If SETI@Home starts up again, you may need to check it for the following problem first:

There is currently a problem with a number of BOINC projects that have switched to https downloads and uploads - those downloads and uploads are blocked, at least for Windows users.

The problem is due to an expired entry in the ca-bundle.crt file that's part of BOINC.

For users who can't edit that file properly, this is likely to require a new version of BOINC containing an updated version of that file.

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=14006

https://boinc.berkeley.edu/forum_thread.php?id=13758&sort=5

SETI is not affected by this problem:

31/05/2020 07:49:12 | SETI@home | Scheduler request completed: got 0 new tasks

That would have failed with an error message on an affected project.

But please do check your other projects, and let us know if you find another one - we know of three so far.

SETI is not affected by this problem

Maybe not this time, but it was couple of months (probably more than a year) ago when I had to update the ca-bundle.crt for my Xperia Ray running NativeBOINC.

---

Does SETI@Home have a good contact with the people running the boinc_dev mailing list?

I haven't seen any email from that list since the problem started for the affected BOINC projects, so I suspect that the ca-bundle.crt problem has disabled the mailing list, and possibly the other mailing lists that connect the BOINC developers.

Does SETI@Home have a good contact with the people running the boinc_dev mailing list?

I haven't seen any email from that list since the problem started for the affected BOINC projects, so I suspect that the ca-bundle.crt problem has disabled the mailing list, and possibly the other mailing lists that connect the BOINC developers.

Saw it on their list a few hours ago. May take them some time to fix.

Does SETI@Home have a good contact with the people running the boinc_dev mailing list?

I haven't seen any email from that list since the problem started for the affected BOINC projects, so I suspect that the ca-bundle.crt problem has disabled the mailing list, and possibly the other mailing lists that connect the BOINC developers.

Saw it on their list a few hours ago. May take them some time to fix.

I emailed them direct (not over the list) over 24 hours ago. Not heard a peep back yet.

I've now found that contacting them through the web site of the list works. Unclear if the email version does, though.

Greetings,

The are remedy procedures in this thread on Rosetta. There are procedures for both Windows and Linux. I'm not sure about Macs. I fixed mine with a Linux procedure yesterday. :)

Have a great day! :)

Siran

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

If any Windows user, 64-bit only, is still affected by this, there is a hotfix v7.16.7 of BOINC available from https://boinc.berkeley.edu/download.php

Thanks for your efforts Richard

Android needs an urgent fix, as most devices are not rooted, so it is impossible to edit the crt file on Android.

I have 2 LHC Tasks that can't report, with a 3rd one in progress.

Not sure about WCG yet, I've got some of those in progress on Android too.

If any Windows user, 64-bit only, is still affected by this, there is a hotfix v7.16.7 of BOINC available from https://boinc.berkeley.edu/download.php

So, running Vista x86 I actually had NO chance to get fix by downloading new client bundle? LoL, BOINC should more carefully grant "expert helper"tags then indeed :D

---

SETI apps news
We're not gonna fight them. We're gonna transcend them.

Thanks for your efforts Richard

Android needs an urgent fix, as most devices are not rooted, so it is impossible to edit the crt file on Android.

I have 2 LHC Tasks that can't report, with a 3rd one in progress.

Not sure about WCG yet, I've got some of those in progress on Android too.

Checked BOINC/Rosetta on my Redmi Note 8 - 2 issues.
1) It had incompleted tasks with deadline... 12 may (!!!!). With BOINC startup settings "auto". That is, BOINC should run at startup and DO progress but it as usually didn't. So, computations on Android really require day by day user intervention, they will do no progress w/o.
2) i aborted those tasks... and dived directly into certificate issue. Phone not rooted so I can't just replace affected file then... So, device more powerful than some of my Windows netbooks now out of fleet completely :/

---

SETI apps news
We're not gonna fight them. We're gonna transcend them.

[snip]

Checked BOINC/Rosetta on my Redmi Note 8 - 2 issues.
1) It had incompleted tasks with deadline... 12 may (!!!!). With BOINC startup settings "auto". That is, BOINC should run at startup and DO progress but it as usually didn't. So, computations on Android really require day by day user intervention, they will do no progress w/o.
2) i aborted those tasks... and dived directly into certificate issue. Phone not rooted so I can't just replace affected file then... So, device more powerful than some of my Windows netbooks now out of fleet completely :/

Have you posted to the boinc_dev mailing list to ask the BOINC developers to produce additional updated versions for earlier Windows and Android operating systems that can't install 64-bit 7.16.7?

BOINC Developers Mailing List <boinc_dev@ssl.berkeley.edu>

I'd expect that to be more useful than posting such complaints here.

Or would you prefer to go to boinc_dev and volunteer to perform those updates of BOINC? You might not need to do anything more than recompile them, but include the updated version of the file instead of the old version.

LoL, BOINC should more carefully grant "expert helper"tags then indeed :D

BOINC doesn't grant those, I do. I give them to people on the BOINC forums who selflessly help out anyone with their problems. They're just a friendly title, just like your volunteer developer title is to show newcomers that the person helping them has the knowledge.

There's no need to continue your disdainful attack on these people because not many of them do development work for BOINC.

If you want to attack anyone on this crapshoot, start with the release manager. It's his responsibility to make sure every file in a new BOINC is updated before releasing a new version. It's also his responsibility to release new versions for every bitness available.

So go yell on the boinc_dev email list. Go yell on Github.

But stop attacking the volunteers.

If any Windows user, 64-bit only, is still affected by this, there is a hotfix v7.16.7 of BOINC available from https://boinc.berkeley.edu/download.php

So, running Vista x86 I actually had NO chance to get fix by downloading new client bundle? LoL, BOINC should more carefully grant "expert helper"tags then indeed :D

As a keen reader of the boinc_dev mailing list, you will have seen my two emails:

Once the dust has settled on this emergency action, could you please clarify for the benefit of other developers exactly where we stand in the 7.16 release cycle, especially in relation to the outstanding bugs that Jord mentions.

Is there going to be a refresh of v7.14.2 for 32-bit users?

- both posted on the morning of Monday 1st June.

Neither has yet been answered, either by email or by action, despite being copied to the direct email address of the Release Manager, David Anderson.

As Jord says, there's no point shooting the messenger. Take it direct to the one person who has the keys - installshield licence, code-signing machine - to unlock the backlog.

I know that the current state of BOINC for Android is a mess.

None of the currently available versions have the updated certificates since May 2020

Google Play now require that all newly uploaded apps to Play Store after 1st August 2019 must contain 64bit code.
From 1 August 2021, they may be removing support for 32bit completely !

I think the current Recommended version of BOINC is still 32bit, the Beta version may be 64bit compatible.

There is intense development work going on for the Android app at the moment, via github. I think the new volunteer arrivals (two seem sound, the third less so) are well aware of the specific complexities of developing for Android and the Play Store. I'm basically standing back and watching, because I have no specific Android skills to bring to the table - but they did acknowledge yesterday that I had given helpful advice on the broad direction of travel.

And additionally, plans are afoot to release a new 7.4 or a replacement 7.4.53 with an updated CA-bundle included, while we wait for the new version to be done.

LoL, BOINC should more carefully grant "expert helper"tags then indeed :D

BOINC doesn't grant those, I do. I give them to people on the BOINC forums who selflessly help out anyone with their problems. They're just a friendly title, just like your volunteer developer title is to show newcomers that the person helping them has the knowledge.

There's no need to continue your disdainful attack on these people because not many of them do development work for BOINC.

If you want to attack anyone on this crapshoot, start with the release manager. It's his responsibility to make sure every file in a new BOINC is updated before releasing a new version. It's also his responsibility to release new versions for every bitness available.

So go yell on the boinc_dev email list. Go yell on Github.

But stop attacking the volunteers.

Hm, well, Jord, when help constituted of "better shut up and switch off" suggestion I will exercise my right to freely express my point of view on such "help".
Even if tag granted directly by you :P

And regarding release manager whoever he is (sorry I dont' follow BOINC development as closely as maybe assumed) - in my current understanding the biggest issue here not in the lack of new binary pack for distribution. But in:
1) set-and-forget system suddenly requires user attention
2) users didn't notified about this.

When I have time to come on boards every day this would probably not attract my attention and resutl in strong reaction. But when I haven't I can fully "test"set-and-forget user situation trusting that fleet busy with productive work, keeping devices ON in such believe and then suddenly realising that all that time fleet just wasted electricity and produced heat.

---

SETI apps news
We're not gonna fight them. We're gonna transcend them.

As a keen reader of the boinc_dev mailing list, you will have seen my two emails:

Not so keen or even not so reader currently.

Neither has yet been answered, either by email or by action, despite being copied to the direct email address of the Release Manager, David Anderson.

Now I know the name of that mighty person - Release Manager, thanks :)

As Jord says, there's no point shooting the messenger.

I know this expression but not sure it applicable to this particular case though, but nevermind, perhaps I "sound" more agressive than intended to be, Ok.

Take it direct to the one person who has the keys - installshield licence, code-signing machine - to unlock the backlog.

Still have no clear picture. Why some projects affected and some not? That could mean smth could be done on SERVER side, not client side - and this is my main point.
Does CLIENT side change absolutely required? (and if so, to exclude such outcomes in the future some mechanism should be applied to notify users that action required (Jord said that BOINC notifications not for this, but I see no reason why not to use them in need or why not to provide smth similar in case religion didn't allow such usage ;) ).

---

SETI apps news
We're not gonna fight them. We're gonna transcend them.