Messages

Some things that may make sense to enhance the project:
(meant ONLY as constructive suggestions)

a review/rethink of the menu items -- as a project grows, some items get misplaced or are not in keeping with the original intent of the menu classifications, such as

    (1) Show Log and Show Rule Log may be better placed under the SHOW list instead of the HELP
    (2) Report Completed Projects may be better placed under the PROJECTS lists to group project-related items
    (3) EXTRA may be better renamed CONFIG or SETTINGS or OPTIONS -- as that is what all the other items really do

A solution to my problem
for those who might need a similar solution.

I re-addressed the temperature issue with the SETI init phase of the GPU tasks.

Now that I know (from Fred - thx) that the tasks are using the temperature based upon the *designated* task type, and not the phase that they are currently in.......

[ for reference, the SETI GPU tasks seem to do a CPU-intensive phase that sets up the GPU task,  and *sometimes* they do not complete this phase correctly, and just seem to get 'stuck' in a never ending loop, going on until manually reviewed and suspended -- lost time, lost CPU power that could be more productive ]

Now the rule checks that the temperature (the GPU temp) is below a certain value for a few minutes, and then suspends the task if that condition exists.

Rule:  Temp < 45 C  for 2 minutes
[ plenty of time for the init phase to get the GPU going, and the GPU temp should rise above this value for the duration of the processing ]

This now works for what I wanted to catch.
It has caught several tasks already, and it keeping the CPU/GPU system productively busy.

Thanks for the rules operations !

Thanks.

Temperature Rule

This is currently one value.
And I suspect that it checks the CPU.
Is there a way to check the CPU and the GPU temperatures separately ?

This is because some recent SETI tasks are having difficulty jumping into GPU mode after initialization, and stay with a high CPU (==> CPU temp), when it should be released and the GPU take over.

Having this breakout will allow me to trap for this occurrence.

Thanks.

After posting this, I looked at this in more detail.

I was thinking that the calc was off by one hour, but I am thinking now that the duration was chnaging on me when I was doing the tests, and that I totally missed that.

I believe that it is indeed working correctly.

I may be worth taking a quick look, but do not waste any time on it.

Sorry.
Thanks.

Running Hydrogen@Home with its tight schedules required me to look at the Deadline Warning settings.

As I understand it (I am guessing), the deadline warning is set as:

Current Time  + Estimated Duration >=  Deadline time

This seems to be fine when the deadline time is less than midnight.

But it seems to be failing (that is, just giving the wrong coloration, nothing else),
when the deadline is over the midnight boundary.

Example:

Current Time + Estimated Duration     Warning Time                   Deadline  
4pm                 6 hrs                               3 hrs                            1 am                     OK
4pm                 6 hrs                               4 hrs                            1 am                     Warning Color Shown

I have sent images to the same email given before.
The clock-time is not shown on the images, but they are about 4pm

Thanks.

Display of timers in Suspend Mode

I have noticed this only a few times before in earlier version, starting in the early .5x , but it is certainly not a regular occurrence.

When I do notice it, I am completely unable to track down any reasonable criteria that makes it happen.

When I suspend an item, *sometimes* the timer of the elapsed time shows some value (typically without hours) that does not correspond to the elapsed time.

I have screen shots of the original time on the item, the changed (incorrect) time shown when suspended, and the recovery of the original time after restarting the item.

Please let me know where to send or upload the images (600K zip file, multiple formats for image clarity)
The affected item is line #2 in the list

Thanks.

I would like to add in here that there is a simular 'follow' process that I have run into.
It was not critical, so I did not report it, as it didn't happen often.
But while you are there in that same code, you might want to look at the following circumstance.

When there are more than one Filter operation (such as Suspended and Waiting to run),
and both combined filters are titled the same (like: 25 [Tasks]),
performing some operations that require a refresh of the set of tasks
will result in the follow process to 'follow' to the one that was not previously selected.
It generally just gets the first in the list that has that same title.


Thanks.

To all...

Thank you for the response on this.  I am a (s)low user of the Boinc systems, but appearently I stepped into a configuration that is of some concern to a number of users.

I will update and reconfigure as I can.

Thanks again.

On updating to the latest (.44) version of BT with the latest Boinc 6.10.36 I am seeing a symptom of performing some BT operations.

I will sometimes suspend a couple of tasks to get a few more running, or to prioritize some in the queue.

What I see happening now that was not happening before is that the Boinc will hit the configured Boinc Preference: "While processor usage is less than ____"
I have had this at 25%, and previously have had no occurrence of the symptom.

What happens is that all of the tasks will go into an 'automatic suspended' mode with a status of 'waiting to start; suspended' (or the like) and the message window will record a "Suspending Computation - CPU usage is too high".  This will clear up and go back to normal withing 10-20 seconds (I use slow refresh rate).

Setting the Boinc Preference higher (like 50% seems to affect the symptom little, as it easily goes into this mode.
Setting it to 0 ('no restriction') fixes the symptom, but then this completely removes the automatic suspension that I do want to have when I am performing other processing on the computer.

What seems to be happening is since BT is not considered an official 'Boinc Process', and is being considered a foreign (ie: normal user computer process) and thus Boinc is automatically suspending all of its processing when BT does some things that take a moment of two to complete.

Maybe Boinc can be either be told that BT is a friendly process and not to be considered foreign, or BT can temporarily set the preference to the max while it is doing some processing that may take a bit of time and reset it back when done.

Just a thought.
Thanks.

A Wish List item...

Addition of a quick(er) way to set and unset the Combined Filters that is based upon the status;
maybe something like a "control-double-click"  that would examime the current status of the item under the cursor, and then set (combine) or unset (expand) those of the same status.

Where this would be helpful is that for some of us who do have hundreds of tasks, and we beleive that it is convenient to leave the task list viewable (with all except actual 'running' tasks being combined), then this would eliminate several clicks and menu selections in order to do a series of expansions-combinations for a quick view of what is pending and back again.

Thanks.