https://boinc.bakerlab.org/rosetta/workunit.php?wuid=258395540
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=258395513
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=258389967

probably the whole series of broken ....

got two same error here.

https://boinc.bakerlab.org/rosetta/result.php?resultid=283583904
https://boinc.bakerlab.org/rosetta/result.php?resultid=283586480

I'm joining the club here with these "histone" work units, and as I see I'm not the only one. I did have one return fine the other day, this is only the second one I have run, but I have 3 more waiting to run. I'd really not like to waste time running them if they are just going to screw up and return errors.
https://boinc.bakerlab.org/rosetta/result.php?resultid=283455906

Tell the server computer that they can have their "histones" back and I'll take a few more of those "Rosetta Beta" work units in their place. 8-)

---

I'm joining the club here with these "histone" work units, and as I see I'm not the only one. I did have one return fine the other day, this is only the second one I have run, but I have 3 more waiting to run. I'd really not like to waste time running them if they are just going to screw up and return errors.
https://boinc.bakerlab.org/rosetta/result.php?resultid=283455906

Tell the server computer that they can have their "histones" back and I'll take a few more of those "Rosetta Beta" work units in their place. 8-)

I'm not sure the rosetta_beta workunits are much better - during the only one I've had lately, the graphics froze and wouldn't go away when I moved the cursor or used the keyboard. It resumed after I did a reboot without using the keyboard or the mouse, though, and I haven't checked if the computation part finished without an error.

That's why I don't let the graphics run as a screensaver at all, and it's a major waste of processing power. It's just easier to have no screensaver, have the video unit power down after 10 minutes of inactivity, and just shut the monitor off when I'm done with the computer. I never understood why anyone would want to leave the monitor on all night long 24/7.

And I have had another "histone" work unit error out on me. I'm just going to abort any of them that get sent my way. I don't feel like wasting time running junk work units.

---

That's why I don't let the graphics run as a screensaver at all, and it's a major waste of processing power. It's just easier to have no screensaver, have the video unit power down after 10 minutes of inactivity, and just shut the monitor off when I'm done with the computer. I never understood why anyone would want to leave the monitor on all night long 24/7.

And I have had another "histone" work unit error out on me. I'm just going to abort any of them that get sent my way. I don't feel like wasting time running junk work units.

Some of the more recent versions of BOINC, including 6.6.36, include a screensaver that goes through the graphics sections of the various workunits in progress, even if the monitor is turned off and the user does not ask for graphics.

Since my computers are in my bedroom, I definitely don't want the monitors on at night, and therefore turn them off. I seldom ask for any other workunit graphics.

Note - the results of the few histone workunits my computers have run, and the results from wingmates, suggest that they may run properly under Windows XP but not under 64-bit Windows Vista. Those of you reporting failed histone workunits may want to include a mention of which type of operating system you're using, in order to check this idea.

Also, any of you interested in histone workunits from previous years may want to search for histone in this thread:

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=1453#14309

And I have had another "histone" work unit error out on me. I'm just going to abort any of them that get sent my way. I don't feel like wasting time running junk work units.

Note - the results of the few histone workunits my computers have run, and the results from wingmates, suggest that they may run properly under Windows XP but not under 64-bit Windows Vista. Those of you reporting failed histone workunits may want to include a mention of which type of operating system you're using, in order to check this idea.

I'm running them now with Vista SP2 and having no problems. I can view properties which show a past checkpoint and see the graphics window.

The one issue is that it I'm currently on Model 0 Step 3875.

Oops. Just double checked and after 20minutes it's on Model 1 Step 920. Is it a problem? Not that I can recognise.

Running Windows 7

---

And I have had another "histone" work unit error out on me. I'm just going to abort any of them that get sent my way. I don't feel like wasting time running junk work units.

Note - the results of the few histone workunits my computers have run, and the results from wingmates, suggest that they may run properly under Windows XP but not under 64-bit Windows Vista. Those of you reporting failed histone workunits may want to include a mention of which type of operating system you're using, in order to check this idea.

I'm running them now with Vista SP2 and having no problems. I can view properties which show a past checkpoint and see the graphics window.

32-bit or 64-bit? Both of my failed histone workunits ran on my 64-bit Vista SP2 machine.

32 bit

---

And I would like to mention that I had one histone that did get some granted credit, but it also ran for a very short period of time.
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=258264335

---

A new version of that idea: Histone workunits tend to run well under 32-bit versions of Vista, but not 32-bit XP, 64-bit Vista, 32-bit Windows 7, or 64-bit Windows 7.

Another possibility: they tend to fail under some of the newer versions of BOINC, such as the 64-bit 6.6.36 I'm using.

A new version of that idea: Histone workunits tend to run well under XP or 32-bit versions of Vista, but not 64-bit versions of Vista or 32-bit Windows 7 or 64-bit Windows 7.

Another possibility: they tend to fail under some of the newer versions of BOINC, such as the 64-bit 6.6.36 I'm using.

Sounds like you're clutching at straws here Robert.

I'm running Vista 64 SP2 with Boinc 6.6.36 and having no problems, aside from very short runtimes despite very few (2) decoys.

histone_loopbuild_run1_14925_72075_1
histone_loopbuild_run1_14925_75345_0
histone_loopbuild_run1_14925_73945_0
histone_loopbuild_run1_14925_24562_1
histone_loopbuild_run1_14925_19382_0

Checking through my last 186 completed jobs I've come up with only 2 errors, both coming from the same type of job.

sel_core_5.0_low50_beta_low200_start_hb_t374__IGNORE_THE_REST_14879_226_1
sel_core_5.0_low50_beta_low200_start_hb_t374__IGNORE_THE_REST_14879_799_1

- Unhandled Exception Record -
Reason: Access Violation (0xc0000005) at address 0x005A7230 read attempt to address 0x00000000

---

And I have had another "histone" work unit error out on me. I'm just going to abort any of them that get sent my way. I don't feel like wasting time running junk work units.

Note - the results of the few histone workunits my computers have run, and the results from wingmates, suggest that they may run properly under Windows XP but not under 64-bit Windows Vista. Those of you reporting failed histone workunits may want to include a mention of which type of operating system you're using, in order to check this idea.

I'm running them now with Vista SP2 and having no problems. I can view properties which show a past checkpoint and see the graphics window.

32-bit or 64-bit? Both of my failed histone workunits ran on my 64-bit Vista SP2 machine.

32-bit SP2 here.

A new version of that idea: Histone workunits tend to run well under XP or 32-bit versions of Vista, but not 64-bit versions of Vista or 32-bit Windows 7 or 64-bit Windows 7.

Another possibility: they tend to fail under some of the newer versions of BOINC, such as the 64-bit 6.6.36 I'm using.

Sounds like you're clutching at straws here Robert.

I'm running Vista 64 SP2 with Boinc 6.6.36 and having no problems, aside from very short runtimes despite very few (2) decoys.

Checking through my last 186 completed jobs I've come up with only 2 errors, both coming from the same type of job.

Now looks like the actual pattern is complicated enough that I'd better leave future guesses to those who have access to a much larger collection of histone workunit results than I do.

Does running them, even if they have too many errors to get granted credit when validated, still help the project team figure out what the real problem with them is?

---

Typically only a small fraction of the models run in to such problems. If everyone started canceling everything that anyone ever had a problem with, they would not only be wasting resources and time, but the project would have no way to assess the scale of the problem.

So, yes, running things, even when they fail, helps the project. This is why credit is given even for failures. On the other hand, you can't encourage failure or people will waste server resources creating more of them.

On the other hand, once you see several posts all pointing to specific problems with specific WUs, it could be wise to suspend the ones of that name that you already have and let other work run until the Project Team has had time to assess and respond with any recommendation to cancel specific tasks. This is one advantage of having a cache of about a day of work.

---

Rosetta Moderator: Mod.Sense

So, yes, running things, even when they fail, helps the project. This is why credit is given even for failures. On the other hand, you can't encourage failure or people will waste server resources creating more of them.

On the other hand, once you see several posts all pointing to specific problems with specific WUs, it could be wise to suspend the ones of that name that you already have and let other work run until the Project Team has had time to assess and respond with any recommendation to cancel specific tasks. This is one advantage of having a cache of about a day of work.

It looks like there's some advantage in having three hands too... ;)

It looks like there's some advantage in having three hands too... ;)

...it's the only way to keep up with the work! :)

---

Rosetta Moderator: Mod.Sense

This is why credit is given even for failures.

I've never been given credit for the failure/error WU, they just say "computer error" or "validation error" and there is no granted or claimed credit. That's why I saw it as a waste of good computing time.
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=258549295

---