 12:30:09 AM : (S)FTP upload of historical log data completed.
11/01/2019 12:30:09 AM : Checking for valid user...
11/01/2019 12:30:10 AM : Valid user/password.
11/01/2019 12:30:13 AM : Creating new datapoint structure on shared data:
11/01/2019 12:30:13 AM : Creating datapoint folder...
11/01/2019 12:30:13 AM : Reading and updating datapoint file...
11/01/2019 12:30:16 AM : Transferred graph data.
11/01/2019 12:30:16 AM : Transferred distribution plot data.
11/01/2019 12:30:17 AM : Transferred trend report.
11/01/2019 12:30:17 AM : Transferred notes.
11/01/2019 12:30:18 AM : Transferred alertsandwarnings.txt
11/01/2019 12:30:18 AM : Transferred coincidences.txt
11/01/2019 12:30:19 AM : Transferred stationsynopsis.txt
11/01/2019 12:30:19 AM : Transferred stationinfo.txt
11/01/2019 12:30:20 AM : Transferred trendreportsummary.txt
11/01/2019 12:30:21 AM : Transferred longtermtrenddata.xml
11/01/2019 12:30:21 AM : Creation of datapoint on shared complete.
11/01/2019 12:31:31 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 12:40:44 AM : Successfully updated shared data viewer
11/01/2019 12:47:32 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 12:48:32 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 12:55:32 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 1:00:37 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 1:04:37 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 1:28:44 AM : Successfully updated shared data viewer
11/01/2019 1:40:44 AM : Successfully updated shared data viewer
11/01/2019 1:43:32 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 1:45:32 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 2:03:38 AM : Error submitting CPM value to radmon.org : The operation has timed out
11/01/2019 2:08:33 AM : Error submitting CPM value to radmon.org : The operation has timed out

My GMC-300 with the 4011 tube has been going well for several years. If you have an internal NiCd battery in the unit remove it and or replace it. Mine is going out and will cause the system to barely stay on between power glitches or shutdown events.

Other things to look out for... has your communication been good till this point or has it always been spotty? I run mine off a Raspberry Pi and I found that the communication is anything but stable. I had to modify the code to make it work with the GMC unit. Are you using the the Pyradmon software? If you want I can share my modifications with you. It's quite stable in many years of running now.

The root cause is the GMC units in the early days were poorly written firmware. There is blocking code that interferes with the RS232 communication so it will not respond properly to commands. At least that is the case in the 300 series and I suspect it's in the 200 as well.

One other thing - do you periodically clean up your logs? Mine can get quite large and I typically will run a truncate command on it to trim it down to the last few entries. Just try to avoid running out of data space due to excessive log size.

Geez, all that typing and it timed out here. Dan, really no errors, just pretty good spikes now and then. Yes, you'll see a gap in data just prior to the spike, its the system hung up and then finally running again. 3 main programs running as far as memory is concerned. GRL3 @ 117k, Radlog @ 25K, and Cumulus @ 22k. All 3 uploading data at various times, 24 hours a day. I did clear the logs and it hung up and you can see that in my chart as of this posting, checked the memory usage again while it was happening and radlog was still around 25k, so no real change there. CPU usage @ 30%, memory @ 38%. When it was happening, those 3 programs were still around the same usage, but CPU was up to 90% usage now, then dropped back to ~35%. The unit out there is feeding 2 sites and other site doesn't show the spike, but that program is running on this PC I am writing on now.