I'm running into problems uploading results right now -- not on all of my systems, but on most of them (and only uploads to Seti).

Prject communication failed, Access to reference site succeeded, project servers may be down.

And then on another workstation on the same network, I can get thru.

Anyone else encountering this?

---

I'm running into problems uploading results right now -- not on all of my systems, but on most of them (and only uploads to Seti).

Prject communication failed, Access to reference site succeeded, project servers may be down.

And then on another workstation on the same network, I can get thru.

Anyone else encountering this?

Matt mentioned 'upload server choking' in today's tech news - I think they're on to it. But a FlushDNS never did any harm.

Same problem here. I uploaded one computer fine, then when I went to do the next, well you know the story. Should be up and running soon.

---

OK -- thought that might be the case, these workstations are behind a proxy server -- first thing I did was flush the DNS cache.

Matt mentioned 'upload server choking' in today's tech news - I think they're on to it. But a FlushDNS never did any harm.

---

At least I know I am not getting the problem alone. But, even with a FlushDNS a WU seems to get stuck

---

It's good to be back amongst friends and colleagues

Even using flushDNS, and on 2 separate rigs, there is both a problem uploading results and then getting an acknowledgement when the WU is "ready to report". But, the main problem seems to be with the results upload.

---

It's good to be back amongst friends and colleagues

Even using flushDNS, and on 2 separate rigs, there is both a problem uploading results and then getting an acknowledgement when the WU is "ready to report". But, the main problem seems to be with the results upload.

I'm not seeing the upload problem per se, but I have noticed a significant delay while the upload is sitting at 100% waiting for the acknowledgement over the last few days. But they all seem to get there in the end so I'm not fretting about it. Perhaps related to the issues Matt has been raising in Technical?

F.

---

see Matt's 'Slow Going - Feb 07 2008'

---

BOINC Wiki . . .

Science Status Page . . .

see Matt's 'Slow Going - Feb 07 2008'

Hi everyone, NO problems, crossing the Atlantic ;)
Only some minot 'glitches'.



6-2-2008 9:17:00||Starting BOINC client version 5.10.30 for windows_intelx86
6-2-2008 9:17:00||log flags: task, file_xfer, sched_ops
6-2-2008 9:17:00||Libraries: libcurl/7.17.1 OpenSSL/0.9.8e zlib/1.2.3
6-2-2008 9:17:00||Data directory: D:\\Program Files\\BOINC
6-2-2008 9:17:00||Processor: 2 GenuineIntel Intel(R) Pentium(R) D CPU 2.80GHz [x86 Family 15 Model 4 Stepping 4]
6-2-2008 9:17:00||Processor features: fpu tsc pae nx sse sse2 mmx
6-2-2008 9:17:00||OS: Microsoft Windows XP: Home Edition, Service Pack 2, (05.01.2600.00)
6-2-2008 9:17:00||Memory: 3.25 GB physical, 5.09 GB virtual
6-2-2008 9:17:00||Disk: 232.88 GB total, 193.52 GB free
6-2-2008 9:17:00||Local time is UTC +1 hours
6-2-2008 9:17:00|SETI@home|URL: http://setiathome.berkeley.edu/; Computer ID: 4181733; location: home; project prefs: default
6-2-2008 9:17:00||General prefs: from SETI@home (last modified 05-Feb-2008 23:59:19)
6-2-2008 9:17:00||Host location: home
6-2-2008 9:17:00||General prefs: no separate prefs for home; using your defaults
6-2-2008 9:17:00||Reading preferences override file
6-2-2008 9:17:00||Preferences limit memory usage when active to 1996.49MB
6-2-2008 9:17:00||Preferences limit memory usage when idle to 3161.11MB
6-2-2008 9:17:00||Preferences limit disk usage to 46.57GB
6-2-2008 9:17:00|SETI@home|Restarting task 22no06ae.27122.2935.12.7.227_1 using setiathome_enhanced version 527
6-2-2008 10:07:17|SETI@home|Restarting task 22no06ae.27122.2935.12.7.227_1 using setiathome_enhanced version 527
6-2-2008 10:26:29|SETI@home|Restarting task 22no06ae.27122.2935.12.7.227_1 using setiathome_enhanced version 527
6-2-2008 10:36:36|SETI@home|Restarting task 22no06ae.27122.2935.12.7.227_1 using setiathome_enhanced version 527
6-2-2008 14:59:15|SETI@home|Computation for task 22no06ae.27122.2935.12.7.227_1 finished
6-2-2008 14:59:17|SETI@home|Started upload of 22no06ae.27122.2935.12.7.227_1_0
6-2-2008 14:59:44|SETI@home|Finished upload of 22no06ae.27122.2935.12.7.227_1_0

***** See ya all later *****

---

OK -- the problem appears to have become a rather persistant one. With some effort I can often enough 'squeeze' results in -- but it is a manual and time wasting effort. For now, I coded SET for no new work so I can clear off the backlog over time and keep other more smoothly running projects well fed and active. I figure that when this problem is resolved, there will be some form of news flash so I will know to bring my SETI work back into the land of the living.

---

An update -- it seems that SETI can be quite intrusive in its desire to upload work units -- as a result, in order to let other projects have a piece of the upload pie, the only thing I could do was detach SETI from a few workstations. SETI was not only being a royal pain about requiring special efforts to upload results, it was not 'letting go' for uploads so that Rosetta, World, Spinhenge, etc work units couldn't get cycle time for their uploads. Detaching SETI resolved that problem -- not the solution I wanted though.

Hopefully their will be not only a resolution to this bug, but also some newsflash regarding it. As it is, my SETI activities are likely to dwindle as I clear out existing workunits.

OK -- the problem appears to have become a rather persistant one. With some effort I can often enough 'squeeze' results in -- but it is a manual and time wasting effort. For now, I coded SET for no new work so I can clear off the backlog over time and keep other more smoothly running projects well fed and active. I figure that when this problem is resolved, there will be some form of news flash so I will know to bring my SETI work back into the land of the living.

---

one member of our Team, the Green Planet became this

08.02.2008 21:43:15|SETI@home|Sending scheduler request: Requested by user. Requesting 6912000 seconds of work, reporting 0 completed tasks
08.02.2008 21:43:20|SETI@home|Scheduler request succeeded: got 0 new tasks
08.02.2008 21:43:20|SETI@home|Message from server: No work sent
08.02.2008 21:43:20|SETI@home|Message from server: (reached daily quota of 4 results)

daily quota of 4....., i think there is something going wrong...

Greetings from Germany NRW
Ulli

one member of our Team, the Green Planet became this

08.02.2008 21:43:15|SETI@home|Sending scheduler request: Requested by user. Requesting 6912000 seconds of work, reporting 0 completed tasks
08.02.2008 21:43:20|SETI@home|Scheduler request succeeded: got 0 new tasks
08.02.2008 21:43:20|SETI@home|Message from server: No work sent
08.02.2008 21:43:20|SETI@home|Message from server: (reached daily quota of 4 results)

daily quota of 4....., i think there is something going wrong...

Greetings from Germany NRW
Ulli

Is he turning in a lot of computation errors or failed WUs?

---

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

Hopefully their will be not only a resolution to this bug, but also some newsflash regarding it. As it is, my SETI activities are likely to dwindle as I clear out existing workunits.

And earlier:

OK -- thought that might be the case, these workstations are behind a proxy server -- first thing I did was flush the DNS cache.

Barry,

You might want to keep an eye on Internet access thru proxy in the BOINC dev boards - it seems as if the long-standing (since v4.45!) bug causing problems with proxy servers is finally getting close to being solved. It doesn't sound as if the latest v5.10.41 is quite ready for showtime yet (unless you want to be a guinea-pig), but the progress is really, really promising.

Thanks for that -- I'd note that the only time I've encountered what seem to be proxy server related communications issues with BOINC is with SETI (I recall this once in the past at least). The work around that sometimes is working now is to direct to my proxy server gateway IP for results uploads (those that get snagged and listed in transfers), and then change back to direct (which my proxy server is configured to accept), exit BOINC, then restart BOINC and select update to get the results posted. Sort of obscure -- works about 50% of the time.

The thing is, until this problem is put back to bed on the SETI side, I'm going to have decreasing activity for SETI -- with an increase in my other projects.

You might want to keep an eye on Internet access thru proxy in the BOINC dev boards - it seems as if the long-standing (since v4.45!) bug causing problems with proxy servers is finally getting close to being solved. It doesn't sound as if the latest v5.10.41 is quite ready for showtime yet (unless you want to be a guinea-pig), but the progress is really, really promising.

---

I have been having the same upload problems since Thursday with all 10 of my machines. The best I can narrow it down it started around 1pm Thursday (eastern). Sometimes I can get a single machine to upload but not all of them.

Are there any resolutions or suggestions yet?


2/8/2008 5:19:18 PM|SETI@home|Sending scheduler request: Requested by user
2/8/2008 5:19:18 PM|SETI@home|Reporting 13 tasks
2/8/2008 5:19:20 PM||Project communication failed: attempting access to reference site
2/8/2008 5:19:21 PM||Access to reference site succeeded - project servers may be temporarily down.
2/8/2008 5:19:23 PM|SETI@home|Scheduler request failed: couldn't connect to server
2/8/2008 5:19:23 PM|SETI@home|Sending scheduler request: Requested by user
2/8/2008 5:19:23 PM|SETI@home|Reporting 13 tasks
2/8/2008 5:19:25 PM||Project communication failed: attempting access to reference site
2/8/2008 5:19:26 PM||Access to reference site succeeded - project servers may be temporarily down.
2/8/2008 5:19:28 PM|SETI@home|Scheduler request failed: couldn't connect to server
2/8/2008 5:19:28 PM|SETI@home|Deferring communication for 14 min 47 sec
2/8/2008 5:19:28 PM|SETI@home|Reason: scheduler request failed

---

This errors i'm getting also: No connection to the Seti server for the reporting of results. Uploads went - well not well - but they work somehow, but the reporting is impossible. I use a local proxy server, which i think, is the problem. If i bypass the proxy and tell Boinc to connect directly, the units were reportet immediately. Since i changed nothing on my end for months, this is definitely a bug on the servers end. I hope this PITA will be corrected ASAP, cause i normally only allow access to the internet through my local proxy. This workaround - which, by the way, is only needed for Seti, all other projects work very well - is not acceptable for long time...

---

Aloha, Uli

I don't believe the problem is proxy related. My machines are on a router behind a firewall with no proxy server involved. And, there has been no changes to the computers on the network or the network itself.

---

I don't believe the problem is proxy related. My machines are on a router behind a firewall with no proxy server involved. And, there has been no changes to the computers on the network or the network itself.

Maybe there is a transparent forced proxy on your providers side involved, that you can't see from your network?
At work i have also a proxy, which i cannot circumvent. The only project having problems is Seti also. That's why i think the problem is proxy related.

---

Aloha, Uli

For a day or 2 i was getting some uploads not uploading on the 1st attempt
8/02/2008 6:22:25|SETI@home|[file_xfer] Temporarily failed upload of 29dc06ai.10424.5798.4.7.106_1_0: system connect
and when they finally did so it was at around 100-400 B/s. Over the last 12 hours or so pretty much all uploads have gone through on the first attempt amd at 10kB/s or better (which is about par for the course).

---

Grant
Darwin NT