Сб 01 мар 2025 05:30:57 | Rosetta@home | [http] HTTP error: SSL peer certificate or SSH remote key was not OK

still no good

Thanks for checking. Seems like I did just get lucky a couple of times in a row. (Checking again: 4 times in a row)
I'd better fix it properly when I get home on Sunday.

s/Sunday/Monday
Now done. Let's see

---

Сб 01 мар 2025 05:30:57 | Rosetta@home | [http] HTTP error: SSL peer certificate or SSH remote key was not OK

still no good

Thanks for checking. Seems like I did just get lucky a couple of times in a row. (Checking again: 4 times in a row)
I'd better fix it properly when I get home on Sunday.

I've been trying to rejoin Rosetta@Home for a while now, and have been discouraged as a result of these technical issues, in combination with a number of signals (lack of communication from the project admins, inconsistent work, news regarded Rosetta focusing on AI-powered models) that suggested that Rosetta@Home had reached the end of its useful life.

Given the most recent news update regarding the project, I made another attempt. One of my computers was able to join the project without any issues, the other ran into the transient HTTP error that I (and others) have previously encountered.

So it seems this technical problem is not yet resolved, unfortunately. :(

Сб 01 мар 2025 05:30:57 | Rosetta@home | [http] HTTP error: SSL peer certificate or SSH remote key was not OK

still no good

Thanks for checking. Seems like I did just get lucky a couple of times in a row. (Checking again: 4 times in a row)
I'd better fix it properly when I get home on Sunday.

I've been trying to rejoin Rosetta@Home for a while now, and have been discouraged as a result of these technical issues, in combination with a number of signals (lack of communication from the project admins, inconsistent work, news regarded Rosetta focusing on AI-powered models) that suggested that Rosetta@Home had reached the end of its useful life.

Given the most recent news update regarding the project, I made another attempt. One of my computers was able to join the project without any issues, the other ran into the transient HTTP error that I (and others) have previously encountered.

So it seems this technical problem is not yet resolved, unfortunately. :(

You're right that it's not fixed.
Editing the hosts file in the way described above (and repeated below) is the only known reliable fix - just once per computer is all
(I do notice you now have tasks on both your PCs so this may be redundant, but bears repetition anyway.)
Also, task availability since mid-January has been the best it's been for a very long time. Almost always something to grab for two solid months now.

To edit the Hosts file, follow these steps:

Open Notepad as Administrator:
Search for "Notepad" in the Windows search bar
Instead of left-clicking it, right-click on Notepad and select "Run as administrator"
When the dialog box pops up select Yes

To open the Hosts File: In Notepad:
Go to File > Open
Navigate to C//Windows/System32/drivers/etc which will appear empty
Change the file type filter from "Text Documents (*.txt)" to "All Files"
Select the "hosts" file and click "Open"

Edit the Hosts File:
On a fresh line at the bottom, type
128.95.160.156 boinc-files.bakerlab.org
Save the file

That should be it
No need to reboot
Return to Boinc manager, select Rosetta and click the "Update" button
Cross fingers

---

You're right that it's not fixed.
Editing the hosts file in the way described above (and repeated below) is the only known reliable fix - just once per computer is all
(I do notice you now have tasks on both your PCs so this may be redundant, but bears repetition anyway.

Yeah, I was aware of that workaround and implemented it earlier.

That being said, asking users to modify the hosts file on their systems (and not even that -- the users have to go on the forum and discover this workaround for themselves) isn't really a feasible way of running a project that relies on crowd sourced computer capacity from users with varying technical capacity. Especially combined with the intermittent work availability, it sends a signal that the project isn't well-maintained and may even be abandoned.

That's why I didn't rejoin until a few days ago. It's not because I didn't know how to get Rosetta working on my systems; it's because I was hesitant to to attach my computers to a project that didn't seem to have anyone maintaining the infrastructure, particularly since doing so essentially means giving permission to the project to download and run binaries on my machines.

That being said, asking users to modify the hosts file on their systems (and not even that -- the users have to go on the forum and discover this workaround for themselves) isn't really a feasible way of running a project that relies on crowd sourced computer capacity from users with varying technical capacity. Especially combined with the intermittent work availability, it sends a signal that the project isn't well-maintained and may even be abandoned.

That's why I didn't rejoin until a few days ago. It's not because I didn't know how to get Rosetta working on my systems; it's because I was hesitant to to attach my computers to a project that didn't seem to have anyone maintaining the infrastructure, particularly since doing so essentially means giving permission to the project to download and run binaries on my machines.

I can only agree.
It's part of the reason I tried to mention all the current issues I'm aware of in response to the Admin's post, hoping they'll look into the responses and pick up those outstanding issues.
Imo at least a couple of those issues are solvable in 5 minutes - not even a whole day of concerted attention to them.

Tbf I don't know what's involved in getting renewed security certificates - that one may take longer if it's up to some other organisation authorising or approving them or issuing them to be employed here.

---

lack of communication from the project admins, inconsistent work, news regarded Rosetta focusing on AI-powered model

Where did they say Rosetta would focus on AI-powered model? I can only find an announcement that mentioned AI and it basically said "deep learning based models have limitations on small molecule and peptide design, so we developed a virtual screening protocol named RosettaVS to cover this".

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=15898