Just wanted to give you a heads up on this Willy. It may be the project site at fault, and not BAM, but just in case....

I apparently had a slightly different URL when I attached to Sztaki than what BAM has. As a result, once my 20 hosts were attached to BAM, they kept trying to attach to Sztaki, and only once they attached did they realize I was already attached to a project named "Sztaki Desktop Grid". The BAM-created host was then disconnected, and I get a message suggesting I detach the old entry so that BAM can connect with the correct URL for that project. Doing that does in fact work.

The problem is that until I discovered this, each attempt was creating a new host record on Sztaki's database. What's worse is that they don't merge properly. When merging, you get something like this:

[SZTAKI Logo here] Unable to handle request

missing or bad parameter: id_1

I've already posted this to Sztaki's site, but figured I'd warn people here about it if they start seeing this message, or large numbers of hosts being created with 0 credits.

Nothing what so ever to do with BAM at all so don't worry on that one. Currently there are 8 projects that need to upgrade there server code to stop this happening, until this happens it will happen to isolated users not everyone will have this problem, until then a lot of multiple hosts will be created on the servers of the projects who don't take any action.

Merging is probably just something broken on SZTAKI, another thing to add on the list of stuff bust there

Well, BAM isn't really at fault, but it did expose the problem in a more obvious way, since the repeated hourly attempts at attaching made for a ton of new hosts in a hurry. I mostly wanted to post here because Willy might get others questioning it in the future.

Thats OK, thought I best mention thats its not BAM as there seems to be one guy going around saying its BAM's fault

One of my hosts that's using the BAM has suddenly started getting multiple entries in the project (as you can see here: http://boinc.bakerlab.org/rosetta/hosts_user.php?userid=784).

This machine (AuthenticAMD AMD Athlon™ XP 3000+), is being repeated over and over. Is this something caused by the BAM (never happened before), and this is the only host with this problem...

Thoughts?

Nope it is a server problem at some projects that only pops its head up now and then, there is a fix for it but it will be a long time coming on Rosetta

But it only *just* started happening... Any idea why it's not happening on any of my other hosts?

Like I said it pops up now and then, you will have to wait for the projects affected to upgrade there servers, there seems to be no pattern to it, it happens to some and not to others

Its a known problem with BOINC and the servers

Got a small problem only with LHC. I suspect it is problem with LHC servers not being up to date. And I don't see that changing any time soon since Chrulle left the project. I have also posted this on their servers.
Every time that my client requests work from LHC a new host is spawned. This means that I have to merge a page full of hosts everyday. This can't be good for their servers if it is happenning unnoticed. I just happened to notice it the other day when I had about 500 host for one computer. This does not happen for any of the other projects that I'm attached to.
Other than that, I'm very very pleased with BAM.
And Willy, keep up the good work, thanks.
<---- BAM Account No.

I've got the same problem with Rosetta and Boincsimap. The machine in question is running XP and on the latest BOINC release.

I don't think this a BAM issue at all. It's probably caused by the server side software of a project.

Experiencing exactly the same problem as described above for Sztaki.

Also the same problem/error when I try to merge the replicated hosts.

I have detached my BAM from this project for now as their host creation numbers seem to be going exponential.... (have they noticed yet?)

Top site Willy - you the man!!

Yes some one has reported the problem to them, and I have put in a quote from D.Anderson to the post over there to point them in the right direction as to where the patch is.

If you are like me and are tired of waiting for a project to fix things on the server, this may help you stop creating new host's with each contact. It worked for me at LHC. Just substitute LHC for what ever project that needs your attention
Below is a copy of my post at LHC.

I figured out what the problem is, at least in my case and how to correct it. This may work for you too. In my case the computer host ID # is not getting passed to the client.

1) You can verify that this is your problem too by checking your client_state.xml file. Look for the section for LHCathome then find the line that states the <hostid>xxxxxx</hostid>. If you have value of 0 for the ID# then this is what causes it to keep making new hosts. Now exit the text editor without saving!
2) If you want to fix this shut down Boinc.
3) Go to your account page at the project and click on View computers then write down the host number from the most recent one there.
4) Go back and open client_state.xml.file with a text editor and replace the 0 with the number that you wrote down.
5) Save the file.
6) Restart Boinc

Hope this helps, it worked for me.
Now I have to find something else to do on this boring rainy saturday. This did not take but a few minutes to figure out, lol.

It worked for me week(s) ago on SZTAKI - a real pain when WUs were like 1 or 2 minutes, hundreds hosts overnight!

Would have been nice if you shared the knowledge with the crunching community like I did.

Yes, Steve, good suggestion.
As I was about to undertake final exams on University today, I wasn't in a mood to write long messages. No it's succesfully done, I guess I can be more active on many projects and forums again :-)

Already put a short answer on SZTAKI several days ago here. There was also other suggestion how to solve the bug.

Added your how-to on chess960 here where this bug aply as well and they have some troubles updating code...