Around 10am this morning gowron's nfsds were all in disk wait. Not sure why, but that pretty much hosed the whole download part of our system. Jeff's been fighting with it all day. I've been at home, chiming in with my two cents every so often. Hopefully he'll get beyond it before too long.

- Matt

---

-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Hopefully he'll get beyond it before too long.

May have just done it.
Network traffic shows the flood gates have opened.

---

Grant
Darwin NT

Well everyone seems to be in a holding pattern, I'm not reporting and continuing to crunch away. Good luck guys ;-D

I think it is interesting and highly creditable that you have been able to identify the ghost units, and re-send them. My one and only ghost was re-issued to me today when I asked for work. Well Done!

What is the problem today? I looked at the server status and it looks like essentially everything is down. I hope it is scheduled downtime and not another catastrophe.

---

What is the problem today? I looked at the server status and it looks like essentially everything is down. I hope it is scheduled downtime and not another catastrophe.

I have no clue; however, Beta is affected by this "whatever-it-is" too... Also, accessing the Websites for SETI and Beta are very problematic for me. At first I thought I was having the usual Time Warner problems; however, I can get into Rosetta, Einstein, and NanoHive just fine - only SETI and Beta Websites and BOINC connections seem to be affected.

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

What is the problem today? I looked at the server status and it looks like essentially everything is down. I hope it is scheduled downtime and not another catastrophe.

I have no clue; however, Beta is affected by this "whatever-it-is" too... Also, accessing the Websites for SETI and Beta are very problematic for me. At first I thought I was having the usual Time Warner problems; however, I can get into Rosetta, Einstein, and NanoHive just fine - only SETI and Beta Websites and BOINC connections seem to be affected.

It's definately Seti server problems of some sort. The Cricket graph tells the tale. Right now the server status page shows everything up but the splitters, but something is definately borking comms in and out, even for the forums.
Uploads are working fine, but no reporting or downloading new work. May have to wait until tomorrow when eveyone gets into work to get it sorted again, unless they are still there working on it.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Looks like I got one of the last work units to get out before the big slowdown. 134190219

12 hours later copies of the work unit have not been sent to additional users.

I have a very slow machine (about 1 hour cpu time per unit of credit.) so I'm limit my que to reduce latency getting results back. I'm currently set to get the next work unit about 12 hours before completing the last. Looks like I should increase that some.

---

It's definately Seti server problems of some sort. The Cricket graph tells the tale. Right now the server status page shows everything up but the splitters, but something is definately borking comms in and out, even for the forums.
Uploads are working fine, but no reporting or downloading new work. May have to wait until tomorrow when eveyone gets into work to get it sorted again, unless they are still there working on it.

On the server status page, notice the large number of workunits (260318) waiting on assimilation. Also notice the 5 hour backlog on the transitioners. Add the symptoms of slow forum response, and trouble reporting completed work and requesting new work but no trouble uploading completed work.

My guess is that sidious (the BOINC database machine) is overloaded due to a large number of things it is being asked to do at the moment. Furthermore, I guess that the ghost workunit aborts over the last two days are slamming the system. Maybe part of the bottleneck is in the assimilators' access to the master science database as well, i dunno. But this theory seems to fit. Everything points to a bottleneck on sidious. Slow forums (but requests are eventually handled), trouble reporting (some are working, though... just had some of mine report a while ago), trouble getting new work downloaded... all depend on sidious. Uploading completed results doesn't really hit sidious that much, so they are somewhat unaffected. The staff likely has the splitters turned off to ease the load on sidious until the backlog clears.

Thats my guess.

Again i have issues with no work available on chicken app. I`m switching temporarily to E@H... in 50h, when Einstein unit`s done, all problems should be sorted out, i hope:)

---

Again i have issues with no work available on chicken app.

There is no work available for anyone.
The Server Status page shows things are up again, but there is still very little coming down the pipes (Cricket shows it's less than 1/4 of what it should be after such an outage).
I think it requires people's presence in the Lab, the remote work doesn't appear to have sorted it out.

---

Grant
Darwin NT

I'm getting the dreaded
"6/11/2007 7:49:33 AM|SETI@home|Scheduler request failed: HTTP service unavailable"
message in my messages tab when trying to report work: both when requesting new work; and when just trying to report.

---

.

Hello, from Albany, CA!...

well
i can just confirm it is broken (again) , hhpt service unavailable
the pain i see here is than afer many failures,app need to download the master file again , i wonder why
cheers
jeanpierrejpp

---

well
i can just confirm it is broken (again) , http service unavailable
the pain i see here is than afer many failures,app need to download the master file again , i wonder why
cheers
jeanpierrejpp

The master file gives the URL of the Scheduler. So loading the master file is just checking to make sure the failures aren't because the URL has changed.

                                                                 Joe

Again i have issues with no work available on chicken app. I`m switching temporarily to E@H... in 50h, when Einstein unit`s done, all problems should be sorted out, i hope:)

Why switch? Why not just assign resource shares and let BOINC do what it can get?

---

Why switch? Why not just assign resource shares and let BOINC do what it can get?

Not as many opportunities for button-ab^H^Hpushing that way. ;)

---