Perfect Storm, Inc. (Jun 16 2010)


log in

Advanced search

Message boards : Technical News : Perfect Storm, Inc. (Jun 16 2010)

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next
Author Message
Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7578
Credit: 6,973,562
RAC: 3,121
United States
Message 1007083 - Posted: 21 Jun 2010, 22:22:27 UTC - in response to Message 1007052.
Last modified: 21 Jun 2010, 22:24:24 UTC

Frankly, and kindly, I do wish that someone at Berkeley would post something short and informative on a daily basis. No blogs, or treatises. Just a simple communication on a more or less daily basis. It would help us all, and help beat back the rioters when rioting commences.


I totally agree
Its very arrogant to leave those of us Volunteers in the dark


I disagree with the assumption that the staff are arrogant-These guys are working their tail off.

I had spoken to Matt awhile back about regular daily updates during outages. They said a couple of things to me--quite plainly. Every time they had posted in the past, people were demanding more and more specifics so no short explanations ever worked. It's extremely difficult to do as some of the updates/work they're doing literally takes all day and the more time they're away from the machines, the less time they're putting in on trying to repair the issues.

Pick your poison---get more updates and less server work time or the other way around.
____________


sheridan
Volunteer tester
Send message
Joined: 25 Nov 00
Posts: 1
Credit: 1,735,583
RAC: 0
United Kingdom
Message 1007085 - Posted: 21 Jun 2010, 22:37:19 UTC

As I'm sure many others here could, I could run SETI better than the current team. With some experience as a project manager it's not so hard - here is the short guide.

1. Several clean sheets of paper to decide how it should work - knowing what is known now, not just what was guessed at when it started.

2. Design the whole thing to work reliably with backups, multi site operation,
disaster recovery and scalability for at least another 10 years.

3. Secure reliable funding for the initial build and ensure ongoing funding for at least another five years so you can...

4. Employ a team of dedicated, capable and enthusiastic people to provide 24/7/365 (I always think that should be 24/7/52 but what the heck) knowing they have a secure and high profile job for at least the next five years.

See - dead easy.

Just a detail though. *IF* anyone does have the money for me to take this project on, you should know one thing. The first people I'd go to - not just for advice but also to offer the jobs to - are the ones already there.

The fact they can do what they have done with so little makes them my first choice to keep doing the job. If anyone else thinks they can do better, then why not move to California (if not already there) and put your skills, dedication and patience to the test. I can't believe the folks there wouldn't welcome the assistance of anyone who actually has the skills required and was reliable and trustworthy.

We live in an imperfect world – deal with it. If you want to know real BOINC frustration, try LHC. Hardly any work units to do and if you miss the few there are, you just slide down the stats. Still, I volunteer and do what I get sent, not gripe that I don’t get enough care and attention from others working hard for the same goal.

____________

jravin
Send message
Joined: 25 Mar 02
Posts: 941
Credit: 102,465,889
RAC: 88,265
United States
Message 1007086 - Posted: 21 Jun 2010, 22:40:07 UTC - in response to Message 1007083.

Frankly, and kindly, I do wish that someone at Berkeley would post something short and informative on a daily basis. No blogs, or treatises. Just a simple communication on a more or less daily basis. It would help us all, and help beat back the rioters when rioting commences.


I totally agree
Its very arrogant to leave those of us Volunteers in the dark


I disagree with the assumption that the staff are arrogant-These guys are working their tail off.

I had spoken to Matt awhile back about regular daily updates during outages. They said a couple of things to me--quite plainly. Every time they had posted in the past, people were demanding more and more specifics so no short explanations ever worked. It's extremely difficult to do as some of the updates/work they're doing literally takes all day and the more time they're away from the machines, the less time they're putting in on trying to repair the issues.

Pick your poison---get more updates and less server work time or the other way around.


Well, then, how about a post when changes are being made on the Production system(s), so we users can at least know why things have gone crazy...this last outage, with the server-side changes - sure would have been nice to know it was coming, and even the reasons for it, and the way the changes were intended to work. (Then we would understand what to look for). Easy to do, if you have written a design document - if you haven't, not so much.
____________

Profile FrostKing9
Avatar
Send message
Joined: 20 Oct 01
Posts: 39
Credit: 23,815,960
RAC: 0
United States
Message 1007091 - Posted: 21 Jun 2010, 22:58:01 UTC
Last modified: 21 Jun 2010, 23:00:10 UTC

All of these great minds hard at work... it's so refreshing to
see so many quarterbacks with no teams to play for, and absent
even one ball between them.

Please do not try attempting to diagnose or to remedy the economy.
____________


I DONATE money to SETI@home.... DO YOU?

I'm just slowly BOINC'ing along.

Hey... ET... you have a sister who likes earthlings?

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7578
Credit: 6,973,562
RAC: 3,121
United States
Message 1007092 - Posted: 21 Jun 2010, 23:00:24 UTC - in response to Message 1007086.
Last modified: 21 Jun 2010, 23:10:24 UTC

I disagree with the assumption that the staff are arrogant-These guys are working their tail off.

I had spoken to Matt awhile back about regular daily updates during outages. They said a couple of things to me--quite plainly. Every time they had posted in the past, people were demanding more and more specifics so no short explanations ever worked. It's extremely difficult to do as some of the updates/work they're doing literally takes all day and the more time they're away from the machines, the less time they're putting in on trying to repair the issues.

Pick your poison---get more updates and less server work time or the other way around.


Well, then, how about a post when changes are being made on the Production system(s), so we users can at least know why things have gone crazy...this last outage, with the server-side changes - sure would have been nice to know it was coming, and even the reasons for it, and the way the changes were intended to work. (Then we would understand what to look for). Easy to do, if you have written a design document - if you haven't, not so much.


Then maybe you should personally try to contact the person who implemented the Boinc changes--DAVID ANDERSON. It is quite well known (historically) that he makes changes without consulting Eric and his staff in any manner-this leaves Eric and his staff to pick up the pieces and the victims of improperly placed blame. Other part of this is that David is equal or above Eric in level so there's nothing Eric can do.

I agree with people that things can be handled better but we need to find a way to make people above Eric (ie DA) acknowledge and accept that.
____________


ront
Send message
Joined: 25 Aug 01
Posts: 77
Credit: 386,336
RAC: 0
United States
Message 1007134 - Posted: 22 Jun 2010, 1:51:36 UTC

I refuse to participate in the nastiness.

I asked a legitimate question as one suggested. If you don't or can't answer the question, please don't bother to respond.

ront
____________

Profile Bill Walker
Avatar
Send message
Joined: 4 Sep 99
Posts: 3400
Credit: 2,148,944
RAC: 2,164
Canada
Message 1007136 - Posted: 22 Jun 2010, 1:58:19 UTC - in response to Message 1007134.
Last modified: 22 Jun 2010, 1:58:47 UTC

ront, the only reliable information is on the project home page. The validator had problems over the weekend, and everything locked up as a result. "Staff" is aware, and working on it.

Almost everything you read on these Forums is speculation (other than posts from "the staff"). Some informed, most not. It should only be considered as entertainment.
____________

ront
Send message
Joined: 25 Aug 01
Posts: 77
Credit: 386,336
RAC: 0
United States
Message 1007138 - Posted: 22 Jun 2010, 2:17:01 UTC

Thank you for your response.

You are quite right and.....generally I do disregard most of what I read.

Perhaps, indeed probably, I share some of the frustration of the others. However, I do realize that beating up on "staff" will avail nothing.

In any event, once again, "staff" has worked their magic, and the system seems to be up and running.

I extend my appreciation and thanks to them...and to you.

ront
____________

Profile Donald L. JohnsonProject donor
Avatar
Send message
Joined: 5 Aug 02
Posts: 6261
Credit: 738,195
RAC: 1,146
United States
Message 1007148 - Posted: 22 Jun 2010, 3:04:39 UTC - in response to Message 1007092.

Well, then, how about a post when changes are being made on the Production system(s), so we users can at least know why things have gone crazy...this last outage, with the server-side changes - sure would have been nice to know it was coming, and even the reasons for it, and the way the changes were intended to work. (Then we would understand what to look for). Easy to do, if you have written a design document - if you haven't, not so much.


Then maybe you should personally try to contact the person who implemented the Boinc changes--DAVID ANDERSON. It is quite well known (historically) that he makes changes without consulting Eric and his staff in any manner-this leaves Eric and his staff to pick up the pieces and the victims of improperly placed blame. Other part of this is that David is equal or above Eric in level so there's nothing Eric can do.

I agree with people that things can be handled better but we need to find a way to make people above Eric (ie DA) acknowledge and accept that.


That's exactly the point I and others made over in the I'm Done Here thread in Number Crunching.

Most of the time, those of us who know where to look (Homepage Notices, Server Status page) know when there is a problem, what might be the cause, and whether or not it is being worked by the Seti@Home staff. We also recognize how much time Over and Above normal working hours Matt, Jeff, and Eric put in to keep this project running (not to mention copious amounts of chewing gum, hook-up wire, and duct tape - or the IT equivalent).

Most of us do not expect minute-by-minute updates during the troubleshooting and repair, especially those of us who have been in similar situations, whether in IT or other lines of work. A simple note, like today's, when they get the chance, will suffice.

But when MAJOR CHANGES are coming, it is not just nice, it is NECESSARY for those of us actually doing the crunching to get a heads-up on it, so we can watch for problems and report them. That is a MANAGEMENT responsibility, and in this case, MANAGEMENT (Dr. Anderson) failed to do his job. Change is much easier to deal with if you know what's coming, than when it jumps up out of the roadside culvert and smacks you upside the head with a club.

Yes, direct communication with Dr. A is called for, just in case he doesn't read these boards (or get a summary from the Moderators or Eric & Matt). Hopefully, learning will take place, and we won't go through this upheaval the next time a major change to BOINC or the server software comes down.

Donald
Infernal Optimist / Submariner, retired


Profile soft^spirit
Avatar
Send message
Joined: 18 May 99
Posts: 6374
Credit: 28,631,148
RAC: 2
United States
Message 1007154 - Posted: 22 Jun 2010, 3:40:45 UTC

I am certain I will not see it.

But what I would really appreciate seeing:

An acknowledgement that they installed untested bad code.

An apology to everyone (staff and volunteers alike) from anyone
involved in that DECISION.

And a promise not to do so again(i.e. TEST IT FIRST!)

Things can still go wrong. It is impossible to see what hardware
needs are if it keeps getting sent bad code. And either cripple a project.

I made the wrong assumption, that after several years seti@home would be past
initial beta testing. Instead I walked into a fresh batch of alpha testing.

This is not what I thought I agreed to do.

I salute those who attempt to maintain the hardware, but unless the incoming code is debugged before system wide implementation, It is fixing the engines on the titanic.

And there are way too many people blowing off those of us aggravated by it.


____________

Janice

Profile edjcox
Avatar
Send message
Joined: 20 May 99
Posts: 68
Credit: 4,024,854
RAC: 1,152
United States
Message 1007155 - Posted: 22 Jun 2010, 4:10:56 UTC

Been around this clubhouse long enough to realize that this effort is at best poorly managed and completely undisciplined in it's operation.

With that said unless and if these boys are wholesale let out to pasture and a professional serious bunch come into place nothing is going to change at SETI.

I would think that configuration management and documented baselines exist so that issues are predictable and manageable. Instead I hear MAtt (the only person diciplined enough to try and keep us informed as to what's happening) complain about changes and ancillary impacts and unforeseen problems all the time.

Long ago and far away I worked on spacetrack radar. We were confounded by jamming and interference and much was done to filter out and remove known signals and unknown interferers. Over 30 years ago... Long before highspeed digital computations etc. Yet SETI is recently touting its ability to filter out interferers and such from its signal analysis as if it's something new, just discovered, etc.

I'm a reluctant partner in SETI as it's the only game being played in the realm. But it doesn't mean I'm happy...

Aw well... rant rant rant... I feel better now never mind...
____________
Never engage stupid people at their level, they then have the home court advantage.....

Profile Dave BarstowProject donor
Send message
Joined: 14 May 99
Posts: 76
Credit: 15,064,044
RAC: 0
Philippines
Message 1007158 - Posted: 22 Jun 2010, 4:20:38 UTC

NOTE: This is addressed TO NO PARTICULAR INDIVIDUAL...

After following these posts for years this is the first time that I have felt compelled to remark on the inane, uninformed, or simply rude commentary coming from some posters.

I am trying to be polite...

If you have a problem with an organization or person, and can not resolve it directly, go over-their-head to someone with the authority to address the issue.

In the current case I would advise that you contact http://boinc.berkeley.edu/anderson/ as a first step.
____________

WinterKnight
Volunteer tester
Send message
Joined: 18 May 99
Posts: 8685
Credit: 24,995,912
RAC: 29,466
United Kingdom
Message 1007182 - Posted: 22 Jun 2010, 5:36:18 UTC - in response to Message 1007158.

NOTE: This is addressed TO NO PARTICULAR INDIVIDUAL...

After following these posts for years this is the first time that I have felt compelled to remark on the inane, uninformed, or simply rude commentary coming from some posters.

I am trying to be polite...

If you have a problem with an organization or person, and can not resolve it directly, go over-their-head to someone with the authority to address the issue.

In the current case I would advise that you contact http://boinc.berkeley.edu/anderson/ as a first step.

I don't think you have quite grasped what the problem is.
The problem(s) have arisen because there has been new BOINC code installed on the Seti servers. It is this new BOINC code that is causing the problems.
As Dr.Anderson is the person responsible and probably the programmer for this new code, certainly CreditNew is his.
Therefore my question to you is who is above Dr.Anderson, and what are the contacts to that person/organisation.

Profile Dave BarstowProject donor
Send message
Joined: 14 May 99
Posts: 76
Credit: 15,064,044
RAC: 0
Philippines
Message 1007201 - Posted: 22 Jun 2010, 7:54:52 UTC - in response to Message 1007182.

The link provided is the U.C. Berkley page for Dr. Andersen, the developer of BOINC.

It is there, as stated, for any who wish to contact him directly.

As others here have mentioned he is the man to contact with any concerns regarding BOINC.

Just trying to aim folks in the right direction...
____________

Stephen Falken
Volunteer tester
Avatar
Send message
Joined: 23 Apr 06
Posts: 63
Credit: 5,159,519
RAC: 0
United States
Message 1007448 - Posted: 23 Jun 2010, 17:10:11 UTC

I have a wonderful idea. Why don't all the critics just donate some money to provide more manpower and equipment and help keep things running smoothly?

KB7RZF
Volunteer tester
Avatar
Send message
Joined: 15 Aug 99
Posts: 9463
Credit: 3,112,738
RAC: 39
United States
Message 1007481 - Posted: 23 Jun 2010, 18:24:18 UTC - in response to Message 1007448.

I have a wonderful idea. Why don't all the critics just donate some money to provide more manpower and equipment and help keep things running smoothly?

Thats the best idea I've heard so far! :-) Still wish I had money to donate to SETI. Damn bills.
____________

Profile Dave BarstowProject donor
Send message
Joined: 14 May 99
Posts: 76
Credit: 15,064,044
RAC: 0
Philippines
Message 1007486 - Posted: 23 Jun 2010, 18:39:05 UTC - in response to Message 1007201.

The link provided is the U.C. Berkley page for Dr. Andersen, the developer of BOINC.

It is there, as stated, for any who wish to contact him directly.

As others here have mentioned he is the man to contact with any concerns regarding BOINC.

Just trying to aim folks in the right direction...


See what happens when I post after 30+ hours without sleep... I spell The Man's name incorrectly! It is Anderson not Andersen. Duh!
____________

Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7578
Credit: 6,973,562
RAC: 3,121
United States
Message 1007501 - Posted: 23 Jun 2010, 19:20:58 UTC - in response to Message 1007448.

I have a wonderful idea. Why don't all the critics just donate some money to provide more manpower and equipment and help keep things running smoothly?


Money is a chunk of the answer but not all...David Anderson will make code changes despite whatever $ is donated.
____________


Profile Blurf
Volunteer tester
Send message
Joined: 2 Sep 06
Posts: 7578
Credit: 6,973,562
RAC: 3,121
United States
Message 1007502 - Posted: 23 Jun 2010, 19:22:34 UTC - in response to Message 1007155.

With that said unless and if these boys are wholesale let out to pasture and a professional serious bunch come into place nothing is going to change at SETI.


Again placing the blame on those not responsible for the situation, not helpful and not appropriate.
____________


Stephen Falken
Volunteer tester
Avatar
Send message
Joined: 23 Apr 06
Posts: 63
Credit: 5,159,519
RAC: 0
United States
Message 1007514 - Posted: 23 Jun 2010, 20:04:36 UTC - in response to Message 1007501.

Perhaps, but more staff will certainly help react with greater expediency.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next

Message boards : Technical News : Perfect Storm, Inc. (Jun 16 2010)

Copyright © 2014 University of California