Folding Forum

I assume that the powers that be are aware that a serious problem exists.

Most links off of the main Stanford server are down.

Several work servers were showing obscure values, at least they were when the server stats page was up.
It, along with the others is presently down.

Please wake up your "Techies" so that we do not have to put up with this for the entire weekend.

Virtually ALL of my Quads have been assigned WUs that utilize slightly over half of their power.
What a waste.

My Macs and Linux dualies have been issued WUs that have absolutely no chance to complete by their preferred deadlines.
(They will meet the final deadlines)

Please Help!

Thanks!

John

Looks as if the fix was underway.

Most pages became reachable immediately after I posted above.

I must have wiggled my nose in a magical way.

Thanks!

It still appears that 171.67.108.24 has a problem in that it had been and currently is showing a net load well beyond reasonable parameters.

This server, along with xx.56 are the PRIMARY servers for MANY Linux and OSX "volunteers".
xx.56 has been A.W.O.L. for most of the afternoon - i.e. too many connections.

John

314159 wrote:Looks as if the fix was underway.

Most pages became reachable immediately after I posted above.

I must have wiggled my nose in a magical way.

Thanks!

It still appears that 171.67.108.24 has a problem in that it had been and currently is showing a net load well beyond reasonable parameters.

This server, along with xx.56 are the PRIMARY servers for MANY Linux and OSX "volunteers".
xx.56 has been A.W.O.L. for most of the afternoon - i.e. too many connections.

John

All the pages are on and off accesible for me to, which rules out any dns errors on our side. It's obviously a Stanford issue, and I hope they get it fixed before the weekend

If not, this won't be a pleasant weekend here on the forums

On another note, I'm sure they are doing their best to keep the servers running, can't break iron with your hands though.

MtM wrote:All the pages are on and off accesible for me to, which rules out any dns errors on our side. It's obviously a Stanford issue, and I hope they get it fixed before the weekend

If not, this won't be a pleasant weekend here on the forums

On another note, I'm sure they are doing their best to keep the servers running, can't break iron with your hands though.

You're saying that you were unable to reach most/all pages subsequent to my post above?

I suspect it IS a major problem.

Back to some more nose wiggling, I guess.
Care to join me?

No need to suspect any more, it was confirmed much earlier. They are fully aware of the problems. Here is one example. http://foldingforum.org/viewtopic.php?f=24&t=8601

Thanks.

Just the serverstatus pages, psummary, and when trying to connect manually to some work servers they tended to either time out ( 404 ) or give me a page not found ( 400 , this for the serverstatus page not the work server ). Idk what happend yesterday, as I couldn't acces the forum as well. That's why I did suspect my dns server, as I had more us based sites I couldn't acces at the time. But through msn I asked UF if he could acces it and he also couldn't ( though he's also in the EU so it might as well have been a bottleneck somewhere between eu and vs idk ).

I am now using opendns, the forum is quicker then I'm used to but I still get the connectivity issues with the other pages I mentioned. Other us based sites and forums don't have issues though, so it's definetly not dns related.

My folding hw is all suspended btw, waiting for my electricty company switch to complete ( I'm going through some rough economics right now, need to save where ever I can even f@h ) so I don't notice the server issues unless I use a browser which I been doing a few times on servers reported in the forums as being troublesome.

Edit: jeez didn't even notice 7im's post above mine. Yeah I knew there were reported problems, I didn't go through all the threads though I guess I didn't expect psummary or serverstats to be affected.

Has it improved for you now? We don't see any problem here, eg serverstat looks fine (but we're inside the net). This may be a problem with our ISP.

Or, it may have been an issue which IT resolved themselves.

171.67.108.24 vsp21v is still showing a rediculous net load.

If you are not actively working on this machine, could someone please restart the binary.

Its being off-line is creating big problems for all Linux and OSX Quads.

Thank you!

John

I haven't seen anything weird or heard any reply back, so is it safe to say that this issue has been resolved on your end?

Sure--binary restarted. The CPU load was minimal, so the binary may have gotten stuck. (The server code sometimes builds up inactive connections for reasons that I don't completely understand; we have a new version of the code that's in testing and appears more robust, but the changeover is a little less than straightforward.)

Whew!

Dr. Pande: If you want to see something weird, I should send you a copy of the WUs that my (many) machines were assigned during this period.

Thank you both.

I hope that you can enjoy your weekends without further incident.

John

7im wrote:No need to suspect any more, it was confirmed much earlier. They are fully aware of the problems. Here is one example. http://foldingforum.org/viewtopic.php?f=24&t=8601

Thanks.

Hey 7im,

Your link is to a post from the 24th relating to two work servers: one for the PS3 platform, the other a "foldathome" server that is a mystery to me.

You really think that these were the cause of the problem? Nah......

Be good, keep up the good work, and fold on,

John

314159 wrote:a "foldathome" server that is a mystery to me.

This is mrshirts' server, which broadcast Double Gromacs core C WUs (Fahcore_7c). These are uniprocessor WUs in the 38xx range.

Aha!

You have "intimate" knowledge of that one as I recall.

I believe that this is the server that is located at the University of Virginia, just down the road from me.

I suspect that mrshirts will get those folks straightened out soon.

Anyway, I hope so!

Staying "off-topic" (sorry - this thread is now hopefully retired anyway).

I suspect that you will be reading this toTOW.

My 12 Quads are SUBSTANTIALLY underutilized at present, running the SMP client which the PG recommended for Quads.
I do not have a clue why this recommendation was made, and then they proceeded to issue a1 Core WUs to these powerful machines.
That stinks!!! and I have to admit to being more than slightly "pixxed".

I have been looking for the thread that I recall being in this Forum related to running two instances of the SMP client on native Linux Quad machines.
Would you be so kind as to point an old man in the right direction?

Thanks!

John