running more than one earthworm at a time

[Mario Aranha]

Is it possible for more than one user to run Earthworm (different
start/stop) at the same time on a computer?
What needs to be done (different ringIDs, etc) to make this happen?

Thanks,

-Mario

[Brett Zimmermann]

Yes you can run multiple instances of Earthworm. I have been testing it out on a virtual machine at the PNSN. You will need to rename certain files and rings to accommodate it. You don't want to do a status check and it bring up the wrong instance, or your startstop start everything under one and not the other. What I did is I made two different /run/params folders, named /run and /run2, /bin and /bin2. I also added some aliases to help navigate to the desired path and to ensure things worked properly such as these

tl () { tail -n50 ~eworm/run/log/*"$@"*`date -u +%m%d`.log; }

tl2 () { tail -n50 ~eworm/run2/log/*"$@"*`date -u +%m%d`.log; }

#status1 () { source ~/bin/ew_linux.bash && status; } <--

#status2 () { source ~/bin2/ew_linux2.bash && status; } <-- These two statements are commented out but were used to take the current status of either instance. Currently the testbed I was using is being used to test something else and these were commented out.

The stuff in the params directory remained the same aside from creating something like WAVE_RING2.  

I would suggest a testbed to try this all out with. As you experiment you will find what works for your system as I figured out what worked for mine. 

--
--
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
https://urldefense.com/v3/__http://groups.google.com/group/earthworm_forum?hl=en__;!!K-Hz7m0Vt54!ln13w-fGwa0mvvFfH5YkFuLcWTgO6OPtCUBXq7ZC78lk_Ucm1ruWhgHLCXKEEsWk9DieXH_wa9skLhdJ3AQHhg$

---
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://urldefense.com/v3/__https://groups.google.com/d/msgid/earthworm_forum/3ece7b5b-0fb1-038e-e920-50d2532a1337*40berkeley.edu__;JQ!!K-Hz7m0Vt54!ln13w-fGwa0mvvFfH5YkFuLcWTgO6OPtCUBXq7ZC78lk_Ucm1ruWhgHLCXKEEsWk9DieXH_wa9skLheksn3lOg$ .

--

Brett Zimmermann

USGS - PNSN

bzimm...@usgs.gov

bzi...@uw.edu

(520)-255-5253

[Paul Friberg]

One of the most important RINGS that you must identify as different on the two instances is the FLAG_RING, which is a somewhat hidden ring. If that is shared between the two instances, if you stop EW in one instance, it will kill both EW's. Set the ID for the FLAG_RING to something different on each instance.

Either way, change all the RING ID's in the earthworm.d file, unless you want to share data between EW instances.

Paul

Paul

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/CAGqicCurfysE1i8H1e-p0HhZ73yGe2%3DT3pDAbETKX2YufZyseg%40mail.gmail.com.

--

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

[Mario Aranha]

Hi Paul,

Turns out there's another hidden ring called NAMED_EVENT_RING which I had to add to "earthworm.d" and also give it a unique ID in order for multiple EW instances to work. Though "ipcs -m" indicated I didn't own any shared memory regions, another user's EW instance was creating a hidden region with this ID and this was killing my instance.  On first glance I don't quite understand your explanation in "ew_nevent_message.c" so if you can explain that would be great.

Found this out the hard way by going into 'startstop' with 'gdb' and stepping through 'tport_create()' which was failing with this message:
"ERROR: tport_create shmctl (1; Operation not permitted) Region: 9991"
And turns out you have:
#define NAMED_EVENT_DEFAULT_KEY 9991

BTW in a few days/weeks I will probably send you a fixed version of 'coaxtoring'  which we are using in EEW. So look out for that.

Thanks,

-Mario

-- 
Mario Aranha
Berkeley Seismological Lab
UC Berkeley
(510)643-3898