Apparently all of the remote computers at that location only have failed to stay attached via LOGMEIN, but are still attached to the internet.
And there's another variant that does affect the video.
I wonder if anyone / everyone else has logmein running?
Off to my meeting.
With LogMeIn you get driver substitution and port capture ( similiar to Remote Desktop connections ) and thus with video hardware you are going to displace/screw-up GPU work at least. I'm not sure of the exact effect on HTTP but I don't think it's helpful for BOINC activity.
Cheers, Mike.
( edit ) I used LogMeIn when overseas a few years ago to access a BOINC running machine and found that it caused the work units on that machine to error out. I worked that out after a few days ( ie. that LogMeIn accesses were triggering the errors ) and stopped using LogMeIn to that machine ( to good effect ).
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
With LogMeIn you get driver substitution and port capture ( similiar to Remote Desktop connections ) and thus with video hardware you are going to displace/screw-up GPU work at least. I'm not sure of the exact effect on HTTP but I don't think it's helpful for BOINC activity.
I hear and understand.
Just another data-point: I've been using LogMeIn with no ill effects on any running work units of any of at least four projects for some months, perhaps as much as more than a year.
I did have trouble with RealVNC or something similar a few years ago and stopped using it. Every log-in caused a driver crash.
Maybe newer drivers or updated LogMeIn clients dealt with this? I don't know. I do know that computers that did not over-fetch work also have LogMeIn installed and running on them.
I'm not having project application errors using LogMeIn. If I am having trouble now, it would be only with Einstein and only with BRP5 work-fetch. I had no problems with BRP4. The trouble, on multiple computers, did not start until BRP5.
Something changed May 26th with BRP5.
So, what changed?
My meeting lasted until after 10:00pm so I did not get out to the remote location tonight.
With LogMeIn you get driver substitution and port capture ( similiar to Remote Desktop connections ) and thus with video hardware you are going to displace/screw-up GPU work at least. I'm not sure of the exact effect on HTTP but I don't think it's helpful for BOINC activity.
Maybe newer drivers or updated LogMeIn clients dealt with this? I don't know. I do know that computers that did not over-fetch work also have LogMeIn installed and running on them.
I'm not having project application errors using LogMeIn.
I've used LogMeIn to access a machine with both AMD and NVidia GPUs for over a year. Never had a driver crash or WU crash yet. Knock on wood.
My experience with LogMeIn is dated to 2011, haven't used it since. I had a search around for deeper details of their methodology to no avail ( but no surprise that they would keep that out of public domain ).
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
Ok... this is
)
Ok... this is new.
Apparently all of the remote computers at that location only have failed to stay attached via LOGMEIN, but are still attached to the internet.
And there's another variant that does affect the video.
I wonder if anyone / everyone else has logmein running?
Off to my meeting.
RE: Ok... this is
)
With LogMeIn you get driver substitution and port capture ( similiar to Remote Desktop connections ) and thus with video hardware you are going to displace/screw-up GPU work at least. I'm not sure of the exact effect on HTTP but I don't think it's helpful for BOINC activity.
Cheers, Mike.
( edit ) I used LogMeIn when overseas a few years ago to access a BOINC running machine and found that it caused the work units on that machine to error out. I worked that out after a few days ( ie. that LogMeIn accesses were triggering the errors ) and stopped using LogMeIn to that machine ( to good effect ).
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal
RE: With LogMeIn you get
)
I hear and understand.
Just another data-point: I've been using LogMeIn with no ill effects on any running work units of any of at least four projects for some months, perhaps as much as more than a year.
I did have trouble with RealVNC or something similar a few years ago and stopped using it. Every log-in caused a driver crash.
Maybe newer drivers or updated LogMeIn clients dealt with this? I don't know. I do know that computers that did not over-fetch work also have LogMeIn installed and running on them.
I'm not having project application errors using LogMeIn. If I am having trouble now, it would be only with Einstein and only with BRP5 work-fetch. I had no problems with BRP4. The trouble, on multiple computers, did not start until BRP5.
Something changed May 26th with BRP5.
So, what changed?
My meeting lasted until after 10:00pm so I did not get out to the remote location tonight.
RE: RE: With LogMeIn you
)
I've used LogMeIn to access a machine with both AMD and NVidia GPUs for over a year. Never had a driver crash or WU crash yet. Knock on wood.
My experience with LogMeIn is
)
My experience with LogMeIn is dated to 2011, haven't used it since. I had a search around for deeper details of their methodology to no avail ( but no surprise that they would keep that out of public domain ).
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal