Hello all! We are glad to welcome Formula BOINC in our project. Since the last Sprint, we changed several parameters that should improve server's work. In case you face any difficulties when sending a results report, try to use the option max_tasks_reported: https://rake.boincfast.ru/rakesearch/forum_thread.php?id=93&postid=458#458

Server doesn't seem to be able to handle upload and download of Work Units. Can anything be done to help handle this Sprint?

Hello Natalia
As each year the same problems. And it will again happen on both ODLK.
Not your fault. The most know you do your best..
We know Rake, OGLK and ODLK1 kan not handle so much traffic load during Sprint, due to not very powerfull server.
But someone on SETIBZH stay pushing, without consideration.
Can you imagine if all your projects gives as much credits as Collatz ?
...
....
Best regards to you and Hoarfrost

Phil

---

Hi folks!

During the phase of tasks flushing, a number of requests in many times higher than average by sprint. My computers also faced with problems but at ~22:00 UTC situation looks normal (except website pages, of course).
Does anyone have problems now?

Thank you for participation! A rate of tasks processing is amazing!

There are still lots of failing file transfers (transient HTTP errors). This can cause computers to run dry, as BOINC does not request new tasks from a project with stalled downloads or too many uploads in progress.

On the other hand, there doesn't appear to be a significant validator backlog and the task generation is much faster than the tasks are sent out. You might even consider slowing task generation down a bit, as all those extra tasks only inflate the database.

All in all, the situation is far from perfect, but I have seen worse things. When PrimeGrid started their Challenge Series back in 2008, it took a few tries until they had figured out which parameters to adjust in which way in order to keep their server alive and their participants happy. And, unlike you, they knew about the challenge well in advance, so they had a chance to open some glass doors instead of just picking up the pieces when it is already too late.

Edit: Sorry for posting this multiple times. I got "504 timeout" errors and retried without checking if it was posted.

Hihi, Mr. Pentadingens hat einen nervösen Zeigefinger (;-)))
Freu' mich schon auf Anfang Mai.
Gruss

All was working here until this mess "chalenges" started... **********
Detto bunkerthlon .. where all "robin hoods" fraudly use not own computers ,not pay electricity, just stealing from taxpayers ,, And i have compete with someone who have hundreds computers from school/company/state institition or other similar?? LOL of course no.)) And if robin hoods aka hide computers do bunkering on this hundreds/ thousand machines ,what i see they do last years..bunkering going to crazy level! WE- real suporters dont have chance!
whole concept of challenges on boinc is bad/old ..also formula chose one project and 10K users start count on one project ..of course it will collapse.. just think omg!

Load during the sprint (especially - in a phase of tasks flushing) in many times higher, than in "usual peaks".
Simple arithmetics - in last days before sprint project server received and processed about 3 ... 5 per second and generate ~1 .. 2 canonical results.
After sprint start, about 220 000 tasks hit to the server simultaneously. Really physical limits stretched out this wave, but load really in tens times higher than usual peaks.

But I like Formula BOINC Sprints because they flash bottlenecks. During the previous sprint main bottleneck was an I/O on hard drives. After - we increase server DRAM, MySQL caches, change data archiving strategy and now I/O not limit the performance. Most operations performed in file system DRAM cache. Network iterations as I see now is the main bottleneck. Several minutes ago we change some settings of apache and now monitor the situation. Really we monitor the server most of the day.

And great thanks for all crunchers for support and attention!

hey i just want to thank everyone at the RakeSearch Project and also all the support people/

Being hit by the swarm in a Sprint is a little mind boggling,the stress and other things that you just didn't count on.

I just want to say thanks for all your time and effort.....

Top Notch Job!

-steve
SystemViper

Hello @jozef j
You should be aware that in the wider BOINC community there are many institutions that make their systems, hundreds of them, available for distributed computing projects as part of their own community engagement policies. Some organisations allow their systems to be controlled by one or more of their system administrators who can choose which projects they participate in. while others choose a single project and stay with it.
There are also a number of people who have more money than space, so they rent system resources in the form of virtual cloud instances, sometimes very large numbers of them at any given time.
While some feel neither situation is in the "spirit" of the challenges, both are participating legally and honestly, neither are the thieves you have accused them of being.

Making use of system resources without proper authorisation is referred to as "borg'ing", named after The Borg, a race of Star Trek aliens. It is strongly discouraged across the wider crunching and folding (Folding at Home project participants) communities and has resulted in people being both banned from projects entirely and the police being involved.

Please do not call people thieves just because they are capable of utilising more resources than you.

Yes, to the point that my munin instance is unable to grab the server status xml from your server, it keeps timing out.

Seems that I can fetch the data :-)

Kiska's site is showing periods with good contact to server status each night between about 01:00 UTC and 07:00 UTC.

It's been like this for at least half a day now. Maybe longer. My machines are idle, with no work.

123995	Rake search of diagonal Latin squares	4/14/2019 7:16:28 AM	[sched_op] Starting scheduler request	
123996	Rake search of diagonal Latin squares	4/14/2019 7:16:28 AM	Sending scheduler request: To fetch work.	
123997	Rake search of diagonal Latin squares	4/14/2019 7:16:28 AM	Requesting new tasks for CPU	
123998	Rake search of diagonal Latin squares	4/14/2019 7:16:28 AM	[sched_op] CPU work request: 5727297.44 seconds; 56.00 devices	
123999	Rake search of diagonal Latin squares	4/14/2019 7:16:28 AM	[sched_op] AMD/ATI GPU work request: 0.00 seconds; 0.00 devices	
124000	Rake search of diagonal Latin squares	4/14/2019 7:17:30 AM	Scheduler request failed: HTTP gateway timeout	
124001	Rake search of diagonal Latin squares	4/14/2019 7:17:30 AM	[sched_op] Deferring communication for 00:01:06	
124002	Rake search of diagonal Latin squares	4/14/2019 7:17:30 AM	[sched_op] Reason: Scheduler request failed	

---

Dublin, California
Team: SETI.USA

The server is now reporting it is out of space when trying to upload :(

I don't think one has to be a rocket scientist to realise that the Formula BOINC event breaks things here at RakeSearch.
The organisers of said event should pick projects that can handle the load.

Hello @jozef j
I understand your frustrations, but sometime things happen.
I totally agreed with Dark Angel. Nowadays, with more and more Universities, institutions, are aware and getting involve with Distributing Computing.
Whether we, as individual contributors, like it or not or feels like they're not legit ... it's still good for science. It's still better than they'd use those computers to mine crypto. Would you agree?
Believe it or not, there are people out there very dedicated to DC. I happened to know a gentleman that has well over 30 machines at his house. His electric bill can probably build a 1950X Threadripper per month. While he doesn't spend money on new hardware, he'd go around town and picked up machines that people threw away and put them to work.
I, myself, have no where near the amount of machines he has.But in the top 10 computers here at Rake, 4 of them are mine. That doesn't mean I stole resources like you said. I also haven't had a vacation for the last 5 years. All the money went to hardware and electric bills. That's the sacrifices I had to make in order to have the output that I have.

So Please do not call people thieves because it's not fair for those who dedicate their time and money to contribute.

Not the best Run this year, no tasks and if I got any it was on 1 or 2 at a time.

And Lots of
504 Gateway Time-out
nginx/1.15.7

Not the best Run this year, no tasks and if I got any it was on 1 or 2 at a time.

And Lots of
504 Gateway Time-out
nginx/1.15.7

Same here.

---

The organizers, due to arrogance and ambitions, got involved in the competition without having any opportunities. It is low, throwing soldiers with sticks under the tanks.