I am running Boinc 4.13/Seti@Home 4.08. I recently switched from Seti Classic. Boinc worked fine for a couple of weeks. However, I have not been able to connect to the server. I have a dozen completed work units sitting for over a week. I get the same messages.

SETI@home - 2005-01-11 20:03:19 - Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
SETI@home - 2005-01-11 20:03:25 - Scheduler RPC to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi failed
SETI@home - 2005-01-11 20:03:25 - No schedulers responded
SETI@home - 2005-01-11 20:03:25 - Deferring communication with project for 1 minutes and 0 seconds

I am runnning Windows 2000 SP4, 512 MB mem, AMD Athlon 1800+.

I've seen related posts, but no answers. Are any Boinc folks reading these messages? Anyone have any suggestions?

Jim H

Just uploaded some results & downloaded more Work Units with no problems.
On the Project tab, right click on Seti@home & then Update & see how things go.

---

Grant
Darwin NT

Tried the update feature (more than once), didn't work. Just tried the "reset project" feature. That did work. Now Boinc is downloading the Seti 4.08 client again. Lost the dozen or so work units I had completed.

Jim H

>
> Just uploaded some results & downloaded more Work Units with no problems.
> On the Project tab, right click on Seti@home & then Update & see how
> things go.
>

I have experienced the identical problem on a WinXP SP2 machine. Some additioal info:

The "no schedulers responded" error will continue indefinately until "reset project" is issued. Following "reset project" the client will download workunits, but only *once* - the next time it attempts to download the "no schedulers responded" error returns.

Uploading workunits completed in this roundabout manner of obtaining them is not a problem and the client indicates these are uploaded sucessfully, however, the units never appear as upload in the computers results listing and eventually expire.

I run boinc on five machines, three on WinXP SP2 machines, only one of which developed this problem. After several weeks of having to reset the project to obtain workunits and failing to receive credit for any that were uploaded sucessfully I gave up and removed boinc from the machine in question.

Please take a look here and the original site here There is an explanation how the two-step upload works. (First upload, Second Report!)

---

BOINC Doc | Team-Site | BOINC-Podcast

My problem was not a matter of uploading. It was lack of contact from the Seti/Boinc server. I could not get a connection for over a week. Completed work units were stuck on the "Work" tab with the status "ready to report". The "Transfers" tab was blank.

I checked the FAQ, but did not see anything that addresses my issue. Can you point to something more specific?

Jim H

> Please take a look here and
> the original site here There is an
> explanation how the two-step upload works. (First upload, Second Report!)

---

I received a second set of work units since I tried a "reset project" last night. I'm crunching the new batch now. I'll keep my fingers crossed and see if uploads continue to work.

Jim H

Jim,

> I checked the FAQ, but did not see anything that addresses my issue. Can you
> point to something more specific?

This is not in the FAQ. It is covered in the Glossary where we talk to "<a>". This probably should be talked about in the FAQ and also in the Owner's Manual.

Sigh ... I will get on it

---

I have the same problem and opened a thread for it a couple of days ago (http://setiweb.ssl.berkeley.edu/forum_thread.php?id=7885) -- however, no solutions up to now.

One additional curious aspect: Despite the "Scheduler RPC to ... Failed" message, BOINC continues to connect and upload WUs to the server. However, the server disregards them and they are lost forever (they don't appear on in the "work" page for the computer in question). It now happened several times ... really annoying.

The last time this happened, I killed BOINC, deleted all *.xml files in the boinc-directory and re-started the program. That way, BOINC looses all knowledge of the previous WUs (so they are still lost forever), but it recognizes that the software (SETI in my case) is already installed on the computer and download ist much faster. Automatically, a new ID is assigned to the computer (you can use "merge" on the webpage to join them again) and SETI now seems to run again (I went through two successfull UL/DL cycles till then). Still, I far from satisfactory "solution".

---

The same problem. No solution on the horizon.
It seems it is relatively frequent for couple of days on the Problems site.

---

I have problem "No schedulers responded" very often. It seems that this problem connected with slow (dial up) conncetion and large(>3) number of WU in state "Ready to report".
I find following solution:
1) stop BOINC program
2) make backup of boinc program directory with all subdirectories
3) accurately delete information connected with some number of wu from 4 files:
client_state.xml
client_state_prev.xml
sched_reply_setiathome.berkeley.edu.xml
sched_request_setiathome.berkeley.edu.xml
4) run BOINC program - check "Work" tab
5) on the tab Projects press Update
6) after "Scheduler RPC to ... succeeded" stop BOINC and very accurately copy information about another WU`s from backup *.xml and ALSO CORRESPONDING FILES from "projects/setiathome.berkeley.edu" subdirrectory

---

> I have problem "No schedulers responded" very often. It seems that this
> problem connected with slow (dial up) conncetion and large(>3) number of WU
> in state "Ready to report".

I doubt it. I have 4Mbit/s / 512 Kb/s line and exerience the same problem all the time.

---

> > I have problem "No schedulers responded" very often. It seems that this
> > problem connected with slow (dial up) conncetion and large(>3) number
> of WU
> > in state "Ready to report".
>
> I doubt it. I have 4Mbit/s / 512 Kb/s line and exerience the same problem all
> the time.
>


.... on the tab Projects press Update - solved the same problem with my PC

---

An update (but no solution) for this problem:
The check whether the "No schedulers responded" is due to my computer or something on the server end, I installed Predictor@home. In the meantime, one of my computers developed the "No schedulers responded" problem again ... but only with the SETI@home application. Predictor now completed several cycles (downloading new WUs, submitting results, credit gets granted). This suggests, that the problem is somewhere in Berkeley (maybe a "blackhole list"? for me, only computers in a local network with 192.168.x.x addresses are affected). Unfortunately, Predictor is a "bad" application for home computers and I'll probably discontinue working with it in a couple of days. The WUs are too large (~4MB each), the connection to their server is unreliable right now and it takes only a short time to process them (3h on my machine). With the slow ISDN connection I have at home that's not what I would like to have ...

---

I had this same problem when I upgraded to 4.14/15/16. My machines were able to upload wu, but not to report themo or download new units. At least from the client side, as if I checked in the host information web page, there were a lot of wu downloaded, but this wu could not be seen in the client.

As soon as I downgraded to 4.13 all worked OK.

---