Validate errors is becoming a rule, day by day.
Once sometime is possible; more is diabolic.
Please investigate and solve the problem.
ASAP.
Or we'll find other projects.

Bye
Franz

---

Hey give the guys a break, it's a weekend for one thing. They have extremely limited funding, problems happen. Some things cannot be avoided.

Relax, it's not as though the world is going to end tomorrow - we live in hope.

Hey give the guys a break, it's a weekend for one thing. They have extremely limited funding, problems happen. Some things cannot be avoided.

Relax, it's not as though the world is going to end tomorrow - we live in hope.

If one believe in a project, he dedicate max effort.
And must be respected.
Both. We, cruncer also.

Bye,
Franz

---

They have asked you to find other projects, in the news sections. They are not in the business to be exclusive.

I also find it interesting that "I" am not having any of these validation issues everyone else seems to be having.

I still am curious as to if it's from too much micromanaging. Forcing uploads when you do not need to, forcing reporting when you do not need to, etc. I leave mine alone, they will upload and update when they are able, and takes the stress off the servers.

---

My movie https://vimeo.com/manage/videos/502242

Hey give the guys a break, it's a weekend for one thing. They have extremely limited funding, problems happen. Some things cannot be avoided.

Relax, it's not as though the world is going to end tomorrow - we live in hope.

If one believe in a project, he dedicate max effort.
And must be respected.
Both. We, cruncer also.

Bye,
Franz

Part of that respect is submitting work which can be validated. If the work can't be validated for legit reasons, then your off the hook!

JeffL

---

I am sure that nobody would like to have this monkey off their back more than Eric. And I am equally sure that he is doing all he can within the constraints of the money and personnel he has available to track down and rectify the problem.

And the kitties say..."Patience is a virtue, my friends".

---

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

Just try to relax and find something else to occupy your time right now. If it's not crunching other projects (if you have run out of SETI work), then find a good book to read or something. Sorry, have to run. My XBox 360 is calling me.

---

http://www.setiusa.net

The problems i've had with results last weekend were my own fault optimizing the apps, nothing to do with the work units from seti. Yesterday i filled the cache on my laptop with over 200 units, no problems so far.

Edit, i have had to errors - each costing me 0.01 credit, so totally destroying my RAC. I am extremely gutted about this.

Sarcasm floating around somewhere.

Same here. I have seen no errors. I have a couple of WUs in the Transfer tab and they wont try to upload again for another 1.5 hrs. It will happen when it happens. It's a long weekend in the USA and we all know the Seti servers don't like to let the crew take 3 days off. They need and demand constant attention or they have a fit.

With 15 projects I'd end up with a coronary if I worried about every little hiccups.

Currently Predictor is MIA, Simaps and Tanpaku like to clutter my log with no work available message before giving me some, LHC only has work for a day at a time and goes weeks between batch's, Spinhenge give errors on almost every up/download before succeeding, and Einstein's Eveready bunny seem to have finally faltered for a couple of weeks before stabilizing again. The other projects just do their thing up/down whatever, I don’t remember the last time I checked the results to see if I was actually contributing anything. I just checked my signature and it looks like they must be doing something right since there RAC appears to be at a reasonable level.

Worrying only shortens your life span. Personably, I intend to be in the Guinness record books as the oldest living human being in another 60+ years.

---

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470

If you think this is bad, well not really...Einstein has had some knarlie problems lately as well if I'm not mistaken. Eventually things get fixed.

---

Account frozen...

The work is not wasted, it helps in finding out what the problem is and it helps in developing a fail safe system because if you dont know what the faults are you can't make it fail safe.

BoB

---

Do you Good Search for Seti@Home? http://www.goodsearch.com/?charityid=888957
Or Good Shop? http://www.goodshop.com/?charityid=888957

Currently Predictor is MIA, Simaps and Tanpaku like to clutter my log with no work available message before giving me some

SIMAP has finished the inital caclulations of the protein sequences (SIMAP) and protein domains (HMMER).
Every 1st every month there will be a new batch of SIMAP WUs and every 15th a new batch of HMMER WUs calculated.

---

Validate errors is becoming a rule, day by day.
Once sometime is possible; more is diabolic.
Please investigate and solve the problem.
ASAP.
Or we'll find other projects.

There was a lot of postings, which recommends to disable the network access and not to upload the results.

If one believe in a project, he dedicate max effort.
And must be respected.
Both. We, cruncer also.

Sorry, this has nothing to do with missing respect.
Did it harm, if you upload or download your WUs today or tomorrow ? - Downloads was still posible. So I did not see any problem.

---

Validate errors is becoming a rule, day by day.
Once sometime is possible; more is diabolic.
Please investigate and solve the problem.
ASAP.
Or we'll find other projects.

Bye
Franz

I'm sure they are trying to fix this ASAP, to be honest. no one is wasting anyone'sCPU time; its all part of the learning curve.

I've lost a few units through validate errors, but I reckon that was 'cos I didn't suspend my network quickly enough.

Totally agree with Pooh Bear - just let it run its course; the update button doesn't exist! BOINC'll sort it in the long run. Personally, I've suspended network for the foreseeable future as the guys at Berkeley have enough to cope with without my boxes mithering for uploads/downloads.

Give it a couple of days and all will be well.

Until the next time ... ;-)

Keep it up folks at Berkeley - we know you're working your respective behinds off, and it is much appreciated.

And the obvious line is to go on about donations for better equipment. I'll confess to not being the biggest donator ;and this month will be the same - but if we're getting all annoyed about stuff like this; give 'em some cash if you can!

Steve.

---

I still am curious as to if it's from too much micromanaging. Forcing uploads when you do not need to, forcing reporting when you do not need to, etc. I leave mine alone, they will upload and update when they are able, and takes the stress off the servers.

Although that can’t do the overall server-health situation any good, I very much doubt that the validate errors are the users’ fault in any way. I’ve never seen them occur except when there are problems at Berkeley. I certainly don’t micromanage my systems in that way—indeed, several of them run essentially unattended—but I’ve been getting little runs of the errors, associated with several of the outages of the past few weeks, from a few different hosts.

---

The message this morning is telling the story, Matt posted about the web server load exceeding the capacity on the 25th of December Eric told more this morning. It is one of those Operations Budget items...

Upgrade to failure tolerant server configuration $13,300

The thread I created for the validate errors has been forwarded to Eric if you look you see the splitter on kryten is turned off to hopefully provide a few more resources. So identifying the earliest occurrence on the next event can help to track what was happening in the logs. Perhaps a bit more relief could be found or what is causing the issue.

Pappa

---

Please consider a Donation to the Seti Project.

Validate errors is becoming a rule, day by day.
Once sometime is possible; more is diabolic.
Please investigate and solve the problem.
ASAP.
Or we'll find other projects.

There was a lot of postings, which recommends to disable the network access and not to upload the results.

If one believe in a project, he dedicate max effort.
And must be respected.
Both. We, cruncer also.

Sorry, this has nothing to do with missing respect.
Did it harm, if you upload or download your WUs today or tomorrow ? - Downloads was still posible. So I did not see any problem.

1) it's not easy to disable network if you have 20 Host to manage, in different locations.
For me this is not a Job and I can't live on PC looking at validate errors.

2) I can stop communication for a week (having a long work buffer) but what happens if I open my comunications in the exact moment of a "validate" problem? I loose a week of job of 20 hosts!

So I don't think that your suggestions are good suggestions.

If the validate problems are once a month, ok, I can agree with you non seeing problems.
But I can see 4 in a week and this for me is TOO MUCH!

Bye,
Francesco

---

1) it's not easy to disable network if you have 20 Host to manage, in different locations.
For me this is not a Job and I can't live on PC looking at validate errors.

You can connect with the BOINC manager or boinc_cmd to the different clients. Also a shell skript or RDP could do this job.

2) I can stop communication for a week (having a long work buffer) but what happens if I open my comunications in the exact moment of a "validate" problem? I loose a week of job of 20 hosts!

Could happen. But, relax, let BOINC the things do. Don't micro manage it and did not care about every single credit.

If the validate problems are once a month, ok, I can agree with you non seeing problems.
But I can see 4 in a week and this for me is TOO MUCH!

????? The upload problems did only happens sometimes, so there is no problem.

Also, you are worried about minor outages. Everybody knows, that SETI has to less funding and the project peoples has a lot of work. They do there best. - I think the project peoples do a great job !
But, with a donation the you would help a lot. Look at the previous posts, a high available solution is about $13300.
Also I think it is wrong to wine at every outage. Peoples should understand, that things in the IT could brake, and that it could happen. If you want a HA solution you have to spend money.

---

[quote][quote]
Validate errors is becoming a rule, day by day.

But I can see 4 in a week and this for me is TOO MUCH!

Bye,
Francesco

Am I reading this correctly?
You have a RAC >9k, and you have had 4 validate errors/week?

[quote][quote]
Validate errors is becoming a rule, day by day.

But I can see 4 in a week and this for me is TOO MUCH!

Bye,
Francesco

Am I reading this correctly?
You have a RAC >9k, and you have had 4 validate errors/week?

No, probably my poor english ...
Don't look at my RAC.
I have seen 4 episodes of "validate error" in a week, as general problem, for all the users. I don't know how many results ar invalidated but looking at the loss of RAC in the stats of a lot of people, I think is a lot of wasting.

I can see that the problem creates a loss of job and RAC for thousand of crunchers. About 10% for every episode.
And that this happens too much frequently.
4 a week.

The problem is not only mine and I have opened this thread non to cry on my RAC but to put the general problem on the table.

Bye
Francesco

---