but boy, are those servers going to get a hammering tonight!

From the way they were acting last evening and into today, I'd think they "got hammered" yesterday morning/afternoon. LOL

I presume this will all be sorted out about 5.00 pm UTC, today, when the servers come down for the weekly planned DB maintenance?

That will not be too long to wait!

If they had fixed it 9 hours ago, we wouldn't have had to wait at all. That's my point. Instead, since they don't care, it'll be 16+ hours down and then whatever the recovery time is.

So?

So, I find it rather insulting that they care so little about those of US that do the actual crunching FOR them.

So!!

Nothing to do with not caring, something to do with a life outside the SSL at Berkeley!

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

---

[I wondered how that worked. My new box wont get anything until they fix the problem, not even the Seti client. The other box that wont cache according to the settings has over 118 Wu's done and, of course, more than 20 of those are validated. I do hit EDF issues on my SMP boxes and the dual core CPU's but I never thought of that as a problem. I suppose there's nothing that can be done about EDF either?

Check you Boinc folder and look to see if you have a Global_prefs file it is a generic file your preferences are stored in Boinc has a habit lately of not creating that file.
If you don't have it you can copy and paste it from a system that does it would have the same settings as your Accounts preferences page.

I posted about this problem awhile ago and the solution as well.
http://setiathome.berkeley.edu/forum_thread.php?id=34722

Dudes!!! For a bunch of computer geeks and IT admins, you guys are wound up too tight!

Maybe its a really big problem and they have been working on it non stop, and just can't get things to work......

Did ya think about that?

[I wondered how that worked. My new box wont get anything until they fix the problem, not even the Seti client. The other box that wont cache according to the settings has over 118 Wu's done and, of course, more than 20 of those are validated. I do hit EDF issues on my SMP boxes and the dual core CPU's but I never thought of that as a problem. I suppose there's nothing that can be done about EDF either?

Check you Boinc folder and look to see if you have a Global_prefs file it is a generic file your preferences are stored in Boinc has a habit lately of not creating that file.
If you don't have it you can copy and paste it from a system that does it would have the same settings as your Accounts preferences page.

I posted about this problem awhile ago and the solution as well.
http://setiathome.berkeley.edu/forum_thread.php?id=34722

The file already exists, now what?

Dudes!!! For a bunch of computer geeks and IT admins, you guys are wound up too tight!

Maybe its a really big problem and they have been working on it non stop, and just can't get things to work......

Did ya think about that?

I did think of that but that's not what happened.

---

I presume this will all be sorted out about 5.00 pm UTC, today, when the servers come down for the weekly planned DB maintenance?

That will not be too long to wait!

If they had fixed it 9 hours ago, we wouldn't have had to wait at all. That's my point. Instead, since they don't care, it'll be 16+ hours down and then whatever the recovery time is.

So?

So, I find it rather insulting that they care so little about those of US that do the actual crunching FOR them.

So!!

Nothing to do with not caring, something to do with a life outside the SSL at Berkeley!

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

There ARE other projects to crunch for, if this one is sooo bad. NO computer will run forever without some kind of crash. These crashed in the middle of the nite. No one there, when they got in a decision was made to go ahead with the daily outage and put a note on the home page, it is there for all to read.
No money and being run at a University that is being run like a 'for profit' organization means anything happens, somebody must pay for any outside help. Seti has no money! You can tell that by the lack of green stars by everyones name. Those that have green stars contributed CASH to Seti, those that do not have the stars, either chose to not display the fact they donated, or did not donate. Nothing that is based on computers can be run for long with no money. Volunteers can only be run thru the mill so often and for so long.

---

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

Wow, so what you say is they should have someone sitting, monitoring 24/7 just so you can have work? I don't think so. These people work their rear ends off trying to keep things running. Stuff happens. Its a computer system for crying out loud. These folks are running on such a tight budget that can only stretch so far. Excuse them that they are tired and need sleep occationally, and not sitting in front of the server's just waiting for something to break down.

---

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

Is it possible your expectations are too high for a volunteer project? I'm sure you've stayed up late to fix problems, and I'm sure you're compensated well for it. Can you say the same for them? Somehow, I doubt they are.

Besides, what's the bigger picture here? Life goes on, right? They will get it fixed eventually, right? It means very little to your life - it's not like you're losing money, time, sleep, etc. just because your machine isn't crunching SETI! Somehow I don't feel so "abused" or "neglected" by them. If others do, then I suggest they look at the bigger picture overall in their lives.

They key thing to remember here is that it's all volunteer. There is nothing personal about this. There is no neglect going on, on anyone's behalf. If I offer my time at a homeless shelter and they're closed due to problems, do I sit there and complain because I can't offer my help there? Do I complain that the homeless can't eat? Or do I not allow a problem to stop me and move on to the next thing?

If everyone stops to complain about something, that is when there's a true loss of productivity. You can find your own solutions/alternatives instead of crying as loud as you can just to be heard - and try to reinforce your own complaints by showing other unhappy people that are just as incapable of moving on.

---

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

Wow, so what you say is they should have someone sitting, monitoring 24/7 just so you can have work? I don't think so. These people work their rear ends off trying to keep things running. Stuff happens. Its a computer system for crying out loud. These folks are running on such a tight budget that can only stretch so far. Excuse them that they are tired and need sleep occationally, and not sitting in front of the server's just waiting for something to break down.

No, what *I* said was that somebody should have fixed it *before* going to bed or got up and fixed it last night instead of making 200,000 people wait for them. Try not to put words in my mouth, you're not very good at it.

Based on results, they clearly care less about us crunchers than sleep or whatever they were doing for the last 20+ hours. As an IT Manager, I've had to fix things in the middle of the night and I didn't have 200,000 people waiting either. They've demonstrated how important we are to them by making us wait so they could sleep. IMO, it's lazyness or a serious lack of respect.

Is it possible your expectations are too high for a volunteer project? I'm sure you've stayed up late to fix problems, and I'm sure you're compensated well for it. Can you say the same for them? Somehow, I doubt they are.

It's possible but I think not. While it is volunteer we don't get paid to crunch either. If they want us to do their work, they should be bending over backwards to make sure we have workunits to crunch. If *I* were begging for CPU cycles I darn sure wouldn't keep my 200,000 volunteers with their 350,000 hosts waiting because I was too lazy to fix it right away. *I* would have fixed it at 5 pm, when it went down, or drove my rear BACK to Berkeley to fix it that night instead. IMO, since they don't respect us enough to do that, it shows how important we are, to them.

It's possible but I think not. While it is volunteer we don't get paid to crunch either. If they want us to do their work, they should be bending over backwards to make sure we have workunits to crunch. If *I* were begging for CPU cycles I darn sure wouldn't keep my 200,000 volunteers with their 350,000 hosts waiting because I was too lazy to fix it right away. *I* would have fixed it at 5 pm, when it went down, or drove my rear BACK to Berkeley to fix it that night instead. IMO, since they don't respect us enough to do that, it shows how important we are, to them.

I don't see it that way. I don't think they should be "bending over backwards" for us. They do what they can. They're not begging for workunits, only making themselves known. Is advertising considered begging?

It's great that you feel you would have fixed it, but that would have been of your own volition. But since you're not working at Berkeley and you didn't fix it, then why worry about it? Why hold them to such high standards when it's all volunteer anyway?

I'm sorry, but I don't feel it's a lack of respect. I think you're reading way too much into the issue. They had an issue with the servers. They didn't fix it right away, for their own reasons that we can only speculate. It doesn't really affect your life in a major way, so why insist you're feeling neglected over it? You weren't hurt in any way; your CPU simply went idle for a while. It picked right back up when they did fix it. The data will eventually get crunched.

I don't think it's a very big deal at all, and I'm having problems understanding why people make such a large, personal issue out of it.

---

There's a 4th option: uninstall boinc and stop helping seti since they seem not to care.

Complaining does serve a purpose, it's lets the "powers that be" know the natives are restless. You don't expect nothing but "yay good job" posts, do you? Are you trying to stiffle any debate over the lack of service and respect for us crunchers? Are we not allowed to complain here?

Oh, geez.

Anything can follow from a false premise.

SETI is staffed 8 hours a day, 5 days a week. That's what the money will pay for -- SETI is historically cash-strapped.

Because the SETI staff does care, they are often checking the servers from home during the night (for those who tend to be night-owls) and in the morning (for those who tend to be early risers) so SETI is only on autopilot about 5 hours per day.

They're checking things on the weekend, even though they aren't "staffed" on the weekend (unless Matt's 5 day workday still includes Sunday).

We need to remember that while this is a passion for many, for the folks who actually work at Berkeley it is still a job. They have other things to do, they have families, and hobbies and you just can't do this all day.

So they've designed software (BOINC) that can work, unattended, through most outages, and as a general observation, it works pretty well. When uploads and/or downloads stop, SETI keeps working off the cache.

It's much better than classic, where the servers needed to be up all the time (or you had to run some add-on to do the caching).

A cache of 7 days is more likely to run out than a cache of 4 days, due to the deadlines for shorter work units, you might want to experiment and find the best value.

If you want SETI to be staffed more, and if you want them to have better equipment that doesn't require so much maintenance, or have redundant servers, then maybe it's time to donate.

---

staffed.....giggle chuckle...
Someone.....hehehe

Matt Lebofsky is THE person who does this. It's not like there's a dozen employees available. And if you'll forgive me, I've seen Matt, Rom, David A working all hours of the nights and on weekends. I can show you copies of emails exchanged while troubleshooting at 11PM Pacific, on Sundays, Saturdays, and at all manner of strange hours.

Matt Lebofsky doesn't make alot and has a second job to make ends meet. Since the start of boinc/seti I've witness outages lasting 3 weeks, 2 weeks, one week and even some short ones lasting 3 days. A 20 hour outage is not even a hiccup. If the participants would make use of the tools given (connect to setting) they wouldn't even be affected by this.

There aren't ANY scientists waiting for your result to finish their work. It (your result) is stored in the MSDB (master science database) until their ready to study them. So, whether it sits on their servers or your computer makes little difference.

Seti has never been good at posting up notices like "hey it's broke, we're working on it", but guess what???? The very minute a problem is detected by those best suited to be the first to know (that's US), there will be atleast ONE thread about it, if not 1/2 a dozen as witnessed here the last day. So, If you just look at the first 3 threads in the NC forum, you'll have the latest head up.

They have something like 7 employees and that includes the boinc employees. I know of one Post grad (or equiv) working at seti/beta/astropulse. Given this small number Boinc AND Seti rely on the volunteer community to give support and to fill in the empty spots. Places like manning the Q@A boards, helping users in these fora, There are a few Volunteer Developers who write the code, even the Boinc.php is maintained by a volunteer.

It's possible but I think not. While it is volunteer we don't get paid to crunch either. If they want us to do their work, they should be bending over backwards to make sure we have workunits to crunch. If *I* were begging for CPU cycles I darn sure wouldn't keep my 200,000 volunteers with their 350,000 hosts waiting because I was too lazy to fix it right away. *I* would have fixed it at 5 pm, when it went down, or drove my rear BACK to Berkeley to fix it that night instead. IMO, since they don't respect us enough to do that, it shows how important we are, to them.

I don't see it that way. I don't think they should be "bending over backwards" for us. They do what they can. They're not begging for workunits, only making themselves known. Is advertising considered begging?

It's great that you feel you would have fixed it, but that would have been of your own volition. But since you're not working at Berkeley and you didn't fix it, then why worry about it? Why hold them to such high standards when it's all volunteer anyway?

I'm sorry, but I don't feel it's a lack of respect. I think you're reading way too much into the issue. They had an issue with the servers. They didn't fix it right away, for their own reasons that we can only speculate. It doesn't really affect your life in a major way, so why insist you're feeling neglected over it? You weren't hurt in any way; your CPU simply went idle for a while. It picked right back up when they did fix it. The data will eventually get crunched.

I don't think it's a very big deal at all, and I'm having problems understanding why people make such a large, personal issue out of it.

I meant "begging" figuratively not literally. We all know they're not on their knees pleading but they do ask that we donate our cpu cycles for free. In return we shouldn't have to wait for them to provide the workunits they want us to crunch.

As I've said, if *I* had a project that required donated cpu cycles, I'd be more respectful of my volunteers. I'd care. That seti doesn't feel the same way, to me, shows a lack of respect.

Some of us have more self-respect than others. *shrug*

There's a 4th option: uninstall boinc and stop helping seti since they seem not to care.

Complaining does serve a purpose, it's lets the "powers that be" know the natives are restless. You don't expect nothing but "yay good job" posts, do you? Are you trying to stiffle any debate over the lack of service and respect for us crunchers? Are we not allowed to complain here?

Oh, geez.

Anything can follow from a false premise.

SETI is staffed 8 hours a day, 5 days a week. That's what the money will pay for -- SETI is historically cash-strapped.

Because the SETI staff does care, they are often checking the servers from home during the night (for those who tend to be night-owls) and in the morning (for those who tend to be early risers) so SETI is only on autopilot about 5 hours per day.

Somebody agreed to keep the network and servers up, no matter what they get paid, if at all. To me, as an IT professional, that means keeping things working, even if I have to stay a little late or get out of bed at 2 am, so my 200,000 volunteers and their 350,000 hosts can keep on keeping on. Nobody put a gun to the admins heads, they took the responsibility and then, imo, let us crunchers wait becaues we weren't important enough to do something different.

Matt Lebofsky is THE person who does this. It's not like there's a dozen employees available. And if you'll forgive me, I've seen Matt, Rom, David A working all hours of the nights and on weekends. I can show you copies of emails exchanged while troubleshooting at 11PM Pacific, on Sundays, Saturdays, and at all manner of strange hours.

Matt Lebofsky doesn't make alot and has a second job to make ends meet. Since the start of boinc/seti I've witness outages lasting 3 weeks, 2 weeks, one week and even some short ones lasting 3 days. A 20 hour outage is not even a hiccup. If the participants would make use of the tools given (connect to setting) they wouldn't even be affected by this.

They have something like 7 employees and that includes the boinc employees. I know of one Post grad (or equiv) working at seti/beta/astropulse. Given this small number Boinc AND Seti rely on the volunteer community to give support and to fill in the empty spots. Places like manning the Q@A boards, helping users in these fora, There are a few Volunteer Developers who write the code, even the Boinc.php is maintained by a volunteer.

As I said above, they took the job and the responsibility no matter how little, if at all, they're paid so that means nothing to me.

Regardless of settings, mine was set to 7 days, the 20+ hour outage affected a whole boat load of clients. Most users don't know enough to change settings so most clients are set to default. Eventhough mine was set to 7 days it still affected me because of reasons already stated. It would have shown appreciation had he/they fixed it right away as you've admitted they sometimes do. In this case they blew US off to do whatever it was they did from the time it went down until they got back in the office.

Maybe it's too much to expect that network and server admins do their jobs and keep things running smoothly for us volunteers without waiting nearly a whole day, but again, I think not. I'm not saying *everyone* has to feel the same way that I do but I feel I'm right in my expecations. That's my story and I'm sticking to it.

---

I vote that all the folks at UCB that run the SETI@HOME project have to give up all their personal lives and sleep in the server closet waiting for it to break down so they can jump at our every whim and fix it immediately apon being notified.

---

I vote that all the folks at UCB that run the SETI@HOME project have to give up all their personal lives and sleep in the server closet waiting for it to break down so they can jump at our every whim and fix it immediately apon being notified.

If that's how you feel, vote the way you see fit. :D

---

I vote that all the folks at UCB that run the SETI@HOME project have to give up all their personal lives and sleep in the server closet waiting for it to break down so they can jump at our every whim and fix it immediately apon being notified.

If that's how you feel, vote the way you see fit. :D

It was a sarcastic reply to the outragous requests your demanding on the boards.

---

While the lack of information flowing from the administrators down to us, the crunchers, is frustrating there is nothing we can do about it. This situation has existed from the very early day's of Seti with no improvement in sight. Seems that nothing is going to change this.

As I stated in another posting they have been known to come into the server closet on weekends and even federal holiday's to kick the boxes. That they don't come in to kick things on each and every failure is not surprising. Boinc is well designed to recover from network failures and does so. Running a cache of several day's or a backup project is also an option to bridge the outages. It is permisible for the project administrators to have a life outside of the server closet. Boinc/Seti is NOT a critical level endeavor. Nothing is lost during a short outage.

In short.....why worry about things that you have no direct control over? Worry about those things you DO have direct control over. Life, family and health.

---

Boinc....Boinc....Boinc....Boinc....

Holeyman, the only one being disrespectful here is you.

The folks who maintain SETI are hard-working and dedicated. When they've needed to put in the hours to get things working again, they've done it.

Don't you find it just a little odd that of the hundreds of thousands of
users who were "inconvenienced", YOU are the only one whining and bitching
about how undedicated the SETI staff are?

If you've just GOT to complain, can't you at least get your facts clear before shouting from the rooftops? As has already been pointed out, there are NOT "millions" of users - not even millions of boxes. There are "only" 553,000 active users and about 1.2 million active boxes.

The vast majority of those users and boxes were not inconvenienced at all; many because they keep a reasonable size cache of work to carry them through small hiccups (or even medium outtages); many were not inconvenienced because they aren't compulsive crunch-watchers and aren't bothered by a few hours of idle CPU (or have alternate projects to work on).

So you had to wait a few hours to get your overclocked box to do some work?

So what?

How were you harmed by waiting a few hundred extra minutes?

What have YOU done to improve SETI? Have you donated money? Have you donated expertise to improve the system? When's the last time you dropped by the shop and helped fixed a broken system?

You've been a member since 1999, and you've crunched a fair number of SETI credits - and your ONLY posts to this board seem to be the whining in this thread.

Looking at your team rankings, it's obvious that missing a day of crunchig won't even affact your standings at all - even if you were the only one without units to crunch.

So - why all the angst?

Chill out, relax, get your box crunching again.

Then come back and apologize to the SETI staff.

Roy

Holeyman, the only one being disrespectful here is you.

The folks who maintain SETI are hard-working and dedicated. When they've needed to put in the hours to get things working again, they've done it.

Don't you find it just a little odd that of the hundreds of thousands of
users who were "inconvenienced", YOU are the only one whining and bitching
about how undedicated the SETI staff are?

[...]

So - why all the angst?

Chill out, relax, get your box crunching again.

Then come back and apologize to the SETI staff.

I'll add my agreement to that.

And as for the supposed lack of communication... Well, for those familiar with the project, a few good guesses can fill in the blanks until Matt has a chance to say what happened.

Also note that he does have to eat and sleep or even go out for a beer sometimes! Boinc has been carefully designed so that he can be IT administrator for a million or so participants and still be able to sleep easy.


It's also good to see the very reasonable tone of all the threads rather than the competitive angst of days long ago over racing and hustling the credits. No cheerleading needed! :-)


Normal operation will be resumed sometime. (Whatever normal might be!)

Happy crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Shamelessly stolen from another thread:

Posted from Here.

November 14, 2006 - 23:45 UTC
So the download server stopped sending out work yesterday afternoon, and it didn't get caught and fixed until this morning. We are in the process of major network upgrades (stay tuned for news on that). A configuration change made for testing the new plumbing tickled a creeping problem where long, cryptic symbolic link chains break unexpectedly. In other words: oops.

So, a simple ooops indeed! (And very easily done.)

No big issue here folks,
now just keep crunching along...

(And all within the normal Boinc design parameters.)

Happy crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)