Has anyone else seen this error recently? I must have had ten (10) of these in the last week on three different computers. Notice the extra "U" on the end of the expected checksum. All of the errors I have seen have this extra "U".

Edit: More than I thought - Found 36 download errors since March 21

3/28/2013 2:17:53 PM | climateprediction.net | [error] MD5 check failed for hadcm3n_u6uq_2020_40_008339423.zip
3/28/2013 2:17:53 PM | climateprediction.net | [error] expected f092b025eafe9ddb67f505a15b7331c2U, got f092b025eafe9ddb67f505a15b7331c2
3/28/2013 2:17:53 PM | climateprediction.net | [error] Checksum or signature error for hadcm3n_u6uq_2020_40_008339423.zip

---

It looks like a new batch of models, so a new problem.
I'll notify the project.

I'm also getting this error on multiple computers runing clients 7.0.59 and 7.0.58

3/28/2013 10:51:24 PM | climateprediction.net | Finished download of DMSSO2NH3_19592001_RCP85v2.gz
3/28/2013 10:51:24 PM | climateprediction.net | Started download of atmos_zbqz_1920_40_008340957_0.gz
3/28/2013 10:51:35 PM | climateprediction.net | Finished download of ocean_zbqz_1920_40_008340957_0.gz
3/28/2013 10:51:39 PM | climateprediction.net | Finished download of atmos_zbqz_1920_40_008340957_0.gz
3/28/2013 10:51:39 PM | climateprediction.net | [error] MD5 check failed for atmos_zbqz_1920_40_008340957_0.gz
3/28/2013 10:51:39 PM | climateprediction.net | [error] expected 8454813f91458b124b708a1e6e8bae4e, got 8454813f91458b124b708a1e6e8bae4e
3/28/2013 10:51:39 PM | climateprediction.net | [error] Checksum or signature error for atmos_zbqz_1920_40_008340957_0.gz

I wonder whether the problem could possibly be with the version of BOINC. You have seven computers:

http://climateapps2.oerc.ox.ac.uk/cpdnboinc/hosts_user.php?userid=8040

The computers with 7.0.28 which is a public stable version can all download models properly. Just the occasional problem as happens to all of us.

The three computers with 7.0.58 which is an alpha version all have the MD5 downloading error. But computer 843661 is different. If you look at the second page of models which downloaded properly, they were downloaded when the computer had 7.0.28.

There's one model on that computer which says in its stderr file that it had 7.0.58 but from the dates of the model I think it probably had 7.0.28 when it was downloaded.

I don't think the error is specific to hadcm or to a particular batch of models; I've seen the MD5 error in a hadam EU as well. In some of WB's workunits another computer has downloaded the model OK and is crunching it.

I don't know whether suspecting the BOINC version is significant or just a coincidence. I haven't noticed any emails about MD5 download errors on the boinc_alpha list.

---

Cpdn news

I've just seen Art Masson's post which reinforces my suspicions. I'm going to ask Jorden and Richard Haselgrove to have a look at this thread.

---

Cpdn news

Just noticed that another of my machines running BOINC 7.0.55 just successfully downloaded a work unit....not sure if this error is related to BOINC client version or data related. The checksum error I'm seeing on BOINC 7.0.59 (and previously on 7.0.58) is showing an extra (unprintable) character (carriage return?) on the last character of target checksum...but otherwise the checksum is showing identical to the actual.

Art Masson
St. Charles, IL

Here is the successful log from my 7.0.55 client:

3/28/2013 5:03:45 AM | climateprediction.net | update requested by user
3/28/2013 5:03:58 AM | climateprediction.net | Sending scheduler request: Requested by user.
3/28/2013 5:03:58 AM | climateprediction.net | Requesting new tasks for CPU and NVIDIA
3/28/2013 5:04:00 AM | climateprediction.net | Scheduler request completed: got 0 new tasks
3/28/2013 5:04:00 AM | climateprediction.net | Project has no tasks available
3/28/2013 11:14:42 PM | climateprediction.net | update requested by user
3/28/2013 11:14:43 PM | climateprediction.net | Sending scheduler request: Requested by user.
3/28/2013 11:14:43 PM | climateprediction.net | Requesting new tasks for CPU and NVIDIA
3/28/2013 11:14:45 PM | climateprediction.net | Scheduler request completed: got 1 new tasks
3/28/2013 11:14:47 PM | climateprediction.net | Started download of hadcm3n_6.07_windows_intelx86.exe
3/28/2013 11:14:47 PM | climateprediction.net | Started download of hadcm3n_data_6.07_windows_intelx86.zip
3/28/2013 11:14:52 PM | climateprediction.net | Finished download of hadcm3n_6.07_windows_intelx86.exe
3/28/2013 11:14:52 PM | climateprediction.net | Finished download of hadcm3n_data_6.07_windows_intelx86.zip
3/28/2013 11:14:52 PM | climateprediction.net | Started download of hadcm3n_um_6.07_windows_intelx86.zip
3/28/2013 11:14:52 PM | climateprediction.net | Started download of hadcm3n_graphics_6.07_windows_intelx86.exe
3/28/2013 11:14:56 PM | climateprediction.net | Finished download of hadcm3n_um_6.07_windows_intelx86.zip
3/28/2013 11:14:56 PM | climateprediction.net | Finished download of hadcm3n_graphics_6.07_windows_intelx86.exe
3/28/2013 11:14:56 PM | climateprediction.net | Started download of hadcm3n_se_6.07_windows_intelx86.zip
3/28/2013 11:14:56 PM | climateprediction.net | Started download of hadcm3n_n6rd_1960_40_008341111.zip
3/28/2013 11:14:57 PM | climateprediction.net | Finished download of hadcm3n_n6rd_1960_40_008341111.zip
3/28/2013 11:14:57 PM | climateprediction.net | Started download of ocean_n6rd_1960_40_008341111_0.gz
3/28/2013 11:14:59 PM | climateprediction.net | Finished download of hadcm3n_se_6.07_windows_intelx86.zip
3/28/2013 11:14:59 PM | climateprediction.net | Started download of atmos_n6rd_1960_40_008341111_0.gz
3/28/2013 11:15:05 PM | climateprediction.net | Finished download of atmos_n6rd_1960_40_008341111_0.gz
3/28/2013 11:15:05 PM | climateprediction.net | Started download of spec3a_sw_3_asol2c_hadcm3.gz
3/28/2013 11:15:07 PM | climateprediction.net | Finished download of spec3a_sw_3_asol2c_hadcm3.gz
3/28/2013 11:15:07 PM | climateprediction.net | Started download of spec3a_lw_3_asol2c_hadcm3.gz
3/28/2013 11:15:08 PM | climateprediction.net | Finished download of spec3a_lw_3_asol2c_hadcm3.gz
3/28/2013 11:15:08 PM | climateprediction.net | Started download of waterfix.ancil.be.32.gz
3/28/2013 11:15:09 PM | climateprediction.net | Finished download of waterfix.ancil.be.32.gz
3/28/2013 11:15:09 PM | climateprediction.net | Started download of NAT_VOLC.gz
3/28/2013 11:15:10 PM | climateprediction.net | Finished download of NAT_VOLC.gz
3/28/2013 11:15:10 PM | climateprediction.net | Started download of DMSSO2NH3_19592001_RCP85v2.gz
3/28/2013 11:15:22 PM | climateprediction.net | Finished download of DMSSO2NH3_19592001_RCP85v2.gz
3/28/2013 11:15:22 PM | climateprediction.net | Started download of sulpc_oxidants_19_A2_1990f.gz
3/28/2013 11:15:29 PM | climateprediction.net | Finished download of ocean_n6rd_1960_40_008341111_0.gz
3/28/2013 11:15:29 PM | climateprediction.net | Started download of SPARC_O3_1960-2001.RCP8.5v2.gz
3/28/2013 11:15:33 PM | climateprediction.net | Finished download of SPARC_O3_1960-2001.RCP8.5v2.gz
3/28/2013 11:15:40 PM | climateprediction.net | Finished download of sulpc_oxidants_19_A2_1990f.gz
3/28/2013 11:15:42 PM | climateprediction.net | Starting task hadcm3n_n6rd_1960_40_008341111_0 using hadcm3n version 607 in slot 4

I've just downgraded my primary machine to 7.0.55 and will report if I now successfully download work units. I've looked at the on line log for this computer on the climatepredition.net site and I see lots of download errors starting March 16th. This might imply the problem started when I installed version 7.0.56 of the client.

Here is my BOINC client install history

current 7.0.55 (just downgraded)
3/28 -- installed 7.0.59 (definitely had check-sum problem)
3/22 -- installed 7.0.58 (definitely had check-sum problem)
3/15 -- installed 7.0.56
3/12 -- installed 7.0.55
3/7 -- installed 7.0.54
2/15 -- installed 7.0.52
2/08 -- installed 7.0.48
2/03 -- installed 7.0.47

Hope this helps...if I get a successful download on 7.0.55 now on my primary computer I will post.

Art Masson

Sorry...my "primary" machine is computer ID 1266353 (HP-CORE-I7)
This is the machine for the upgrade record above.

Art Masson

Les and I have asked two people who know more than we do about BOINC testing versions to look at what you've both said.

Have either of you got a computer with 7.0.58 or 7.0.59 that can't download from CPDN but can download from other projects? (I could work out the answer myself but you will know immediately.)

---

Cpdn news

Yes. My clients running 7.0.55 and previously 7.0.59 and 7.0.58 are successfully running projects from SETI@HOME, EINSTEIN@HOME, ENGIMA@HOME, and MILKYWAY@HOME.

The copy/paste from the event log (below) is not showing the "funny" character at the end of the expected check-sum....I have a graphic of this I could provide...but can't quite figure out how to attach this jpg in this message....

My main computer now running 7.0.55 just had the check-sum failures on additional downloads. Here is the log:

3/29/2013 4:15:43 AM | climateprediction.net | update requested by user
3/29/2013 4:15:43 AM | climateprediction.net | Sending scheduler request: Requested by user.
3/29/2013 4:15:43 AM | climateprediction.net | Requesting new tasks for CPU and NVIDIA
3/29/2013 4:15:45 AM | climateprediction.net | Scheduler request completed: got 2 new tasks
3/29/2013 4:15:47 AM | climateprediction.net | Started download of hadcm3n_4j5l_1980_40_008341281.zip
3/29/2013 4:15:47 AM | climateprediction.net | Started download of ocean_4j5l_1980_40_008341281_0.gz
3/29/2013 4:15:49 AM | climateprediction.net | Finished download of hadcm3n_4j5l_1980_40_008341281.zip
3/29/2013 4:15:49 AM | climateprediction.net | Started download of atmos_4j5l_1980_40_008341281_0.gz
3/29/2013 4:15:49 AM | climateprediction.net | [error] MD5 check failed for hadcm3n_4j5l_1980_40_008341281.zip
3/29/2013 4:15:49 AM | climateprediction.net | [error] expected fb2fdf85b671d23ba0b6e784c2db787b, got fb2fdf85b671d23ba0b6e784c2db787b
3/29/2013 4:15:49 AM | climateprediction.net | [error] Checksum or signature error for hadcm3n_4j5l_1980_40_008341281.zip
3/29/2013 4:15:57 AM | climateprediction.net | Finished download of atmos_4j5l_1980_40_008341281_0.gz
3/29/2013 4:15:57 AM | climateprediction.net | Started download of DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 4:15:57 AM | climateprediction.net | [error] MD5 check failed for atmos_4j5l_1980_40_008341281_0.gz
3/29/2013 4:15:57 AM | climateprediction.net | [error] expected 2ca2a12b146ff96e5703eb86f05723b0, got 2ca2a12b146ff96e5703eb86f05723b0
3/29/2013 4:15:57 AM | climateprediction.net | [error] Checksum or signature error for atmos_4j5l_1980_40_008341281_0.gz
3/29/2013 4:15:59 AM | climateprediction.net | Finished download of DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 4:15:59 AM | climateprediction.net | Started download of hadcm3n_4f1f_1980_40_008341290.zip
3/29/2013 4:15:59 AM | climateprediction.net | [error] MD5 check failed for DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 4:15:59 AM | climateprediction.net | [error] expected c8d7bf9a026379061f12fe8589a93340, got c8d7bf9a026379061f12fe8589a93340
3/29/2013 4:15:59 AM | climateprediction.net | [error] Checksum or signature error for DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 4:16:00 AM | climateprediction.net | Finished download of hadcm3n_4f1f_1980_40_008341290.zip
3/29/2013 4:16:00 AM | climateprediction.net | Started download of ocean_4f1f_1980_40_008341290_0.gz
3/29/2013 4:16:00 AM | climateprediction.net | [error] MD5 check failed for hadcm3n_4f1f_1980_40_008341290.zip
3/29/2013 4:16:00 AM | climateprediction.net | [error] expected b5253bbb65ffa9141c36a239166146c1, got b5253bbb65ffa9141c36a239166146c1
3/29/2013 4:16:00 AM | climateprediction.net | [error] Checksum or signature error for hadcm3n_4f1f_1980_40_008341290.zip
3/29/2013 4:16:06 AM | climateprediction.net | Finished download of ocean_4j5l_1980_40_008341281_0.gz
3/29/2013 4:16:06 AM | climateprediction.net | Started download of atmos_4f1f_1980_40_008341290_0.gz
3/29/2013 4:16:15 AM | climateprediction.net | Finished download of atmos_4f1f_1980_40_008341290_0.gz
3/29/2013 4:16:15 AM | climateprediction.net | [error] MD5 check failed for atmos_4f1f_1980_40_008341290_0.gz
3/29/2013 4:16:15 AM | climateprediction.net | [error] expected 61824c828e7810814377cec211bd6da3, got 61824c828e7810814377cec211bd6da3
3/29/2013 4:16:15 AM | climateprediction.net | [error] Checksum or signature error for atmos_4f1f_1980_40_008341290_0.gz
3/29/2013 4:16:20 AM | climateprediction.net | Finished download of ocean_4f1f_1980_40_008341290_0.gz

[/img]

I'm going to try downgrading my client to 7.0.28 and see if that makes any difference.

I have three Windows computers with 7.0.58. CPDN is the only project getting the MD5 errors. SETI, Milkyway, and Einstein work OK.

Also, all CPDN downloads do not get the error. Once in a while the download is successful.

If I "downgrade" to an earlier version of BOINC, do the workunits keep running. I seem to remember I read that all the workunits abort on a downgrade.

---

I thinkyou only lose all the running tasks if you downgrade from BOINC 7 to BOINC 6. This is what Jorden says in the BOINC 7 FAQ here in the section called Returning to BOINC 6. Different versions of BOINC 7 should be compatible with each other.

http://boincfaq.mundayweb.com/index.php?language=1&view=592&sessionID=e4d421bff1d42c68ff1480e002f59c1c


You could wait to see whether Art is successful in downgrading to the public, stable version of BOINC 7. Do not even think about downgrading to BOINC 6.

---

Cpdn news

I've successfully downloaded two work units on 7.0.28 on my primary machine. So not sure what client release level the problem starts (somewhere between 7.0.28 and 7.0.55 I assume). Strange that my other machine running 7.0.55 did successfully download two units last night and they are still running OK.

..so this problem has some kind of a random impact or data-driven component. For whatever it's worth both machines are running 64-bit Windows 7 and I'm using the 64 bit version of BOINC. At this point I have two active BOINC clients processing climate-prediction work units -- one on 7.0.28 and one of 7.0.55.

Will you guys report this...or do I need to log the issue as an alpha users with BOINC ???? If you want to see a capture from my log which shows the "strange" character, please give me an email address to forward a jpg screen captures to.

Here is the successful log dump on 7.0.28:

3/29/2013 8:54:49 AM | climateprediction.net | update requested by user
3/29/2013 8:54:52 AM | climateprediction.net | Sending scheduler request: Requested by user.
3/29/2013 8:54:52 AM | climateprediction.net | Requesting new tasks for CPU
3/29/2013 8:54:54 AM | climateprediction.net | Scheduler request completed: got 2 new tasks
3/29/2013 8:54:56 AM | climateprediction.net | Started download of hadcm3n_z7vy_1960_40_008340938.zip
3/29/2013 8:54:56 AM | climateprediction.net | Started download of ocean_z7vy_1960_40_008340938_0.gz
3/29/2013 8:54:58 AM | climateprediction.net | Finished download of hadcm3n_z7vy_1960_40_008340938.zip
3/29/2013 8:54:58 AM | climateprediction.net | Started download of atmos_z7vy_1960_40_008340938_0.gz
3/29/2013 8:55:07 AM | climateprediction.net | Finished download of atmos_z7vy_1960_40_008340938_0.gz
3/29/2013 8:55:07 AM | climateprediction.net | Started download of DMSSO2NH3_19592001_RCP85v2.gz
3/29/2013 8:55:17 AM | climateprediction.net | Finished download of ocean_z7vy_1960_40_008340938_0.gz
3/29/2013 8:55:17 AM | climateprediction.net | Finished download of DMSSO2NH3_19592001_RCP85v2.gz
3/29/2013 8:55:17 AM | climateprediction.net | Started download of SPARC_O3_1960-2001.RCP8.5v2.gz
3/29/2013 8:55:17 AM | climateprediction.net | Started download of hadcm3n_4bqi_1980_40_008340381.zip
3/29/2013 8:55:19 AM | climateprediction.net | Finished download of hadcm3n_4bqi_1980_40_008340381.zip
3/29/2013 8:55:19 AM | climateprediction.net | Started download of ocean_4bqi_1980_40_008340381_0.gz
3/29/2013 8:55:20 AM | climateprediction.net | Finished download of SPARC_O3_1960-2001.RCP8.5v2.gz
3/29/2013 8:55:20 AM | climateprediction.net | Started download of atmos_4bqi_1980_40_008340381_0.gz
3/29/2013 8:55:20 AM | climateprediction.net | Starting task hadcm3n_z7vy_1960_40_008340938_2 using hadcm3n version 607 in slot 11
3/29/2013 8:55:28 AM | climateprediction.net | Finished download of atmos_4bqi_1980_40_008340381_0.gz
3/29/2013 8:55:28 AM | climateprediction.net | Started download of DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 8:55:30 AM | climateprediction.net | Finished download of DMSSO2NH3_1900_RCP.anc_be.be.32.gz
3/29/2013 8:55:37 AM | climateprediction.net | Finished download of ocean_4bqi_1980_40_008340381_0.gz

I think I can see a problem which might be accounting for this - there seem to be faults both at the CPDN server end, and with the new client. Maybe that's why it hasn't been observed before - one error by itself is survivable, both together are a train wreck. I need to do a bit more investigation and writing up, then I'll let both groups know what I've found.

I'm sending you my email address by PM, Art.

I was about to ask you to hold off reporting this directly to the BOINC alpha email list until one of the people familiar with both BOINC and CPDN (and the current problem) has seen this thread, particularly as it's a long weekend in Europe until Tuesday. And as I had hoped, Richard has appeared.

For those not aware of the problem I've referred to, please have a look at the last few posts in the News thread at the top of Number Crunching.

---

Cpdn news

Thanks! I'll send you the jpg.
I'll hold off reporting this further.

One question...should I let my one machine currently running two units on 7.0.55 continue on that client level, or downgrade it to 7.0.28?. I would not want it to fail on uploading later and lose the processing...or is this problem confined to downloads??

I have three machines actively processing climate-predication units

1175567 on 7.0.55 -- processing two units (local machine)
1266353 on 7.0.28 -- processing two units (my local primary machine)
1267479 on 7.0.44 -- processing three units (this unit is remote)

Art Masson

I downgraded my three computers to 7.0.44 (the last version I had). I believe that version was working correctly.

It will be a day or so before I can draw any conclusions because of the one workunit per day rule (failed downloads count).

---

To display an image:

Open a new message window.
Looking at the words to the left of the window, click on Use BBCode tags to format your text.
Read the line about displaying an image.

PS
Anyone can look at this without needing to post.

Just quit from message window with the red X way up in the top right hand corner of the screen once you've looked at the BBCodes.
Or use the "go back to a previous page" browser button in the top left hand corner of the screen. (Should be a left pointing arrow.)