|
Re: Pytroll/SatPy for EUMETCast
Ferdinand Valk
Maurizio,
I ran your Sentinel-3A.py with my own file paths and it works as advertised. As you do see the folders appearing in the tmpdir but no image, the only thing I can imagine is that you point the overlay line and the Logo lines to the wrong location or that you didn’t add the required files there. If you run with the debug function and attentively sift through the produced debug info it should be clear what is going on.
Cheers, Ferdinand
From: MSG-1@groups.io On Behalf Of maurizio.calvitti@...
Sent: Thursday, 17 December, 2020 08:39 To: MSG-1@groups.io Subject: Re: [MSG-1] Pytroll/SatPy for EUMETCast
Ernst, I checked point by point, so much so that I noticed an oversight in the guide you linked : ETCdata is correct? You mean EMCdata Unfortunately I still have no images in the dedicated directories (ECMdata / images / Sentinel-3A or 3B), but only decompressed files in EMCdata/tmpdir xsen3a and x3senb. It should be clear than with this mail, I do not dare to ask for ready solutions, but only the confirmation that on your PC (I am referring to those who use SatPy on Windows) the attached shared S3 scripts (without any other further amendment) is working.
My best regards, Maurizio
Da: MSG-1@groups.io <MSG-1@groups.io> Per conto di Ernst Lobsiger via groups.io
Maurizio,
|
|
|
|
Re: Pytroll/SatPy for EUMETCast
Ernst Lobsiger
Mauriziao,
It's always EMCtools and EMCdata (not ETCdata nor ECMdata). Maybe I typod it somewhere in a post or *.txt as well. 1) These files work under Windows, Ferdinand has just recently published images on MSG-1. 2) Do you have at least a S3A.png or S3B.png in the respective tmpdirs??? 2a) If not check the environment variables part of the installation HOWTO. At the prompt command " echo %PPP_CONFIG_DIR%" should not be empty! 2b) If yes check that you have made the subdirectory ...EMCdata\logos and copied the two image files from the Sentinel-3 Update there. Regards, Ernst
|
|
|
|
Re: Pytroll/SatPy for EUMETCast
maurizio.calvitti@...
Ernst, I checked point by point, so much so that I noticed an oversight in the guide you linked : ETCdata is correct? You mean EMCdata Unfortunately I still have no images in the dedicated directories (ECMdata / images / Sentinel-3A or 3B), but only decompressed files in EMCdata/tmpdir xsen3a and x3senb. It should be clear than with this mail, I do not dare to ask for ready solutions, but only the confirmation that on your PC (I am referring to those who use SatPy on Windows) the attached shared S3 scripts (without any other further amendment) is working.
My best regards, Maurizio
Da: MSG-1@groups.io <MSG-1@groups.io> Per conto di Ernst Lobsiger via groups.io
Maurizio,
|
|
|
|
Re: experiences and lessons learned with a fresh Eumetcast install for WIndows and Linux
George Sz
Ernst,
I did take care of smcroutectl kill and put phyint disable entries into smcroute.conf, essentially accomplishing what you did. As for dvb-eumetcast stop, I seldom issue that command. I mean, why would I? Regards, George
|
|
|
|
Re: experiences and lessons learned with a fresh Eumetcast install for WIndows and Linux
Ernst Lobsiger
On Wed, Dec 16, 2020 at 01:54 PM, George Sz wrote:
everything works fine so far.George, if you run the service with EUMETSAT's script "dvb-eumetcast" chances are that "dvb-eumetcast stop" does not work. It should deconfigure all the multicast routes, kill szap and the daemon smcroute. But now the daemon is smcroute(d!!). I also use an /etc/tc-smcroute.cfg file where I enable only my DVB-Interfaces with lines like "phyint dvb0_0 enable" and start the daemon with the command "smcrouted -c 0 -N -f /etc/tc-smcroute.cfg > /dev/null 2>&1". Without the -N switch the multicast router wants to take care of all interfaces it finds. I do not want that for my network card. I kill the daemon with "smcroutectl kill > /dev/null 2>&1". The new version has a well written manual page ... Regards, Ernst
|
|
|
|
Re: experiences and lessons learned with a fresh Eumetcast install for WIndows and Linux
George Sz
Ernst,
I remember reading through your message when you posted it but I completely forgot about the -c part. I was aware that the smcroute command is merely a wrapper script now, however. I did not split traffic to dummy0/1/2, but regardless, everything works fine so far. Regards, George
|
|
|
|
Re: Giant iceberg A-68A is spinning.
toggle quoted messageShow quoted text
On Wed, Dec 16, 2020 at 05:34 PM, <geojohnt@...> wrote: Hello Ernst,
|
|
|
|
Re: Giant iceberg A-68A is spinning.
geojohnt@...
Hello Ernst,
Nice collection of animations.
Only the pointed tip of the iceberg still pointing down is showing in cloud today but it looks as Ferdinand said with his diagram of the coastal shelf around South Georgia that it may well move away from South Georgia.
Interesting.
Regards,
John. +++++++++++++++++++++++++++++++
-----Original Message-----
From: Ernst Lobsiger via groups.io <ernst.lobsiger@...> To: MSG-1@groups.io Sent: Tue, 15 Dec 2020 19:58 Subject: Re: [MSG-1] Giant iceberg A-68A is spinning. John,
the iceberg first turned CCW a little and is now spinning CW. I have put six videos on receiver Io (Pytroll/Satpy + ffmpeg). All *.ogv run in browsers like Edge and FireFox as players. http://5.153.116.236:81 Cheers, Ernst
|
|
|
|
Re: Pytroll/SatPy for EUMETCast
Ernst Lobsiger
|
|
|
|
Re: experiences and lessons learned with a fresh Eumetcast install for WIndows and Linux
Ernst Lobsiger
Markus,
this is not new (see message 29108). Good old smcroute has been split in two parts. EUMETSAT has not noticed that either. I ran into this with Debian 10 and Devuan 3.0. It has nothing to do with the kernel. Good news: More than one daemon can run. smcroute is only a wraper script for old installs. You might run into problems when stopping the EUMETcast service as well. Regards, Ernst
|
|
|
|
Re: Pytroll/SatPy for EUMETCast
maurizio.calvitti@...
Good Morning to all, In particular to SATPY users and developers, after reading and learning,
Now, I’m able to “SatPy” process all the Geostationary and Polar Satellites (Suomi_NPP, NOAA_20, Metop, Aqua..etc).
For what it concerns the testing of Sentinel-3 OLCI, once launched the python files I see the OLCI files processed, but I don’t produce any saved images nor .png In tmpdir. Dind’t I read enough? Maybe!
Regards, Maurizio
Da: maurizio.calvitti@... <maurizio.calvitti@...>
Da: MSG-1@groups.io <MSG-1@groups.io> Per conto di Ernst Lobsiger via groups.io
On Sat, Dec 5, 2020 at 11:55 AM, Graham Woolf wrote:
O.K.
|
|
|
|
Pytrolll/SatPy for EUMETCast
Roland Ertl
Ernst,
followed this thread HOWTO install and use Pytroll/SatPy for EUMETCast under Win 10 64-bit,i managed to produce MSG-4/Himawari and NOAA-20 composite image. Thank you for the beginner instructions.
Regards, Roland DK1NJ
|
|
|
|
Re: experiences and lessons learned with a fresh Eumetcast install for WIndows and Linux
George Sz
I upgraded my Eumetcast box to Ubuntu 20.04.1 LTS. As expected, it did not go without issues. After a manual update to kernel 5.9 (I did this before 5.10 was released) and recompiling TBS drivers, everything seemed to be going fine. Or so I thought.
I noticed that after a while, Tellicast started to throw "no data received" errors and indeed, no data was being received at all. I modded dvb-eumetcast according to Markus' entry [link] but that did not help, unfortunately. After digging through the documentation of smcroute and some trial and error, I managed to fix the issue by applying smcrouted -c 0 on startup. Per smcroute documentation: -c Flush unused dynamic (*,G) multicast routes every SEC seconds ... Default is 60 sec, set to 0 to disable.So I suspect that there is a bug somewhere in smcroute that disconnects multicast routes when a dynamic route flushing occurs. Disabling this behavior fixes the issue. I'm running smcroute version 2.4.2. I'm aware that there is a newer version available but I did not manage to manually install it correctly yet. For now I'm taking the "if it ain't broke, don't fix it" route and leave it as is.
|
|
|
|
Re: Giant iceberg A-68A is spinning.
Christian Peters
Ernst,
toggle quoted messageShow quoted text
really nice videos! It’s turning a lot now...! :-O Regards, Christian
|
|
|
|
Re: Giant iceberg A-68A is spinning.
Ernst Lobsiger
John,
the iceberg first turned CCW a little and is now spinning CW. I have put six videos on receiver Io (Pytroll/Satpy + ffmpeg). All *.ogv run in browsers like Edge and FireFox as players. http://5.153.116.236:81 Cheers, Ernst
|
|
|
|
Re: Pytroll/SatPy for EUMETCast..Help
Ernst Lobsiger
On Tue, Dec 15, 2020 at 10:29 AM, <rnosborne@...> wrote:
There is still a lot to learn but at least I am on my way and thanks for the beginner instructions.Richard, welcome to the Club. Stay with GNU/Linux whenever possible! I do all my development there. Much more versatile and performant. You find a very old HOWTO for GNU/Linux below. It has never got much attention and is certainly deprecated. Only with the Windows HOWTO V1.0 (one good thing out of the first Corona Lockdown) I tried to make things more simple and less technical. Maybe you still find a couple of ideas below. When I will issue a V2.0 for Windows towards the end of 2020 I will also publish all my scripts for GNU/Linux. All this is also based on Miniconda3 for a better synchronization with Windows 10 even if we have other possibilities under GNU/Linux. Stay tuned, Ernst https://groups.io/g/MSG-1/message/28324
|
|
|
|
Re: Pytroll/SatPy for EUMETCast..Help
rnosborne@...
Ernst,
Having followed this thread, I decided that I would have a go at Pytroll/Satpy. As I already had a Conda installation on a Linux Mint machine, I elected to try it there and eschew Windows for once. After following your HOWTO instructions very carefully, I managed to produce an MSG-4 composite image reasonably quickly so I now have a working installation. There is still a lot to learn but at least I am on my way and thanks for the beginner instructions. Regards, Richard
|
|
|
|
Course of A68
Ferdinand Valk
I attach an image of today with the sea floor profile next to it. It has all
the looks of it that it is following this profile with the local current, maybe thus avoiding stranding after all?
|
|
|
|
Re: A68
Ernst Lobsiger
On Tue, Dec 15, 2020 at 06:35 AM, Graham Woolf wrote:
Thanks - I have managed to do that :-)Graham, scripting is certainly possible (though out of range for most users here). You might remember that the next things a script would check is for Sentinel-1A Decemebr 19th and for Sentinel-1B December 21th. But A68a is a moving target and considering its speed also might slip out of an automated fixed search window soon. Makes scripting not easy ... Regards, Ernst
|
|
|
|
Re: A68
Graham Woolf
Hi Ernst
Thanks - I have managed to do that :-) Its just not the level of automation that I was looking for All my scripts run on a schedule in Task Manager so if Im not at the computer I still get images I was hoping to do the same with Sentinel 1 data but it seems not Regards Graham
|
|
|
