SMP Update

[SteveI]

Don't get your hopes up. There appears to be no bug fixes, only a deadline extension. I am attaching the update file only. If you need to load SMP from scratch, please download from the FAH site.

Stop the core using ctl+C, just replace the exe, and restart. Do this only with a new WU.... just in case it corrupts and restarts on you..... Have fun Thunder, lol. You've got a lot of machines to update.

[ThunderRd]

Well, ain't that grand. I was hoping for more.

They keep talking about the *major* release. I'm not holding my breath. Right now I'm up to my eyeballs in network problems since my ISP decided to *improve* their service several days ago. Hanging clients, lost WUs, EOEs, I/O errors, the lot. It is MUCH more sensitive to network changes than I ever expected.

I'm slowly getting control again, after I "discovered" that they changed their primary DNS server address. Hope to have the aircraft all flying soon.

[keithwalton]

Thanks Steve, i'll update the quad next wu

[Samuknow]

Originally Posted by ThunderRD

Well, ain't that grand. I was hoping for more.

They keep talking about the *major* release. I'm not holding my breath. Right now I'm up to my eyeballs in network problems since my ISP decided to *improve* their service several days ago. Hanging clients, lost WUs, EOEs, I/O errors, the lot. It is MUCH more sensitive to network changes than I ever expected.

I'm slowly getting control again, after I "discovered" that they changed their primary DNS server address. Hope to have the aircraft all flying soon.
___________

Those bastages. Don't they know what they are screwing with?

[ThunderRd]

Originally Posted by Samuknow

Those bastages. Don't they know what they are screwing with?

Should replace "what" with "WHO". As in "our team".

WUs are starting to move now. Today is an improvement, tomorrow should be better. If all goes well I'll be back at 27k over the weekend.

[Samuknow]

Yee Haa. 27K.

That is what I like to see. Should have me a temporary 939 board to get me up and going soon. Hopefully before the weekend. I do have my eye on a A8N32 SLI Delux for about 100 bucks. Seems like a sweet ride.

[ccperf721p]

I'm trying to keep up with Sam. We've been playing leap frog the last few days. Not going to get the AM2 setup going (offer I can't refuse) so I will have to pull out the little guns. T2300 Yonah on a micro-atx board. If I can keep it cool it will run with an A64 at the same speed, but max clock is 2.4..

[SteveI]

When I looked around, it seems that the fixes are still works in progress.

While the update is not necessary atm, it is still a good idea. The current exe expires soon, maybe the first week of october(?) .... not sure. But the last thing we need is to find a core hanging for a few days after the expiration.

Updated 4 rigs.... 2 to go. While I haven't had a single problem, Im still waiting until WUs finish.


@Thunder,
Careful with any IP resets/changes, could cause WUs to dump.

[Samuknow]

So your CPDME. Did not catch on to that until now...

Once I get my x2 back up, it will be a little harder to keep up......he he

[ccperf721p]

I am just wrapping up the OC on my Q6600 so I will be loading Gentoo on it over the next couple of days and bring it into the mix over the weekend..

[Samuknow]

So much for my theory of staying ahead of you....LOL

[ccperf721p]

I'll be thinning out my rigs so it may end up equal or close by the time I am actually finished. Just too many around here so I will sell them off and donate some stuff to someone that will fold.

[Samuknow]

Send a couple this way. You know I will keep em crunching....LOL

[keithwalton]

I have now had two wu's EUE at 86% whilst running the smp core, both times they were a 2652 wu and another at 45% in the last two days. All since i updated the client, cpu clocks wise they are the same as they have been for the last few months.
I have completed a 3050 units in between though.

This is todays error

[20:25:33] Completed 850000 out of 1000000 steps (85 percent)
[20:33:46] Writing local files
[20:33:46] Completed 860000 out of 1000000 steps (86 percent)
[20:37:25] Warning: long 1-4 interactions
[20:37:25] Quit 101 - NaN detected: (ener[20])
[20:37:25]
[20:37:25] Simulation instability has been encountered. The run has entered a
[20:37:25] state from which no further progress can be made.
[20:37:25] This may be the correct result of the simulation, however if you
[20:37:25] often see other project units terminating early like this
[20:37:25] too, you may wish to check the stability of your computer (issues
[20:37:25] such as high temperature, overclocking, etc.).
[20:37:25] Going to send back what have done.
[20:37:25] logfile size: 493860
[20:37:25] - Writing 494410 bytes of core data to disk...
[20:37:25] ... Done.
[20:37:25] - Failed to delete work/wudata_08.arc
[20:37:25] No C.P. to delete.
[20:37:25] - Failed to delete work/wudata_08.bed
[20:37:25] - Failed to delete work/wudata_08.chk
[20:37:25] - Failed to delete work/wudata_08.goe
[20:37:25] - Failed to delete work/wudata_08.xvg
[20:37:25] Warning: check for stray files
[20:39:25]
[20:39:25] Folding@home Core Shutdown: EARLY_UNIT_END
[20:39:25]
[20:39:25] Folding@home Core Shutdown: EARLY_UNIT_END
[20:39:28] CoreStatus = 7B (123)
[20:39:28] Client-core communications error: ERROR 0x7b
[20:39:28] Deleting current work unit & continuing...

and these are yesterdays

[16:44:44] Completed 440000 out of 1000000 steps (44 percent)
[16:53:13] Writing local files
[16:53:13] Completed 450000 out of 1000000 steps (45 percent)
[16:55:49] Warning: long 1-4 interactions
[16:55:49] Gromacs cannot continue further.
[16:55:49] Going to send back what have done.
16:55:49] logfile size: 237560
[16:55:49] - Writing 238096 bytes of core data to disk...
[16:55:49] ... Done.
[16:55:49] - Failed to delete work/wudata_06.arc
[16:55:49] No C.P. to delete.
[16:55:49] - Failed to delete work/wudata_06.bed
[16:55:49] - Failed to delete work/wudata_06.xvg
[16:55:49] Warning: check for stray files
[16:57:49]
[16:57:49] Folding@home Core Shutdown: EARLY_UNIT_END
[16:57:49]
[16:57:49] Folding@home Core Shutdown: EARLY_UNIT_END
[16:57:53] CoreStatus = 7B (123)
[16:57:53] Client-core communications error: ERROR 0x7b
[16:57:53] Deleting current work unit & continuing...

[10:26:39] Completed 850000 out of 1000000 steps (85 percent)
[10:34:39] Writing local files
[10:34:39] Completed 860000 out of 1000000 steps (86 percent)
[10:38:13] Warning: long 1-4 interactions
[10:38:14] Quit 101 - NaN detected: (ener[0])
[10:38:14]
[10:38:14] Simulation instability has been encountered. The run has entered a
[10:38:14] state from which no further progress can be made.
[10:38:14] This may be the correct result of the simulation, however if you
[10:38:14] often see other project units terminating early like this
[10:38:14] too, you may wish to check the stability of your computer (issues
[10:38:14] such as high temperature, overclocking, etc.).
[10:38:14] Going to send back what have done.
[10:38:14] logfile size: 320542
[10:38:14] - Writing 321091 bytes of core data to disk...
[10:38:14] ... Done.
[10:40:14]
[10:40:14] Folding@home Core Shutdown: EARLY_UNIT_END
[10:40:14]
[10:40:14] Folding@home Core Shutdown: EARLY_UNIT_END
[10:40:18] CoreStatus = 7B (123)
[10:40:18] Client-core communications error: ERROR 0x7b
[10:40:18] Deleting current work unit & continuing...

[ThunderRd]

I have seen that error three times on my machines and logged the event; my backup routine saved the WU each and every time. I just rolled back to an hour or so before it plotzed and started the client again-and it finished after that.

[SteveI]

FYI...
I forgot to update my notebook... It's been idling for a day after my kids reboot it.

Updated the core, and back folding again.

[Samuknow]

Those sneaky kids.....LOL

I have to watch them here too.

[keithwalton]

its seems to be going again fine here, though i havent had anymore of those 2652 units. i seem to remember these causing my client to become unstuck before.