I've had a lot of downloads fail to pass the checksum since Nov 16. Here is a copy of the messages from the BOINC Manager:

11/17/2006 8:59:48 AM||Starting BOINC client version 5.4.11 for windows_intelx86
11/17/2006 8:59:48 AM||libcurl/7.15.3 OpenSSL/0.9.8a zlib/1.2.3
11/17/2006 8:59:48 AM||Data directory: C:Program FilesBOINC
11/17/2006 8:59:48 AM|rosetta@home|State file error: result (– not found
11/17/2006 8:59:48 AM|rosetta@home|State file error: result (– not found
11/17/2006 8:59:48 AM||Processor: 2 GenuineIntel Intel(R) Pentium(R) 4 CPU 3.20GHz
11/17/2006 8:59:48 AM||Memory: 1022.98 MB physical, 2.40 GB virtual
11/17/2006 8:59:48 AM||Disk: 74.52 GB total, 52.77 GB free
11/17/2006 8:59:48 AM|rosetta@home|URL: https://boinc.bakerlab.org/rosetta/; Computer ID: 323022; location: home; project prefs: default
11/17/2006 8:59:48 AM||General prefs: from rosetta@home (last modified 2006-11-12 09:24:27)
11/17/2006 8:59:48 AM||General prefs: no separate prefs for home; using your defaults
11/17/2006 8:59:48 AM||Local control only allowed
11/17/2006 8:59:48 AM||Listening on port 31416
11/17/2006 8:59:48 AM|rosetta@home|Started download of file BAR_R13_cc1opd_09_05.200_v1_3.gz
11/17/2006 8:59:48 AM|rosetta@home|File 1acf_.psipred_ss2.gz exists already, skipping download
11/17/2006 8:59:48 AM|rosetta@home|Started download of file 1acf.pdb.gz
11/17/2006 8:59:49 AM|rosetta@home|Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
11/17/2006 8:59:49 AM|rosetta@home|Reason: To fetch work
11/17/2006 8:59:49 AM|rosetta@home|Requesting 86400 seconds of new work
11/17/2006 8:59:49 AM||Project communication failed: attempting access to reference site
11/17/2006 8:59:51 AM||Access to reference site failed - check network connection or proxy configuration.
11/17/2006 8:59:51 AM|rosetta@home|Temporarily failed download of BAR_R13_cc1opd_09_05.200_v1_3.gz: http error
11/17/2006 8:59:51 AM|rosetta@home|Temporarily failed download of 1acf.pdb.gz: http error
11/17/2006 8:59:51 AM|rosetta@home|Started download of file BAR_R29_cc1acf_03_05.200_v1_3.gz
11/17/2006 8:59:51 AM|rosetta@home|Started download of file BAR_R29_cc1acf_09_05.200_v1_3.gz
11/17/2006 8:59:52 AM|rosetta@home|Temporarily failed download of BAR_R29_cc1acf_03_05.200_v1_3.gz: http error
11/17/2006 8:59:52 AM|rosetta@home|Temporarily failed download of BAR_R29_cc1acf_09_05.200_v1_3.gz: http error
11/17/2006 8:59:52 AM|rosetta@home|Started download of file 1acf__R29_cheat.bar
11/17/2006 8:59:52 AM|rosetta@home|Started download of file 1scjB.fasta
11/17/2006 9:03:09 AM|rosetta@home|Temporarily failed download of 1acf__R29_cheat.bar: http error
11/17/2006 9:03:09 AM|rosetta@home|Started download of file 1scjB.psipred_ss2.gz
11/17/2006 9:03:09 AM|rosetta@home|Scheduler request failed: couldn't resolve host name
11/17/2006 9:03:09 AM|rosetta@home|Deferring scheduler requests for 1 minutes and 0 seconds
11/17/2006 9:03:10 AM|rosetta@home|Temporarily failed download of 1scjB.fasta: http error
11/17/2006 9:03:10 AM|rosetta@home|Started download of file 1scj.pdb.gz
11/17/2006 9:03:31 AM|rosetta@home|Temporarily failed download of 1scjB.psipred_ss2.gz: http error
11/17/2006 9:03:31 AM|rosetta@home|Started download of file BAR_R14_cc1scjB03_05.200_v1_3.gz
11/17/2006 9:03:32 AM|rosetta@home|Temporarily failed download of 1scj.pdb.gz: http error
11/17/2006 9:03:32 AM|rosetta@home|Started download of file BAR_R14_cc1scjB09_05.200_v1_3.gz
11/17/2006 9:04:09 AM|rosetta@home|Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
11/17/2006 9:04:09 AM|rosetta@home|Reason: To fetch work
11/17/2006 9:04:09 AM|rosetta@home|Requesting 86400 seconds of new work
11/17/2006 9:04:45 AM|rosetta@home|Scheduler request failed: failure when receiving data from the peer
11/17/2006 9:04:45 AM|rosetta@home|Deferring scheduler requests for 1 minutes and 0 seconds
11/17/2006 9:05:45 AM|rosetta@home|Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
11/17/2006 9:05:45 AM|rosetta@home|Reason: To fetch work
11/17/2006 9:05:45 AM|rosetta@home|Requesting 86400 seconds of new work
11/17/2006 9:05:50 AM|rosetta@home|Scheduler request succeeded
11/17/2006 9:05:50 AM|rosetta@home|Message from server: Not sending work - last request too recent: 95 sec
11/17/2006 9:05:50 AM|rosetta@home|No work from project
11/17/2006 9:08:57 AM|rosetta@home|Finished download of file BAR_R14_cc1scjB09_05.200_v1_3.gz
11/17/2006 9:08:57 AM|rosetta@home|Throughput 5819 bytes/sec
11/17/2006 9:08:57 AM|rosetta@home|Started download of file 1scjB_R14_cheat.bar
11/17/2006 9:08:57 AM|rosetta@home|MD5 check failed for BAR_R14_cc1scjB09_05.200_v1_3.gz
11/17/2006 9:08:57 AM|rosetta@home|expected 469cb1977024c8bacb33d5aded75a7e1, got b9033e9a5129a9ad6a7554981de0156d
11/17/2006 9:08:57 AM|rosetta@home|Checksum or signature error for BAR_R14_cc1scjB09_05.200_v1_3.gz
11/17/2006 9:08:59 AM|rosetta@home|Finished download of file 1scjB_R14_cheat.bar
11/17/2006 9:08:59 AM|rosetta@home|Throughput 27 bytes/sec
11/17/2006 9:08:59 AM|rosetta@home|Started download of file cc1acf_03_05.200_v1_3.gz
11/17/2006 9:09:24 AM|rosetta@home|Finished download of file BAR_R14_cc1scjB03_05.200_v1_3.gz
11/17/2006 9:09:24 AM|rosetta@home|Throughput 2407 bytes/sec
11/17/2006 9:09:24 AM|rosetta@home|Started download of file cc1acf_09_05.200_v1_3.gz
11/17/2006 9:09:24 AM|rosetta@home|MD5 check failed for BAR_R14_cc1scjB03_05.200_v1_3.gz
11/17/2006 9:09:24 AM|rosetta@home|expected c404453321d4f5f305dada1b15d386e1, got 1816715f728146dd2056cf0689d17463
11/17/2006 9:09:24 AM|rosetta@home|Checksum or signature error for BAR_R14_cc1scjB03_05.200_v1_3.gz
11/17/2006 9:09:56 AM|rosetta@home|Sending scheduler request to https://boinc.bakerlab.org/rosetta_cgi/cgi
11/17/2006 9:09:56 AM|rosetta@home|Reason: To fetch work
11/17/2006 9:09:56 AM|rosetta@home|Requesting 86400 seconds of new work

---

You had a lot of network errors besides the checksum there.

I agree that these should not affect the checksumming process, but it is possible that the download mechanism got confused by the stops and restarts in the download process.

(As you may know, the download mechanism attempts to restart a download part way through if it is interrupted - it does not go back to the beginning. This restart is usually reliable but not quite so reliable as a download that completes in one connection.

Do you see the checksum error when there are no other network errors? (ie no faile ddownload messages for other reasons, http errors, etc, in the few minutes before)

River~~

---

I do see the checksum error when there are no other network errors (http, etc). The last time I tried was this morning.

Is it possible that installing Internet Explorer 7 may have interfered with something? This is on a Windows XP Media Center PC with SP2 and all current patches installed.

Thanks,
Karl

---

I do see the checksum error when there are no other network errors (http, etc). The last time I tried was this morning.

Is it possible that installing Internet Explorer 7 may have interfered with something? This is on a Windows XP Media Center PC with SP2 and all current patches installed.

Thanks,
Karl

I imagine that if IE 7 is interfering with file downloads in general, that this will be noticed very quickly. I'd look again today to see if new patches (for IE 7 or XP MC) have been posted.

Did the proxy settings change when you installed IE 7? If they did, should they have done? (These are more likely to cause a complete failure to download than a checksum).

Next, I'd try re-installing BOINC. It is possible (but unlikely) that the way IE takes over as your default browser has somehow interfered with BOINC's own settings for downloading. The reason I say it is unlikely is that BOINC uses its own code for this, it does not go through IE.

Stop BOINC and re-run the install file if you still have it, or download it first if you deleted it.

When you stop BOINC will lose any work that has not been checkpointed, but the re-install should not make you lose any more than that. When you restart the client should pick up from where the previous installation had got to.

These suggestions are grasping at straws - I don't really see why installing IE 7 should have broken anything at all, and others may have better suggestions to make than mine.

R~~