The database has gotten hung up a couple of times with slow queries. I've restarted it once and that seems to have helped. It'll take several hours to be sure.

If that didn't solve the problem, it probably means the index the server uses to choose workunits to send has gotten corrupted. If I need to rebuild it, we'll probably be down most of tomorrow while I do so.

---

@SETIEric@qoto.org (Mastodon)

No worries Eric thanks for the update..

As an IT guy myself, I feel your pain. Unfortunately, I think the worst case scenario is playing out. I've been days without being able to get work units, up to, and including, just before writing this message.

Appreciate the update. Good luck.

---

Thanks for the update.
Presently Server Status is green, but many values not updated for 2 hours & still unable to get work- "Project has no tasks available"

Edit- Server Status numbers now updated.
Still no joy getting work, and it's not just me; Server Status In-progress numbers continuing to fall.

---

Grant
Darwin NT

Thank you for the update, Eric.
Looks like things still need 'adjustment'.
Appreciate your efforts last night to try to fix it.
Hope the rebuild goes well, as it appears that it may be necessary.

---

"Time is simply the mechanism that keeps everything from happening all at once."

Thanks for the update, have a good weekend.

---

Member of the People Encouraging Niceness In Society club.

. . Hi Eric,

. . It looks like the worst case scenario indeed. None of my rigs are getting any work and they are now competely out so beddy byes for me and for them :(

Stephen

:(

Thanks for the heads up. As I can burn through 200 work units in a single day, I have activated my backup project so if I run out of work, I will still have something to crunch.

---

My first letter here after 18 years

Sorry, I know you sure work a lot there at SETI. I have been working for over 18 years.
But please: How do you want to recognize and interpret an extraterrestrial signal, if you do not even have the earthly technology under control?

Here is another important request: In BOINC you can secure yourself for SETI under the calculations 10 plus 10 working days, so that in case of a system failure everyone has enough work units. Unfortunately that has never worked! If there is a failure, I can work with my computers just 8 hours, then the work units are used up.
If there is a secret attitude with which one can bypass the Stone Age bunkering of work units, then I ask for tips.

For all users, this is a huge loss of work for SETI. And then the failures that have been going on for years from Tuesday to Wednesday due to maintenance work. If there were at least work units for those two days (real 48 hours), the project would have taken a big step forward and we'd all come closer to ET.

Sorry, but I'm really upset about the many failures for years.

---

any news on the database repair? Hope to get this laptop back to work soon.
hope everybody has a peaceful Easter holiday with family and freinds.

perhaps we need to donate funds to build a server that can handle the work load. consider donating parts . I have a 650 watt power supply collecting dust. I'm willing to donate to this cause.

I'm willing as I'm sure others are to help in any means possible. I understand not everybody has liquid funds to donate. lets get this project going friends.. do what you can......

18+ years of Seti enjoyment. Thanks for the updates and good work, Eric and Team.

---

perhaps we need to donate funds to build a server that can handle the work load. consider donating parts . I have a 650 watt power supply collecting dust. I'm willing to donate to this cause.

$250,000 would go a long way to getting an All Flash Array and 100GB Ethernet backbone & switches.

---

Grant
Darwin NT

There hasn't been any posts about what they spent the donations on from the last fundraiser back in December. Are the funds just sitting idle? I would like to them to use some of that in rebuilding the servers to handle the increased work output from all the crunchers.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

There hasn't been any posts about what they spent the donations on from the last fundraiser back in December. Are the funds just sitting idle? I would like to them to use some of that in rebuilding the servers to handle the increased work output from all the crunchers.

That money was for the Parkes data store & development
Reallocating money that was given for a particular purpose to a different one has killed many a project over the years. Using surplus funds, not so much of a problem. But re-allocating the bulk of funds raised for one purpose to another purpose is a great way of ensuring you never get donations again.

Better to state what the problem is, what is required to sort it out, and have another specific fund raiser.

---

Grant
Darwin NT

I still think it's important to think about raising money for new/extra staff, to spread the load and deal with all those niggling and long-standing glitches. Don't put all your faith in hardware alone.

I still think it's important to think about raising money for new/extra staff, to spread the load and deal with all those niggling and long-standing glitches. Don't put all your faith in hardware alone.

True.
It used to be 1 splitter for each file, and splitter output was very good with only a few splitters. It's been a while since that was the case. The more splitters you have on a single file, she slower the splitting rate tends to be. So it would be nic if that issue was sorted out.
You have to have the staff to support the hardware.

But it doesn't matter how many staff you have, if the hardware isn't up to the job.

Hence a post telling us exactly what the problem is (or most likely problems. The issue (from the outside looking in) is most likely I/O contention within the file servers- even a rack of enterprise HDDs can't match the random or sequential performance of a few high end consumer SSDs, so an AFA would sort that out. Of course to take full advantage of that storage would require new servers with plenty of Cores & plenty of RAM. And of course improved network bandwidth between the systems so the internal network itself doesn't become the bottleneck), what's required to fix it, and keep it fixed for a reasonable period of time.
And it's no good getting something that's affordable, if it's not noting to meet future demands- you just have to raise more money from scratch at that time. Better to save up the money till you can afford what's required to meet current & future needs, or do the upgrades in stages. Replace the current bottleneck, with equipment that will meet our likely needs 10 years from now. When funds are next available, replace the next bottleneck, and so on.

---

Grant
Darwin NT

I agree with the right marketing raising those kinds of funds shouldn't be an issue considering anybody can participate in the project and see their own dollars in effect,.

I know I'm the new guy here, and im not trying to step on anybody's toes. Just having a friendly conversation that I'm sure has come up many times.. I enjoy this kind of stuff. hope you all do as well.