Trig2disk problem

98 views
Skip to first unread message

cellectronic

unread,
Oct 14, 2022, 10:08:52 AM10/14/22
to Earthworm Community Forum
Hi all

I have EW7.10 installed on a RPi (OS Raspbian).
Carlsta and Carlsub are used along with Trig2disk.
I can see that if an event is declared by Carlsub it is observed in the Hypo-Ring,  I have Trig2disk looking at this ring to capture the events.
The problem is there are no events being recorded by Trig2disk and the log shows the below information....

*** processing new trigger message after (561 bytes) ***
v2.0 EVENT DETECTED     20221014 09:10:12.00 UTC EVENT ID: 7218 AUTHOR: 033014255

Sta/Cmp/Net/Loc   Date   Time                       start save       duration in sec.
---------------   ------ ---------------    -----------------------------------------
 FIR * GB 00 P 00000000 00:00:00.00 UTC save: 20221014 09:09:42.00      69
 WBK * GB 00 P 00000000 00:00:00.00 UTC save: 20221014 09:09:42.00      69
 WBK * GB 00 P 20221014 09:07:36.00 UTC  save: 20221014 09:09:42.00      69
 WBK * GB 00 P 20221014 09:07:36.00 UTC  save: 20221014 09:09:42.00      69

 WARNING: in event 7218 WBK * GB 00 was either multiply requested, or not found
 WARNING: in event 7218 WBK * GB 00 was either multiply requested, or not found
 problem retrieving FIR SHZ GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHZ GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHN GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHE GB 00 - SnippetMaker: wave server returned problem flag; continuing.
*** Done processing message (561 bytes) ***


I cant see what is causing this error and would be gratefull for some help.

Thanks in advance 


Paul Friberg

unread,
Oct 14, 2022, 10:21:54 AM10/14/22
to earthwo...@googlegroups.com

Trig2disk is not finding the data requested in the EW wave-server that you have it pointed at. This error:

SnippetMaker: wave server returned problem flag; continuing.

Also, you seem to have a subnet with  the same station multiply defined. That is not recommended, instead use one channel per station in a subnet in carlsubtrig.

Paul

--
--
You received this message because you are subscribed to the Google
Groups "Earthworm Community Forum" group.
 
To post to this group, send an email to earthwo...@googlegroups.com
 
To unsubscribe from this group, send an email to
earthworm_for...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/earthworm_forum?hl=en

---
You received this message because you are subscribed to the Google Groups "Earthworm Community Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to earthworm_for...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/earthworm_forum/716c8cf1-ceef-4e9b-be78-5425419b3ca2n%40googlegroups.com.


--
===================================
Paul Friberg   p.fr...@isti.com
CEO/Seismologist
ISTI==Instrumental Software Technologies, Inc.
Phone +1.518.602.0001  

Steve Shuff

unread,
Oct 16, 2022, 4:42:14 AM10/16/22
to earthwo...@googlegroups.com
Hi Paul,

Thanks for the info.
What is confusing is when looking at the trigger info all the components are marked with an "*" (asterix)
ie" WBK * GB 00" however  these are defined as SHZ SHN SHE in waveserver.d and the Carl triggering modules and also the trig_station file .
I have the heli2ew module running fine, it's only the triggering I am having problems with.

It is possible I have a misconfigured  waveserver ! I have attached this in case you spot anything obvious  .
 I setup the trig2disk module to create a Seisan format file, on detection of an event the file is created, However opening this file in Seisan reveals a flat
 trace with no data.
Any help would be much appreciated 

Thanks again. 



--
Steve Shuff.
wave_serverV.d

Paul Friberg

unread,
Oct 16, 2022, 8:04:11 AM10/16/22
to earthwo...@googlegroups.com

Glad to help Steve. We also need to see your trigstation, carlsubtrig and carlstatrig files. The * means that there is a setting in the carlsubtrig saying use wildcard for all stations and since you have multiple channels for the same station defined, it uses that setting.

Paul

Steve Shuff

unread,
Oct 16, 2022, 9:26:45 AM10/16/22
to earthwo...@googlegroups.com
Hi Paul,

I have attached the requested files.

Best regards



--
Steve Shuff.
carlsubtrig.d
seismet_trig.sta
trig2disk.d
carlstatrig.d
Message has been deleted

cellectronic

unread,
Oct 26, 2022, 5:05:43 AM10/26/22
to Earthworm Community Forum

Hello Paul,
I am not sure if you managed to look into the Trig2disk problem, No woorries if your busy.
The additional files you requested should all be available to you now,
If there is anything more I can do to help please shout out. Thanks again.

El Berto

unread,
Nov 2, 2022, 10:34:41 AM11/2/22
to Earthworm Community Forum
Are you sure to store enaugh data into waveserver?
I suppose trig2disk is requesting unavailable data: let's say trig2disk is requesting wave from 20221014 09:09:42.00, duration 60 seconds, but in waveserver you have data from 20221014 09:10:50.00 to 20221014 09:20:53.00.
Check using getmenu to see time window of available data.

Steve Shuff

unread,
Nov 2, 2022, 10:44:35 AM11/2/22
to earthwo...@googlegroups.com
Hi Berto,

Yes I used "Get menu" to check the quantity of data in Waveserver and it seemed to hold several hours of data .

I also used "Swarm" and I had data backfill from several hours previous , so I need to dig a bit deeper on this one .

The Edata2ew module you developed works really well, Thanks for helping out with that to.


   Steve 

Paul Friberg

unread,
Nov 2, 2022, 8:55:26 PM11/2/22
to earthwo...@googlegroups.com
Hi Steve,

Apologies for not getting back to you sooner. Can you get us your carlsubtrig logs for a trigger and also the result from a getmenu from the waveserver so we can see what is in the wave server. From your original errors:

WARNING: in event 7218 WBK * GB 00 was either multiply requested, or not found
 WARNING: in event 7218 WBK * GB 00 was either multiply requested, or not found
 problem retrieving FIR SHZ GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHZ GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHN GB 00 - SnippetMaker: wave server returned problem flag; continuing.
 problem retrieving WBK BHE GB 00 - SnippetMaker: wave server returned problem flag; continuing.
*** Done processing message (561 bytes) ***

It looks like there are no data in the waveserver at the time of request!! There is a DelayTime setting in the trig2disk.d that you can increase to a bit longer if your data are buffered from the data source. I see you have that set to 5 seconds, but perhaps the EarthData loggers buffer data longer and the trig2disk request to the wave server is still too quick....push the delay time to say 30s and see if the log messages go away.

The "multiply defined" error is because you have the WBK station listed in the subnet for multiple channels and you have the CompAsWild flag set in your carlsubtrig file which says request all chans for a given station....yet you have WBK listed for each chan in the subnet. This part is confusing, but we typically only include ONE chan from a station in the carlsubtrig subnet definition...otherwise one station can vote multiple times....

Best,

Paul

Steve Shuff

unread,
Nov 6, 2022, 12:37:46 PM11/6/22
to earthwo...@googlegroups.com
Hi Paul 
No worries , 
 I have been busy myself , Decorating , So most of my PC stuff has been under dust sheets for the last few days.

Anyway I have attached the requested files and hope they can help you .
The only one I was unsure of was the contents of the waveserver from the  "Getmenu" command,
I did "Getmenu" and copied the results into a text file.

I have increased the Delay time to 30 seconds as suggested and also removed the Compaswild statement.

Currently I am running the 3 sensors from WBK but I will soon be installing another station at a remote location so I will change my trigger sets to suit.
Also, I have attached an image of the EW map .
Many thanks for your help with this.

Best regards 



--
Steve Shuff.
carlsubtrig.d
carlstatrig.d
getmenu
carlsubtrig_20221106.log
seismet_trig.sta
wave_serverV.d
trig2disk.d
EW Map.jpg

cellectronic

unread,
Nov 21, 2022, 4:44:35 AM11/21/22
to Earthworm Community Forum
Hi All


As Paul suggested in a seperate correspondance "compiling EW7.9 on another Pi to see if it is a compile error"

The question now is before I go down that route , Has EW 7.9 been compiled on the RPI 3 ? 
Also, is there some info anywhere that suggests how to setup the ew_linux.bash for a successfull compilation.


Any info appreciated.

El Berto

unread,
Nov 21, 2022, 5:53:28 AM11/21/22
to Earthworm Community Forum
I'm trying to compile EW7.9 on Pi3, but I'm getting errors regardless ew_linux.bash settings

sudsputaway.c:451:32: error: ‘INT32_MIN’ undeclared (first use in this function); did you mean ‘INT_MIN’?
sudsputaway.c:453:38: error: ‘INT32_MAX’ undeclared (first use in this function); did you mean ‘INT_MAX’?
startstop_unix_generic.c:1923:22: warning: ‘%s’ directive writing up to 49 bytes into a region of size between 1 and 256 [-Wformat-overflow=]

and I'm just compiling libraries...

Steve Shuff

unread,
Nov 21, 2022, 6:00:42 AM11/21/22
to earthwo...@googlegroups.com
Hi Berto,

I will try it later today and send you a report  , see if we can fix it..

In your notes you describe using the CarlSta/Sub and trig to disk, 
  • "Single station event detection example using Earthworm modules"
Which version did you use to test that and was it on a RPI ?

Thanks,



--
Steve Shuff.

El Berto

unread,
Nov 21, 2022, 6:43:04 AM11/21/22
to Earthworm Community Forum
EW7.9 on Pi3:

- add <stdint.h> in sudsputaway.c

- removed srusb2ew module in makefile in data_sources:
UNIX_MODULES = \
        grf2ew\
        k2ew \
        nmxptool \
        psnadsend \
        q2ew \
        q3302ew \
        scream2ew \
        srpar2ew
#       srusb2ew


- removed C32 value in /libsrc/qlib2/Makefile:

#C32    = -m32
C32     =

- changed flags in ew_linux.bash:

# warning flags for compiler:
#export WARNFLAGS="-Wall -Wextra -Wno-sign-compare -Wno-unused-parameter -Wno-unknown-pragmas -Wno-pragmas -Werror=format"
export WARNFLAGS="-Wall -Wextra -Wno-sign-compare -Wno-unused-parameter -Wno-unknown-pragmas -Wno-pragmas"

# commentato io
#export GLOBALFLAGS="-m${EWBITS} -Dlinux -D__i386 -D_LINUX -D_INTEL -D_USE_SCHED  -D_USE_PTHREADS -D_USE_TERMIOS -I${EW_HOME}/${EW_V$
# messo io
export GLOBALFLAGS="-Dlinux -D__i386 -D_LINUX -D_INTEL -D_USE_SCHED  -D_USE_PTHREADS -D_USE_TERMIOS -I${EW_HOME}/${EW_VERSION}/inclu$


In this way I can compile.

El Berto

unread,
Nov 21, 2022, 6:44:56 AM11/21/22
to Earthworm Community Forum
Yes, it was on RPi but it was first version of RPi..... I don't remember EW version but it was earlier than 7.7.

Steve Shuff

unread,
Nov 21, 2022, 7:10:59 AM11/21/22
to earthwo...@googlegroups.com
Well done Berto,

Hopefully I will get chance to try it out later.  
I have found Trig2disk wont work on EW10 . So I will try to get it working on 7.9 then copy the executable into 10 once compiled and see what happens ...  Having to go backwards to go forwards !!   

Steve Shuff

unread,
Nov 21, 2022, 1:51:54 PM11/21/22
to earthwo...@googlegroups.com
Hello Berto.
I have tried to follow your notes from earlier today but it fails to compile.  It is regarding the ahputaway . I have attached the log.
 Also in case anyone else is trying to follow this, make sure to add the " at the end of the Globalflags line... 
20221121_184542.jpg

El Berto

unread,
Nov 24, 2022, 5:50:41 AM11/24/22
to Earthworm Community Forum

It's simply beacuse you have wrong GLOBALFLAGS, it should be something:
[...] -I/home/pi/ew/earthworm_7.9/include [...]

and not
[...] -I/home/pi/ew/earthworm_7.9/inclu [...]


Check again your environment/ew_linux.bash

Steve Shuff

unread,
Nov 25, 2022, 2:51:24 AM11/25/22
to earthwo...@googlegroups.com
Hello Berto,

Yes you are correct, I am not sure how that error got through as I thought I had seen and corrected it !
I will try to take a look at it over the weekend as it's all under dust sheets at the minute as the room is being decorated. 

On a separate note , How can you tell if a module has compiled correctly? As in, I don't think my Trig2Disk is working which I suppose could be a compile error and I would like to check it.

Thanks 
 



--
Steve Shuff.

Larry Baker

unread,
Nov 28, 2022, 10:47:43 AM11/28/22
to Steve Shuff, 'jona...@ntu.edu.sg' via Earthworm Community Forum, Larry Baker
Steve,

One approach may be to try your Earthworm configuration on a Linux PC or laptop to find out if Trig2Disk v7.10 is really a problem.  You should be able to download binaries for v7.9, v7.10, and v7.10.1 for a PC Linux OS.  Use that platform to trouble shoot your configuration.  And, if you find a bug that way, you can pass that along to the developers to fix the code for all platforms.

I compared the changes to Trig2Disk from v7.9 to v7.10 and v7.10.1.  I didn't see any logic changes.  I doubt that Trig2Disk is your problem.  If anything, given the lack of code changes, if one version of Trig2Disk has a problem, all versions of Trig2Disk will likely have the problem.  If you can nail it down, it can be fixed.

Be sure to enable debug output options for the log files in your .d configuration files.

And, pass along the .d, .desc, and .log files for your configuration when the problem occurs.


On Nov 25, 2022, at 1:15 PM, Larry Baker <larry...@stanfordalumni.org> wrote:

Steve,

Here are the steps I follow to make sure there have been no errors creating Earthworm.

source the environment variables from your modified Linux bash script:

# source earthworm-7.10.1/environment/ew_linux_arm.bash

cd into your Earthworm src directory:

# cd earthworm-7.10.1/src

cleanup any debris leftover from the previous build:

# make clean_unix 2>&1 | grep '[*][*][*]'
# make clean_bin_unix 2>&1 | grep '[*][*][*]'

build Earthworm and save the make output in a log file:

# make unix &>../make-debian-${EW_BITS}.log

If the build failed, the end of the log file will not look "normal".  That is, it will have one or more error messages, as you had, and then the make will quit:

# tail ../make-debian-${EW_BITS}.log

Search the log file for errors in the module builds:

# grep -i ': err' ../make-debian-${EW_BITS}.log
# grep '[*][*][*]' ../make-debian-${EW_BITS}.log

You will not even have a binary if a module does not compile correctly.

Beginning with Earthworm 7.10, the very first few lines of the log file will print out the build settings and the execution environment.


Larry Baker

unread,
Nov 28, 2022, 10:47:49 AM11/28/22
to Larry Baker, earthwo...@googlegroups.com
Also, beware of 64-bit builds from code so old.  I recall I fixed 64-bit bugs in the main transport.c code and wave_serverV.  Try to build for 32-bit Linux on releases older than EW 7.10.  If you have to build for 64-bit Linux, do your best to keep ring buffers and wave tanks within the limits of 32-bit addressing.


On Nov 21, 2022, at 10:30 AM, Larry Baker <larry...@stanfordalumni.org> wrote:

I added support for ARM in Earthworm 7.10.  There is a ew_linux_arm.bash that is used.  Compare that with the ew_linux.bash you are using for the older EW's.

Also, there were source code changes required that you will be missing in the older EW's.  As I recall, Steim compression will not work because ARM gcc char is unsigned, while Intel gcc char is signed.

I think you will have better luck starting with the latest EW, since there have been so many fixes to the code since EW 7.7.

Larry Baker

unread,
Nov 28, 2022, 10:47:49 AM11/28/22
to Steve Shuff, Larry Baker, 'jona...@ntu.edu.sg' via Earthworm Community Forum
Steve,

Here are the steps I follow to make sure there have been no errors creating Earthworm.

source the environment variables from your modified Linux bash script:

# source earthworm-7.10.1/environment/ew_linux_arm.bash

cd into your Earthworm src directory:

# cd earthworm-7.10.1/src

cleanup any debris leftover from the previous build:

# make clean_unix 2>&1 | grep '[*][*][*]'
# make clean_bin_unix 2>&1 | grep '[*][*][*]'

build Earthworm and save the make output in a log file:

# make unix &>../make-debian-${EW_BITS}.log

If the build failed, the end of the log file will not look "normal".  That is, it will have one or more error messages, as you had, and then the make will quit:

# tail ../make-debian-${EW_BITS}.log

Search the log file for errors in the module builds:

# grep -i ': err' ../make-debian-${EW_BITS}.log
# grep '[*][*][*]' ../make-debian-${EW_BITS}.log

You will not even have a binary if a module does not compile correctly.

Beginning with Earthworm 7.10, the very first few lines of the log file will print out the build settings and the execution environment.

On Nov 24, 2022, at 11:51 PM, Steve Shuff <steves...@gmail.com> wrote:

Larry Baker

unread,
Nov 28, 2022, 10:47:49 AM11/28/22
to 'jona...@ntu.edu.sg' via Earthworm Community Forum, Larry Baker
Check your files that setup the environment variables.  I can see in the command line that fails that the include directory name is truncated.  That is why you got the error that earthworm.h cannot be found.

Larry Baker

unread,
Nov 28, 2022, 10:47:49 AM11/28/22
to earthwo...@googlegroups.com, Larry Baker
I added support for ARM in Earthworm 7.10.  There is a ew_linux_arm.bash that is used.  Compare that with the ew_linux.bash you are using for the older EW's.

Also, there were source code changes required that you will be missing in the older EW's.  As I recall, Steim compression will not work because ARM gcc char is unsigned, while Intel gcc char is signed.

I think you will have better luck starting with the latest EW, since there have been so many fixes to the code since EW 7.7.
On Nov 21, 2022, at 1:44 AM, cellectronic <steves...@gmail.com> wrote:

Steve Shuff

unread,
Dec 6, 2022, 5:33:33 AM12/6/22
to earthwo...@googlegroups.com
Problem Solved.
The main issue was down to the time stamp of the digitizer,
Once that was corrected the Carlsub/Trig2disk functioned correctly.

The second issue was that within Carlsub  the defined subnets had a "Space " right at the start before "Subnet" , that needed to be removed !

I now have two Pi,s running EW 7.10 on "Bullseye" and so far all looks good.

Thanks to all those that have responded to my queries. 



--
Steve Shuff.
Reply all
Reply to author
Forward
0 new messages