Messages

everything that we're saying is correct, but not so correct, because the computer (Laptop) I connect
has no dns known available that the fixed computers at home should know in the remote_host.cfg,
so, is it anyway of giving a fix name to the laptop and this intro in the remote_host.cfg of the fixed ones??

And what about giving also the laptop a DynDNS name? The same way as the router at home?

P.S. Writing any text longer then some 12 lines is a torture, the input field is being scrolled up after each typed character    

For clarification - I've meant writing a longer text into the Forum.  Could anything be done with this? 

Could it be possible that the project and download pages were obviously visible from this forum? Or am I just not able to see it?

Be it a separate category in Generals, or at least a question in FAQs pointing to them, or even maybe the small "eFMer  BoincTasks  TThrottle" banner in the upper right forum corner could point to their web pages?

[...]you could understand that what I'm asking is a very strange thing, on the way that from one computer
outside the router, firewall, etc, you want to control all the computer behind the router, etc.

Not strange at all, but very common task.

if one remote desktop connection is done, the CUDA applications of accessed computers will fail to work.

I've read about this problem on the mailing lists. Yes, remote desktop is out, and it is also too bandwidth-hungry for your needs.

but I think that if you know your public IP of the router...

Sure, DynDNS and the like services.

...and forwarding every 31416 port to every computer that you have in your network, could be easy to do it, I'll try and I let you know.

Exactly! You will forward your router's incoming ports e.g. CorsairNet.DynDNS.org:2010 ... CorsairNet.DynDNS.org:2035 to your local network's clients 192.168.4.10:31416 ... 192.168.4.35:31416 and will see all 26 BOINC clients at a glance

but the problem is the name (or IP) of the controlling computer, that must be in the "remote_hosts.cfg"
of the controlled computers.

so, could it be names without IP numbers is the "remote_hosts.cfg"??, as the controller computer use to
have a dynamic IP and unknown.

Right. The remote_ hosts.cfg can also contain domain names, not just IPs, like CorsairCruiser.DynDNS.org 

How about a color setting available for task that execeed their completion time?

What about different colors for tasks, which are already past the deadline and tasks, which will exceed their completion time?

I've sent back my translation attempt 

I will not be able to check if you made any spelling errors.

Perfect that's good...
That's what I've mentioned - someone will have to laugh me out 

  could be possible in the future to run from e.g. laptop away home and accessing to home computers with BT??

[...]
Another possibility is to create separate SSH tunnels from your B.T. laptop to BOINC clients and access the clients over these tunnels. The following B.T. version is promised to support this.

I can now at least confirm that I could successfully connect BT 0.23 over a SSH tunnel (using a PuTTY terminal) from localhost's port 31417 to a client hanging on a remote machine on port 31416 

I've sent back my translation attempt 

Just a minor cosmetic problem, seen in TTh 1.69 and at least one previous version:

• In the Programs tab, the "Temperature / Core" field has a gray background, which also indicates that its value can not be edited. The "Run time / Running Cpu" field next to it has a white background, which may suggest that it can be edited, although the field is also read-only. This field should be IMO grayed too.

At some point in time I'd like to make a Slovak translation.

It should be also understandable for most Czech users, or a good starting point to create their native translation. (I could also create a rough Czech translation in paralell, but I do not intend to be laughed out .)

This trailing ".exe." problem for PrimeGrid has been hopefully fixed in TThrottle 1.68.

A picture ? To make sure we understand each other.

OK, here it goes:

The percentage is somewhere between upper and lower 10º line.

Percentage just reached the lower 10º line. Note that there is still just the 1º wide area below the line. Pretty sufficient.

Percentage just crossed the lower 10º line, it reached 1º below it. The graph was enlarged by one additional 10º area below the line.

Percentage is reaching the upper 10º line. There is still just the 1º wide area above the line.

Percentage just reached the upper 10º line. The graph was already enlarged by one additional 10º area above the line, although the 1º wide area (like in the example with the lower bound) would be sufficient. Especially in the case of the Percentage graph at 100% it is not necessary at all, because it will never exceed 100%.

Percentage just went above the upper line, by 1º and the upper area off course have to be larger.


P.S. Writing any text longer then some 12 lines is a torture, the input field is being scrolled up after each typed character    

The scaling is now done with all settings e.g. Temperature and Percentage even if the Temperature of Percentage is not there.
If you mean this let me know and I make it a "wish", to scale only on the displayed item.

Hmm, and I can see the opposite: if I select just the temperature(s) or just the percentage, the graph is scaled accordingly. I believe it is also right this way.

I've meant something else. Just encountered it actually behaves differently at the bottom and at the top of the graph (so my description was not exact, the bottom graph part already behaves exactli how I like it so let me rephrase this):

Currently when any curve touches any 10º line from bootom up (e.g. 90,1º), the graph is resized accordingly, that an additional free 10º area is visible above the touched line.
When any curve touches any 10º line from upside down (e.g. 70,0º), the graph still not resized and just an additional free 1º area is visible below the touched line.
But only when any curve crosses any 10º line from upside down and goes below (to e.g. 69,9º), the graph is resized accordingly, that an additional free 10º area is visible below the crossed line.
An example: When the Throttling curve is constantly drawn at exactly 100% (and just the throttling Percentage graph is being displayed), the graph is displaying the area from 99º to 111º. (With this extreme example, I would expect either 99º to 101º or 89º to 111º.)

Could the resizing be redefined to "If any curve crosses any 10º line (i.e. should be displayed at e.g. 91º or 69º), then the graph is resized by one more completely free 10º area above/below"? (So my example throttling percentage graph would be drawn in a 99º to 101º area graph.)

One wish to the automatic graph scaling:
Currently when any curve touches any 10º line, the graph is resized accordingly, that an additional free 10º area is visible on both sides of the touched line.
When the Throttling curve is constantly drawn at 100% (and the temperature curves are below it), the graph is unnecessarily resized by one more completely free 10º area above the line (100º - 110º). But there is actually one more free area 1º wide.
Could the resizing be redefined to "If any curve crosses any 10º line (i.e. should be displayed at e.g. 91º or 69º), then the graph is resized by one more completely free 10º area above/below"?

A bug:
When my target temperature is set to 75ºC, after changing the preference to use ºF, the target temperature is recalculated to 167ºF, but the graph remains displaying the target temp line at 75ºF. Even after changing the target Fahrenheit temperature, the graph keeps displaying the target line at its Celsius temperature.

• In the Computers window, I can see a computer with password in cleartext and the port is displayed as "*****". (This happens after creating a computer.)
  After restarting BT (or closing and reopening Computers window), the password is displayed as "*****".
• I've two BOINC clients accessible, one over localhost:31416, another (actually remote, tunnelled through local port 31417) one as localhost:31417. Although I've defined just the remote computer and specified a password for it, BT still obtains the password from localhost's client folders, instead of using the specified password:
  16:34:14 computers ---- Computer = localhost get local password: 460c71e1a7a6f5398g3ce74hb963fk7c
  16:34:14 computers ---- remotemachine , localhost
  16:34:14 Connect, passwd ---- Host: localhost,remotemachine,Port: 31417, scheduler request host ID doesn't match authenticator (Passwrd)
  Just like the manual states for the port numbers "Leave the port column empty in the computers window, unless you know exactly what your are doing.", the same should be true for passwords and if a password is specified for a computer line, BT should obey it and not guess according to the "localhost" in the IP address.