Hey buddy can you spare a Gromacs?

[FlybyKnight]

First, my power company blinked me off 5 times over 10 hours, now
for nearly 20+ hours Stanford won't give me a Gromacs.:banghead:

There are Gromacs being shown as current projects, but I can't seem to hook one.

What's the deal?

Is everybody seeing this? :hic:

[Medic193]

I've never seen a problem on any of my boxes getting a gromacs. Course that's only on my home farm. Don't know what happens with the unmonitered ones.

[Bruce]

Originally posted by FlybyKnight
First, my power company blinked me off 5 times over 10 hours, now
for nearly 20+ hours Stanford won't give me a Gromacs.:banghead:

There are Gromacs being shown as current projects, but I can't seem to hook one.

What's the deal?

Is everybody seeing this? :hic:

What command line switches are you using?

Post applicable portions of your FAHlog. (probably better at folding-community.org than here, but either one might help.) Maybe somebody can see a problem.

[FlybyKnight]

As luck would have it I am now finally grabbing Gromacs again.

I do notice that the ones that I am getting are on the server 171.64.122.123

p537_BBA5_N in water and
p544_BBA5_ext

These were not showing up on the list of current projects for many hours.

During my Gromacs famine, there were Gromacs showing up on the list of current projects, associated with other servers IP addresses.

I.E.
171.64.122.110 p340_gnra_nat
171.64.122.110 p341_gnra_den
171.64.122.144 p909_vill_str0

For some reason my machines would not be given any of these. (Though I have received them before)

I use the BETA 3.25 software and switches -advmethods -forceasm -local -verbosity 9
I use an OS mix of win98 win2k redhat 7.2 7.3 and 8.0

After every failure to load a Gromacs, I was -delete (ing) the work from my queue
I even deleted and re-installed the FAH software on several machines.

I am not really looking for a investigation. I really wanted to know if this was a problem for many or just me.

I guess at this point, I am trying to understand why with Gromacs available, the FAH program would act as if they did not exist.

This is not the first ime that this has happened, but it was the longest period of time that it persisted.

[Bruce]

Originally posted by FlybyKnight
During my Gromacs famine, there were Gromacs showing up on the list of current projects, associated with other servers IP addresses.

I.E.
171.64.122.110 p340_gnra_nat
171.64.122.110 p341_gnra_den
171.64.122.144 p909_vill_str0

For some reason my machines would not be given any of these. (Though I have received them before)
I can't answer your specific question, other than to say that each server has a set of characteristics and the Assignment Server tries to match your request to the most appropriate server. Servers may provide WUs only to Win or Lin or Mac users or some combination. They may provide WUs only for those preferring Advmethods or preferring GAH or preferring FAH or to a combination of those. A server may provide WUs only through port 8080 or to either 8080 or 80. Occasionally (rarely) a project is distributed by more than one server. Each server has a "Weight" and a "Minimum Speed" (rarely used) so the proportions can also change.

All of those factors change from time to time, and it's really difficult to predict what's going to happen.

I use the BETA 3.25 software and switches -advmethods -forceasm -local -verbosity 9
I use an OS mix of win98 win2k redhat 7.2 7.3 and 8.0

After every failure to load a Gromacs, I was -delete (ing) the work from my queue
I even deleted and re-installed the FAH software on several machines.

You realize, of course, that that is really bad for F@H. Everytime somebody deletes a WU, the project stalls until the WU expires. The WU is not reassigned until the server has given you a fair chance to return it.

I am not really looking for a investigation. I really wanted to know if this was a problem for many or just me.

I guess at this point, I am trying to understand why with Gromacs available, the FAH program would act as if they did not exist.

This is not the first time that this has happened, but it was the longest period of time that it persisted.

It's because there is scientific value in each of the projects, and there is no guarantee that any single project has any WUs to distribute to your OS at the particular moment you were looking for a new assignment.