Is there anyway to see in the standard BoincManager what the vatious Debts are?

My concern is that there may be a logic flaw in the Manager that I stumbled into. Would my machine just have kept thrashing back and forth, until the H/D failed? It's just blind luck that I was here watching when it started...

The only way to see the debts without an add-on is looking through the client_stat.xml file manually.

It is unlikely that it would have kept thrashing. It is common for a new install or a one project host that just had a second project added to thrash like that for a bit before it settles into a rythym. The reason was stated in another post, the STD is very close together and close to zero so any little thing can cause it to decide it is time to switch. Another reason is that each time a file downloads the client re-evaluates what it is working on just in case you get a task with a very tight deadline.

---

BOINC WIKI

BOINCing since 2002/12/8

Update to my last post:

Since the team just a made a post to the general public regarding the upcoming release of ESAH, I decided it might be a good time to give it a try over at SETI Beta.

Now that I have a third project attached to one of my "slugs" the STD displays normally in BV (no negative values), however the question about what the story is with negative values in the 2 project scenario remains.

<edit> I take that back I am still seeing negative vales for STD even with 3 projects.

Alinator

Understood, on point one. I was wondering if the task switching would have stopped when all the DL's had finished, but it looks like you interceeded before that happened. FWIW, I have never personally seen BOINC task switch every second like that in the past.

Yes STD is short term debt, but unfortunately it is not displayed with the stock Manager. You can check it by looking at the client_state.xml file with IE, even while BOINC is running however.

I put a question up regarding this STD reporting difference between 4x and 5x on the BOINC message board, but haven't gotten an answer to it yet, or at least the last time I checked yesterday.

Alinator

Just a quick question John,

Since you're on DU, how many results were you expecting to get from SAH when you connected this time?

Also, related to WK's remarks about STD, I have noticed on my hosts with 5x the way BV displays the STD has changed from what you saw with 4x. What I mean is you never saw negative values in a 2 project scenario, whereas now it shows like the LTD. Maybe this has a bearing on what you saw.

Alinator

First Point:
I just d/u and pray. I have no idea whether I'll get two or twenty. I do about 30 a day on this machine, and I've set a 1.5 day cache set in Preferences. This time I had about 15, got 5, and 10 minutes later, 6 more.

Second Point:
Is STD Short Term Debt? I mean, besides being Se... But I digress...

Is there anyway to see in the standard BoincManager what the vatious Debts are?

My concern is that there may be a logic flaw in the Manager that I stumbled into. Would my machine just have kept thrashing back and forth, until the H/D failed? It's just blind luck that I was here watching when it started...

---

Just a quick question John,

Since you're on DU, how many results were you expecting to get from SAH when you connected this time?

Also, related to WK's remarks about STD, I have noticed on my hosts with 5x the way BV displays the STD has changed from what you saw with 4x. What I mean is you never saw negative values in a 2 project scenario, whereas now it shows like the LTD. Maybe this has a bearing on what you saw.

Alinator

I think the reason for this behavior is because when a unit is downloaded BOINC looks to see if it should start processing that project, at the same time I guess the STD for both projects was almost equal and the few seconds crunching flipped the switch saying crunch the other project.
I've seen my computer switch projects before the expected time because a downloaded caused a request_reschedule_cpus.

---

Hi... I'm on dial-up access, running Set and Einstein with equal weight, 24/7 right now, thru dial-up access, using BOINC 5.2.12. This morning I was uploading the overnight work, and downloading more, when the following occured (starting at 8:06:01):

09/03/2006 8:00:19 AM|SETI@home|Started download of 01jl01ab.4166.6465.673552.1.195
09/03/2006 8:00:20 AM||request_reschedule_cpus: files downloaded
09/03/2006 8:02:57 AM|SETI@home|Finished download of 11ap03ab.5070.15665.642322.1.145
09/03/2006 8:02:57 AM|SETI@home|Throughput 2279 bytes/sec
09/03/2006 8:02:57 AM|SETI@home|Started download of 13ap03ab.6621.6946.504824.1.60
09/03/2006 8:02:58 AM||request_reschedule_cpus: files downloaded
09/03/2006 8:02:59 AM|SETI@home|Finished download of 01jl01ab.4166.6465.673552.1.195
09/03/2006 8:02:59 AM|SETI@home|Throughput 2272 bytes/sec
09/03/2006 8:02:59 AM|SETI@home|Started download of 14ap03aa.15728.33345.386086.1.230
09/03/2006 8:03:00 AM||request_reschedule_cpus: files downloaded
09/03/2006 8:04:03 AM||request_reschedule_cpus: project op
09/03/2006 8:06:00 AM|SETI@home|Finished download of 14ap03aa.15728.33345.386086.1.230
09/03/2006 8:06:00 AM|SETI@home|Throughput 2025 bytes/sec
09/03/2006 8:06:01 AM||request_reschedule_cpus: files downloaded
09/03/2006 8:06:01 AM|SETI@home|Restarting result 11ap03aa.6531.688.15918.1.194_2 using setiathome version 418
09/03/2006 8:06:01 AM|Einstein@Home|Pausing result r1_0226.0__1220_S4R2a_0 (removed from memory)
09/03/2006 8:06:02 AM||request_reschedule_cpus: process exited
09/03/2006 8:06:02 AM|SETI@home|Pausing result 11ap03aa.6531.688.15918.1.194_2 (removed from memory)
09/03/2006 8:06:02 AM|Einstein@Home|Restarting result r1_0226.0__1220_S4R2a_0 using albert version 437
09/03/2006 8:06:03 AM||request_reschedule_cpus: process exited
09/03/2006 8:06:03 AM|SETI@home|Restarting result 11ap03aa.6531.688.15918.1.194_2 using setiathome version 418
09/03/2006 8:06:03 AM|Einstein@Home|Pausing result r1_0226.0__1220_S4R2a_0 (removed from memory)
09/03/2006 8:06:04 AM||request_reschedule_cpus: process exited
09/03/2006 8:06:04 AM|SETI@home|Pausing result 11ap03aa.6531.688.15918.1.194_2 (removed from memory)
09/03/2006 8:06:04 AM|Einstein@Home|Restarting result r1_0226.0__1220_S4R2a_0 using albert version 437
09/03/2006 8:06:05 AM||request_reschedule_cpus: process exited
09/03/2006 8:06:05 AM|SETI@home|Restarting result 11ap03aa.6531.688.15918.1.194_2 using setiathome version 418
09/03/2006 8:06:05 AM|Einstein@Home|Pausing result r1_0226.0__1220_S4R2a_0 (removed from memory)
09/03/2006 8:06:06 AM||request_reschedule_cpus: process exited
09/03/2006 8:06:06 AM|SETI@home|Pausing result 11ap03aa.6531.688.15918.1.194_2 (removed from memory)
09/03/2006 8:06:06 AM|Einstein@Home|Restarting result r1_0226.0__1220_S4R2a_0 using albert version 437

And it kept right on going... Flipping between the two apps over 50 times without any sign of stopping. During the flipping, BOINC reported the successful download of another Result, and just kept on going. I stopped it at 8:07:35 by Suspending SETI, leaving BOINC with nothing to do but Einstein.

Just now, I Resumed Seti:

09/03/2006 9:20:24 AM|Einstein@Home|Computation for result r1_0226.0__1220_S4R2a_0 finished
09/03/2006 9:20:24 AM|Einstein@Home|Starting result r1_0226.0__1219_S4R2a_0 using albert version 437
09/03/2006 9:20:26 AM|Einstein@Home|Started upload of r1_0226.0__1220_S4R2a_0_0
09/03/2006 9:21:22 AM|Einstein@Home|Finished upload of r1_0226.0__1220_S4R2a_0_0
09/03/2006 9:21:22 AM|Einstein@Home|Throughput 2618 bytes/sec
09/03/2006 9:28:00 AM||request_reschedule_cpus: project op
09/03/2006 9:28:01 AM|SETI@home|Restarting result 11ap03aa.6531.688.15918.1.194_2 using setiathome version 418
09/03/2006 9:28:01 AM|Einstein@Home|Pausing result r1_0226.0__1219_S4R2a_0 (removed from memory)

BOINC flipped me out of Einstein into Seti, and and everything seems stable...

Any explanation would be appreciated...

Edited for BBS code...

---