Hello it is me again,

I have a new issue with another pc that won't upload Rosseta completed WU's. I have asked there if there are issues and there are none. From MilkyWay I get no new tasks. I have to error messages:
1/29/2012 2:47:28 PM rosetta@home Temporarily failed upload of T0568_boinc_cm_test_abinitio_cmiles_IGNORE_THE_REST_36289_16141_0_0: HTTP error
and
1/29/2012 2:45:45 PM Milkyway@home Scheduler request failed: HTTP file not found

What must I do to get things working again?
Thanks for the help in anticipation.

---

Greetings from,
TJ

Set up a core client configuration file (cc_config.xml) with the following lines in it:

<cc_config>
<log_flags>
<http_debug>1</http_debug>
<http_xfer_debug>1</http_xfer_debug>
</log_flags>
</cc_config>

Make sure BOINC reads this config file (Advanced view->Advanced->Read config file). Then let it contact Rosetta once and let us see the output of that, completely.

After this disable the flags for now.

<cc_config>
<log_flags>
<http_debug>0</http_debug>
<http_xfer_debug>0</http_xfer_debug>
</log_flags>
</cc_config>

---

Hi again Jord,
This is the output.

1/29/2012 4:27:19 PM rosetta@home Temporarily failed upload of T0609_boinc_cm_test_abinitio_cmiles_IGNORE_THE_REST_36313_16098_0_0: HTTP error
1/29/2012 4:27:19 PM rosetta@home Backing off 3 hr 8 min 10 sec on upload of T0609_boinc_cm_test_abinitio_cmiles_IGNORE_THE_REST_36313_16098_0_0
1/29/2012 4:27:20 PM Internet access OK - project servers may be temporarily down.
1/29/2012 4:30:23 PM Re-reading cc_config.xml
1/29/2012 4:30:23 PM Re-read config file
1/29/2012 4:30:23 PM log flags: file_xfer, sched_ops, task, http_debug, http_xfer_debug
1/29/2012 4:30:33 PM rosetta@home update requested by user
1/29/2012 4:30:35 PM rosetta@home Sending scheduler request: Requested by user.
1/29/2012 4:30:35 PM rosetta@home Not reporting or requesting tasks
1/29/2012 4:30:35 PM [http_debug] HTTP_OP::init_post(): http://srv4.bakerlab.org/rosetta_cgi/cgi
1/29/2012 4:30:35 PM [http_debug] HTTP_OP::libcurl_exec(): ca-bundle set
1/29/2012 4:30:36 PM [http_debug] [ID#0] info: timeout on name lookup is not supported
1/29/2012 4:30:36 PM [http_debug] [ID#0] info: About to connect() to srv4.bakerlab.org port 80 (#5)
1/29/2012 4:30:36 PM [http_debug] [ID#0] info: Trying 128.95.160.145...
1/29/2012 4:30:36 PM [http_debug] [ID#0] info: Connected to srv4.bakerlab.org (128.95.160.145) port 80 (#5)
1/29/2012 4:30:36 PM [http_debug] [ID#0] Sent header to server: POST /rosetta_cgi/cgi HTTP/1.1
User-Agent: BOINC client (windows_x86_64 6.10.17)
Host: srv4.bakerlab.org
Accept: */*
Accept-Encoding: deflate, gzip
Content-Type: application/x-www-form-urlencoded
Content-Length: 7979
Expect: 100-continue


1/29/2012 4:30:37 PM [http_debug] [ID#0] info: Done waiting for 100-continue
1/29/2012 4:30:37 PM [http_debug] [ID#0] Received header from server: HTTP/1.1 100 Continue

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: HTTP/1.1 200 OK

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: Date: Sun, 29 Jan 2012 15:29:11 GMT

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: Server: Apache/2.2.3 (Red Hat)

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: Connection: close

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: Transfer-Encoding: chunked

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server: Content-Type: text/xml

1/29/2012 4:30:38 PM [http_debug] [ID#0] Received header from server:

1/29/2012 4:30:38 PM [http_xfer_debug] HTTP: wrote 1288 bytes
1/29/2012 4:30:38 PM [http_xfer_debug] HTTP: wrote 1523 bytes
1/29/2012 4:30:38 PM [http_debug] [ID#0] info: Expire cleared
1/29/2012 4:30:38 PM [http_debug] [ID#0] info: Closing connection #5
1/29/2012 4:30:40 PM rosetta@home Scheduler request completed

---

Greetings from,
TJ

The issue with MilkyWay I resolved by Detach and then Attach again, as I was not having any WU's of them. That worked as I get WU's now again.
I won't like to that with Rosetta as I loose the 3 WU's then.

---

Greetings from,
TJ

Please try with this cc_config.xml and post its output.

<cc_config>
<log_flags>
<file_xfer_debug>1</file_xfer_debug>
</log_flags>
</cc_config>

Although we can try one quick solution as well.
Use the following cc_config.xml to change the client's use of HTTP from 1.1 to 1.0; some ISPs fare better with HTTP 1.0 connections.

<cc_config>
<options>
<http_1_0>1</http_1_0>
</options>
</cc_config>

This is with 6.10.60, right?

---

The output is very short with that file:
1/29/2012 6:06:47 PM Re-reading cc_config.xml
1/29/2012 6:06:47 PM Re-read config file
1/29/2012 6:06:47 PM log flags: file_xfer, sched_ops, task, file_xfer_debug
1/29/2012 6:06:54 PM rosetta@home update requested by user
1/29/2012 6:06:58 PM rosetta@home Sending scheduler request: Requested by user.
1/29/2012 6:06:58 PM rosetta@home Not reporting or requesting tasks

The version on this AMD pc is 6.10.17 works normally flawlessly.
I just remember that Rosetta have moved databases a few weeks ago and indeed this pc was of since then. Could that be it?

---

Greetings from,
TJ

Although we can try one quick solution as well.
Use the following cc_config.xml to change the client's use of HTTP from 1.1 to 1.0; some ISPs fare better with HTTP 1.0 connections.

<cc_config>
<options>
<http_1_0>1</http_1_0>
</options>
</cc_config>

I did this too but seems not to work either.

---

Greetings from,
TJ

The version on this AMD pc is 6.10.17 works normally flawlessly.
I just remember that Rosetta have moved databases a few weeks ago and indeed this pc was of since then. Could that be it?

Didn't 6.10.17 have the DNS caching Bug?, you could try upgrading to the latest Boinc 6.10.x version (6.10.60) and see if that is any better.

Claggy

The version on this AMD pc is 6.10.17 works normally flawlessly.
I just remember that Rosetta have moved databases a few weeks ago and indeed this pc was of since then. Could that be it?

Didn't 6.10.17 have DNS caching Bug?, you could try upgrading to the latest Boinc 6.10.x version (6.10.60) and see if that is any better.

Claggy

I didn't know of that bug, but a few weeks back this version was working for me.
I will update later but when I do that now I loss these 3 WU's and the fourth running.

---

Greetings from,
TJ

I let the debug from Jord on and see what happend with the last runed WU.
It uploaded well so BOINC version 6.10.17 is working fine.

1/29/2012 7:00:46 PM rosetta@home Computation for task prednorlx_ECH13_hisremstart_lrN_572_0001_37278_116_1 finished
1/29/2012 7:00:47 PM rosetta@home Started upload of prednorlx_ECH13_hisremstart_lrN_572_0001_37278_116_1_0
1/29/2012 7:00:47 PM rosetta@home [file_xfer_debug] URL: http://srv5.bakerlab.org/rosetta_cgi/file_upload_handler
1/29/2012 7:00:49 PM rosetta@home [file_xfer_debug] FILE_XFER_SET::poll(): http op done; retval 0
1/29/2012 7:00:49 PM rosetta@home [file_xfer_debug] parsing upload response: <data_server_reply> <status>0</status> <file_size>0</file_size></data_server_reply>
1/29/2012 7:00:49 PM rosetta@home [file_xfer_debug] parsing status: 0
1/29/2012 7:00:53 PM rosetta@home [file_xfer_debug] FILE_XFER_SET::poll(): http op done; retval 0
1/29/2012 7:00:53 PM rosetta@home [file_xfer_debug] parsing upload response: <data_server_reply> <status>0</status></data_server_reply>
1/29/2012 7:00:53 PM rosetta@home [file_xfer_debug] parsing status: 0
1/29/2012 7:00:53 PM rosetta@home [file_xfer_debug] file transfer status 0
1/29/2012 7:00:53 PM rosetta@home Finished upload of prednorlx_ECH13_hisremstart_lrN_572_0001_37278_116_1_0
1/29/2012 7:00:53 PM rosetta@home [file_xfer_debug] Throughput 57742 bytes/sec

That the other 3 files won't upload is thus not directly a problem of mine?

---

Greetings from,
TJ

Eventually I did a detach and reatach to Rossetta and that seems to work.
However at this moment there a 2 new WU's that won't upload so it must be a problem at Rosetta, but all silent there as usual.

---

Greetings from,
TJ

If you like folding, try Human Proteome Folding, Phase 2, which is a stable Rosetta version. Run at WCG by a former associate of Dr.B, the project has accumulated over 100,000 runtime years of contributions. Rarely a false note.

We've had our share of http issues, one of them was that 6.12.34 and up had upload problems that do not show with 6.12.33. For/by WCG, 6.10.58 is recommended, as installed with their many college/high school and corporate contributors to the grid.

--//--

If you like folding, try Human Proteome Folding, Phase 2, which is a stable Rosetta version. Run at WCG by a former associate of Dr.B, the project has accumulated over 100,000 runtime years of contributions. Rarely a false note.

We've had our share of http issues, one of them was that 6.12.34 and up had upload problems that do not show with 6.12.33. For/by WCG, 6.10.58 is recommended, as installed with their many college/high school and corporate contributors to the grid.

--//--

I will try it later indeed. I like project for medicine that is way I am with Rosetta, but the support from them is awfull if one has issues.
Earlier problem has to do with the move of the servers and someone from the crunching community gave a workaround instead of from th project.
I run 6.10.17, 6.10.58 and 6.10.60 with no problems for other projects and normally no problems for Rosetta as well. It started last Friday and overnight on another pc again.

---

Greetings from,
TJ

The problem is still not over. I have reset rosetta once more. I am now getting new WU's and need to wait if they upload.
However after a long wait LHC has work and on one PC they run fine and upload.
You guess already. On this particular system they won't upload. Again the HTTP error.
What can I do?
Einstein and MilkyWay run on this machine too and they upload without any problem.
AMD Phenom II X4 with win vista home premium 64 bit.

---

Greetings from,
TJ

When other projects work without problems on that machine, it isn't your BOINC version that causes this but the project --it may just be dropping the connection when your system tries to do uploads. Many server connection problems are project problems, so you'll have to continue to bombard them with your problem.

---

It could also be a routing problem with your ISP, try a different ISP, be it via WiFi Hotspot, 3G Dongle etc,
or try a different DNS server, or Proxy Server,

Claggy

I tried a few things and posted on the site of Rosetta, but it seems I am the only one. Or there are not enough questions for the admins to react.
Strange thing is that Rosetta is giving me new work and when ready it won't upload. Same with LHC (who has only very limited work). Einstein and MilkyWay running fine, so it couldn't be my system was my idea and Ageless confirms.
I will see it a few more days and then no more Rosie and LHC on this machine.

---

Greetings from,
TJ