Create a shortcut to BOINC Manager on the Desktop and change it's properties to read:
"c:\Program Files\BOINC\boincmgr.exe" /m /n 127.0.0.1 /g 31418 /p xxxxxxxxxxxxxx

after following all the steps as described, I could get the second BOINC manager to open, but right thereafter the message "the password you have provided is incorrect - please try again" shows up :-(

I had replaced the many xxx with exactly what was contained in the original gui_rpc_auth.cfg. So - no idea why this password is not being accepted.

So, unfortunately, the second BOINC manager does not work here :-(

Ok, there were two ways I can reproduce the error. First was to mistype the password in the properties of the Manager start command. Second way was to append the gui_rpc_auth.cfg with a txt suffix, which causes the client to create a new gui_rpc_auth.cfg when it is started with the .bat file. Check that the gui_rpc_auth.cfg file that was copied over does not have .txt suffix. If you created the file by right-clicking and choosing create new text file and then renaming it to .cfg, it may still have the .txt suffix. The best was to do it is to have both data directory folders open and drag from the first client to the second client choosing the Copy option.

The best was to do it is to have both data directory folders open and drag from the first client to the second client choosing the Copy option.

I did this right now, but it didn't solve the problem :-(

You did stop/start the client, correct? The gui_rpc_auth.cfg is only read when the client starts, so any changes to it require the client to be stopped and then started again.

You did stop/start the client, correct?

Yes, I did.


But by now I guess that the problem may be somehow related to what Richard was saying above:

...But that assumes that the second client is running, that it's listening on the correct port, etc. etc. etc. I think you get the same "incorrect password" error message whichever part of your configuration is wrong.

:-( :-( :-(

Let's see if the client has even started properly. Check your data directory for the second client and if the client has started properly, you should see a list of files similar to these:

all_projects_list.xml
client_state.xml
client_state_prev.xml
coproc_info.xml
daily_xfer_history.xml
global_prefs_override.xml
gui_rpc_auth.cfg
stderrdae.txt
stderrgpudetect.txt
stdoutdae.txt
stdoutgpudetect.txt
time_stats_log

There should also be one folder named Notices. If only the gui_rpc_auth.cfg is present, then the second client has not started properly.

except for "global_prefs_override.xml" everything is there, also the folder named "Notices".
So the second client obviously has startet properly.

At this point, I don't know what to try other than stopping the second client if running, edit the gui_rpc_auth.cfg and change the password to something simple like the word test then save the file. Start the client again, change the properties of the Manager so the password is test and see if that works.

Instead of changing the password (honestly, I didn't dare to make this change in the original BOINC instance, not risking that this one then wouldn't work either), I closed down both BOINC instances and again copied the gui_rpc_auth.cfg directly. Then I started the first Manager and then the second one.
And now, in contrast to as it acted before, the second one does NOT say "wrong password" right after trying to connect once, but I can see in the lower right corner of the Manager window "connecting to 127.0.0.1" and "disconnected" - in about 2-3 seconds intervals, all time long, until I close the Manager.

This is all very strange, and by now I suspect that this constellation simply will NOT work here (as, from what one can read in the forums, it does not work for many others as well, for what reason ever).

And now, in contrast to as it acted before, the second one does NOT say "wrong password" right after trying to connect once, but I can see in the lower right corner of the Manager window "connecting to 127.0.0.1" and "disconnected" - in about 2-3 seconds intervals, all time long, until I close the Manager.

That is caused be either the client not having started properly or some security software blocking the communication between the two. The fact that earlier you were getting incorrect password errors should indicate that security software is not the problem as the client and manager had to talk to each other to see the password was not correct.

Well, the situation is such that the trial period for the RAMDisk ends in 4 days. It would not make sense to buy it, if I cannot run the second BOINC for GPUGRID computing (more details see my initial posting above).
Which means that if I don't get the second BOINC to function, I will give up this plan and install BOINC on an external USB-3 drive. In fact, I have a spare one around here, so I wouldn't even have to buy it.
From what I have been told here in the forum, speed-wise the difference between RAMDisk and USB3 SSD should, in case of BOINC, not play much of a role.

It's really too bad that 2 BOINCs don't work here, but - as said - this seems to be a common problem :-(

...
Which means that if I don't get the second BOINC to function

From what I've read, your problem is with BOINC Manager not connecting to the right BOINC client, but the client itself seems to be functioning, so if need be you can address it through the boinccmd tool.

You can also attach the first BOINC to GPUGRID, then move the account_*.xml file over to the second BOINC instance's data directory. This if BOINCCMD also proves not to work.

... I will give up this plan and install BOINC on an external USB-3 drive.

Well, not so much install BOINC on the USB drive, but move the data directory there and tell BOINC through the installer (or registry) where the new place is.

You can move the data directory to a directory on the USB drive, then in HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup edit the DATADIR entry to point to the new place of the data directory. Then just start BOINC.

---

You can also attach the first BOINC to GPUGRID, then move the account_*.xml file over to the second BOINC instance's data directory. ...

sorry, Jord, I cannot follow you here :-(

The current situation is that GPUGRID is attached to the first BOINC instance (the 2nd one never worked so far).
So, if I move the account_www.gpugrid.net.xml file to the second BOINC instance's data directory - I guess then GPUGRID will no longer work, neither on the first nor on th second BOINC, right?

Or did I misunderstand something?

When you add a project through the command line, BOINC Manager or an account manager, adding the project generates an account_URL_of_project.xml file that holds the details about your account at that project.
So if you want to run GPUGRID only on the second BOINC instance, you move the account*.xml file for GPUGRID from the first BOINC's data directory to the second one.

When you then start the first BOINC, it will no longer be attached to GPUGRID, but when you start the second instance, it will be.

Of course, if you want both attached, you just copy the file over.

---

When you then start the first BOINC, it will no longer be attached to GPUGRID, but when you start the second instance, it will be.

okay, that's clear to me. My intention is to have GPUGRID attached to the second instance (the one on the SSD) instead to the first, current one (on the RAMDisk).
However, to move the GPUGRID-xml from the 1st to the 2nd BOINC, I have to get the second one to work. In which I did not succeed so far :-(

You don't have to get it to work first.
You have access to the computer system, and the various folders and files.
Shut down BOINC, go to the location of the file, copy it, (or "move" if you want to move it), go to the "same/similar" location in the other instance, and paste.

Then restart BOINC.

What I understand from reading this thread so far is that you managed to get the second client to run, but you cannot get the second BOINC Manager to access it and to add the projects and such. My approach doesn't use BOINC Manager.

You can try to use boinccmd, but it may also like the first BOINC only. I don't think there is a way to tell boinccmd to point to a second running client.

So, you can do everything manually:
- add the account_*.xml files of the projects you want to run.
- add the local preferences file global_prefs_override.xml to set things such as amount of cache and when to run.
- add a cc_config.xml file with specific flags or options.

Now, it may at times happen you want to suspend this BOINC, but you cannot do that via the boinccmd tool. No sweat, you can edit its client_state.xml file, where you have entries for the Activity menu:

<user_run_request>2</user_run_request>
<user_gpu_request>2</user_gpu_request>
<user_network_request>2</user_network_request>

1 = 'run always' / 'always network'
2 = 'based on preferences'
3 = 'suspended'

After you change the value to for instance suspend the CPU use (run_user_request) you exit & restart BOINC. It'll suspend on the CPU then.

---

You can try to use boinccmd, but it may also like the first BOINC only. I don't think there is a way to tell boinccmd to point to a second running client.

Works similar to the shortcut properties for the Manager, except the port has to be entered after the IP separated by a colon. Here is an example using get_cc_status and set_network_mode on the test second client I am using:

C:\Program Files\BOINC>boinccmd --host 127.0.0.1:31418 --passwd xxxxxxxxxx --get_cc_status
network connection status: don't need connection
CPU status
    not suspended
    current mode: according to prefs
    perm mode: according to prefs
    perm becomes current in 0 sec
GPU status
    not suspended
    current mode: according to prefs
    perm mode: according to prefs
    perm becomes current in 0 sec
Network status
    not suspended
    current mode: according to prefs
    perm mode: according to prefs
    perm becomes current in 0 sec

C:\Program Files\BOINC>boinccmd --host 127.0.0.1:31418 --passwd xxxxxxxxxx --set_network_mode never

C:\Program Files\BOINC>boinccmd --host 127.0.0.1:31418 --passwd xxxxxxxxxx --get_cc_status
network connection status: don't need connection
CPU status
    not suspended
    current mode: according to prefs
    perm mode: according to prefs
    perm becomes current in 0 sec
GPU status
    not suspended
    current mode: according to prefs
    perm mode: according to prefs
    perm becomes current in 0 sec
Network status
    suspended: user request
    current mode: never
    perm mode: never
    perm becomes current in 0 sec

C:\Program Files\BOINC>

Ah OK, but if he's already having trouble connecting two BOINC Managers, he may as well have trouble connecting boinccmd to the right directory.

@Bobcat, Thinking about the BOINC Manager trouble, do you have any specific exceptions set in the firewall - and if so, which firewall do you use?

---

There is very little GPUGrid work at the moment anyway, so it could be much ado about nothing.

Maybe I mentioned this elsewhere, but at the risk of repetition: In comparable situations, I just do Folding on the GPU, and use BOINC for the CPU projects. They get along together fine in most cases, except when a VirtualBox project is involved, in which case Folding can crash. But Folding uses its own client, which can install only on your SSD, and not use any ramdisk memory. The write rate of Folding is very low, so you don't need to protect the SSD anyway. And it always has plenty of work, and is scientifically important. It gives you another option at least.

@Bobcat, Thinking about the BOINC Manager trouble, do you have any specific exceptions set in the firewall - and if so, which firewall do you use?

Privatefirewall 7.0 and the exceptions for Boinc and Boinc Manager/BoincTasks are local lan only, allowed on remote ports 31416-31418.