Holy Crap!! SciLINC

[Bender10]

I just had SciLINC try to ASSIMILATE my computer!!

I was just poking around, looking at stats and smurfing the interweb...when I noticed that I had not checked for SciLINC wu's in a few dayz. So I did. Holy Crap!! It started downloading ok, but then I noticed it was trying to download about a gazillion 3k files!!

My BOINC manager went unresponsive (5.8.16) and jerky. I shut it down a few times, and un-installed it.

(My next thought was to find the chainsaw, and watch out for little green critters.....)

I finally got it to respond to the 'Detach' button. After that everything was ok.

Nice Project... :shocked:...(jerks)

[PoorBoy]

:icon_lol: ... Their Bitching about the Project over in the Senergy Forum too ... I don't know what this guy was thinking when he set the Wu's up but he's going to have a hard time getting people to run the Project the way it is set up now ...

[Bender10]

Wonder what kind of testing he did not do...?

[Bubben]

Just read this at their page:

June 15, 2007 Server Processes Running Again
The process that gathers new work for SciLINC got a little carried away and filled the drive with everything it could find in the vaults. Images, PDFs, everything. This has been corrected and the server should now be running correctly again.

:icon_cool:

[AMDave]

That was definitely an alpha-woopsie.
I could not get the client to respond in any way even after reboots so I had to uninstall and reinstall.

I saw the complaints too.
That's dumb.
No one has any right to complain really.
This is what alpha projects are all about - knocking out the bugs, both in the code and in the project processes.

When you connect to an alpha project you acknowledge the risks.
If they don't like the prospect of spending hours diagnosing, testing, reinstalling or even rebuilding their pharm then don't connect to alpha projects.

The experience was worth it.
SciLinc figured out one of their settings is really critical.
The rest of us just figure out that BOINC still get's itself in horrible knots when dealing with high frequency network transfers and log entries.
Honestly I thought that had been corrected in previous versions.
BOINC 5.9.4 (Windows) however got slammed as did a couple of other versions (eg 5.8.x)
Interestingly BOINC 5.10.x (Linux-experimental) did not get into a locked state.

So for me, I observed some points of interest out of it.

I did stuff up a days worth of Tanpaku though (haha).
But like I said above that was my responsibility.

[Lagu]

I got a flood of work and at the same time Tanpaku ran. Boinc stalled and ended answer. I tried pause Tanpaku and the SciLINC WU´s was executed and as far it was done 2 new Tasks downloaded. I never stopped and Boinc was unstable and didn’t run. At least I was forced quiet that project. First when it was done Boinc began work again. I couldn’t se how many but I guess I got 500 WU´s. Shit!

And there is 5 as will verify the same WU.:icon_twisted:

[Nflight]

I Have since stopped and restarted the project, this seems to have helped somewhat. I am still receiving those pesky 1 second work units, but my system is not dead it is working trough each one of them. Last night it seemed like the world was going to come to an end with 3 gazillion WU's. Now we are at less then a gazillion but still and awful lot to process!

[Ototero]

I'm still glad I have my non-Boink stance. I don't seem to have any of the issues you guys have.

[PoorBoy]

I'm trying to get the Project to run with some sort of reasonable confusion ... :icon_lol:

I don't think I can run it on my Quads as the Uploads just keep piling up over time & would eventually just choke the System again as other Projects tried to Upload their Wu's too.

So I'm trying it on my Duo Core X6800 & will see what happens, it looks like pretty much the same thing. Another problem is I'm not even getting any Credit, what little there is to get for over half the Wu's, they run so fast that they don't ask for any Credit. I ran 10 of them on the X6800 & 8 of them never asked for any Credit and didn't receive any Credit, so I'm running them for free I guess & taking away from the other Projects.

I think a slower CPU is in order for this Project so as the Wu will run at least 1 sec or more and ask for Credit. I'll try it on my P4 3.4 HT and see what happens. Well at least the P4 3.4 HT asks for Credit, a whole whooping 0.01 and gets it anyway. But it still eventually managed to choke the Uploads to a crawl, have 5 Wu's trying to Upload now so I'm going to have to just shut this Project down I guess.

Another problem with the Project is there's no Message or Question Section, actually there is one but you can't post anything there to let the Guy know whats going on with his Wu's ... :icon_rolleyes:

[PoorBoy]

I'm still glad I have my non-Boink stance. I don't seem to have any of the issues you guys have.

This Project is an extreme case I think, I haven't really had this kind of problem with a Project except for the RCN Project. My systems would get choked down with that Project too trying to Upload all the Files you were required to.

I stopped running that Project for some time and went back recently and things seemed to be a lot better, at least I could run the Project without getting bogged down with Uploads.

A lot has to do with the Projects servers, most of these Projects start out on a wing and a prayer & have to be coaxed into getting better Servers if they expect any sort of participation ...

[mbgrparker]

Hello all,

I am the <insert epithet> that setup the SciLINC project and its workunits.

We did perform internal testing on the project and did not see these issues.

Of course the latency of the Internet is much higher than our local network. Unfortunately, I did not foresee that causing these problems.

Updates will be posted to the main SciLINC page (http://www.scilinc.org) as we get our database back on-line and are able to address the work unit issue.

Please, see a discussion on BOINCstats for hints on improving SciLINC performance (http://boincstats.com/forum/forum_thread.php?id=2164&nowrap=true#21154) so that your machine may recover better once the database is back up.

Ron Parker

[Bender10]

Hi Ron,

Thanks for taking the time to visit and explain. Your project sure caused a few moments of 'mild stress' around here. But everyone was sure it would be straightened out.

Anyway, thanks for the update, and I hope things go smoother for you in the future!

[Frederic Brillouet]

I laughed at the <insert epithet> part, though I don't have a clue on what it means :icon_lol:

[Bender10]

<insert epithet> means to use your favorite expression of dislike...

Jerk
Dumba$$
Weasle

You get my meaning ??

[Frederic Brillouet]

<insert epithet> means to use your favorite expression of dislike...

Jerk
Dumba$$
Weasle

You get my meaning ??
I guessed it would mean something like that :icon_mrgreen: (btw I didnt know ass was censored here :icon_neutral: )

EDIT: It isn't haha

[PcManiac]

I guessed it would mean something like that :icon_mrgreen: (btw I didnt know ass was censored here :icon_neutral: )

EDIT: It isn't haha

Hey! for once I am not the only one who doesn't get it :icon_lol:
I mean, I got this one, but you know what I mean lol.

[AMDave]

Welcome to the AMD Users forum, Ron.

Thanks for popping in and letting us know you are on top of it.
That BOINCview setting is handy to know, too.

Also glad to see your sense of humour is intact.
("/me dawns his flame proof fire suit." - I have got to get me one of those for work LOL)

Stick with it, Ron.
We are with you all the way

( PS - Note to Fred. - I'm watching your a$$ :icon_twisted: heh heh )

[Frederic Brillouet]

Welcome to the AMD Users forum, Ron.
...
( PS - Note to Fred. - I'm watching your a$$ :icon_twisted: heh heh )
Yikes :5eek:

I'd think you'd rather be watching Lisa and Jess' ass :icon_twisted: heh heh