Did the Detach/Reattach thing last night on this computer last night to clear around 1800 ghosts. When the server came back up, at 00:47 UTC I was allocated 31 new units but none of them appear on the machine.

T.A.

its going to happen every 7 day cycle ... it's a systemic issue.

---

This is something on the Seti radar at the moment......
It has not been totally figured out yet.
Seems to be related to the bandwidth and the servers being maxxed out, which they probably be for some time to come, given the extent of the latest outage.

I don't think there is a definitive answer as to why they occur yet from the Seti staff.

But they are certainly working on it.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Its going to happen every 7 day cycle ... it's a systemic issue.

What makes you say that? And what kind of a systemic issue, do you mean?
The capacity and speed of the UP & Down Load server's.

The creation (rate) of Ghosts, is it pinned down and understood?
I noticed having Ghosts, as well, but it isn't clear wheather it's
SERVER problem, a bandwith problem, or both.

Just got a second load of MB tasks on my VISTA host (Q6600+GTS250),
it takes some time, but has come through.

Only I can not UP-Load, MB tasks on SETI Bêta, cause it's still down,
hope they don't time out! Would be a pitty and a waste of recourses.
Maybe it'll be fixed before the weekend, but got a few days, before they time out.
Anyway, SETI@home is UP & Running, the SETI crew did a great job.

---

While they may still have a ghost unit creation issue (wow computer is sluggish, welcome back work units!) I look at what I got and I got tons of 2's and 3's.. So hopefully we can make some progress in getting that glut knocked down.

some ghosts might be inevitable, but some things have definately changed. It will be interesting seeing how much so!

---

Janice

Speaking of ghosts I just checked my i7. OMG It says I have 1272 tasks in progress, Well My machine only has 6 running. And here I wondered why I was getting one new work unit at a time. Most wont time out untill late Oct.

So I am thinking I should Just do a dettach and reattach. So here is my first silly question. Im running milkyway also, And have about 20 tasks left to run. Will dettaching from SETI hurt milkyway?

---

[/quote]

Old James

Nope, shouldn't hurt Milkyway at all but you might want to set MW for no new tasks first so it doesn't get carried away at being the only project on the block. Leave it at NNT until you get SETI back up and running some WUs before you release MW again.

---

PROUD MEMBER OF Team Starfire World BOINC

I have Milkway at NNT also only have 3 SETI wu to finish crunching. Now I need to find the thread on how to safley do a reattachment. Then find the right files to save.

---

[/quote]

Old James

Get the Ghost Detector
http://setiathome.berkeley.edu/forum_thread.php?id=61519
http://www.mediafire.com/?yyy64lh5953d7
(get the small .zip)

In the .zip there is .pdf which explains how to do detach/re-attach and keep opt apps

---

 


- ALF - "Find out what you don't do well ..... then don't do it!" :)
 

Well I read in another thread that after they went through all the hastle they ended up with just as many ghosts that they started with.

I Have had my NNT set and will let every ghost time out. Will whack my rac but thats the breaks.

---

[/quote]

Old James

Since i was having trouble getting new tasks, i've got the server to resend me all my Ghosts,
now got enough Cuda work to last a day or two, and enough ATI work to last a few more days,
now just picking one or two fresh Wu's at a time,

Claggy

Uhhh, how did you get the server to resend?

---

PROUD MEMBER OF Team Starfire World BOINC

Uhhh, how did you get the server to resend?

Set NNT, do a report (with at least one Wu to report), wait a few seconds, then before Boinc gets the reply, disconnect from Wireless/Disable Wireless Card/Disconnect from Net/maybe even setting Boinc to Network Activity Suspended might work,
use another computer to see if host has reported on host page, wait until Boinc has timed it's connection out, (if you have set Network Activity to suspend, you'll have have wait at least 5 minutes)
then depending if Boinc has managed to report, unset NNT, and ask for work, if you're managed to report that Wu twice, the Server will resend Ghost work instead of fresh work,
if the host page didn't update, wait for Boinc's connection to time out, then try again, waiting a few more seconds this time, the number of seconds you have to wait depends on your connection, how many tasks you're reporting, and how sluggish the Database is at that moment, try 5 seconds first, then work up from that,

Claggy

Great it worked doing the resend trick.

Did 5 tries, first tried suspending network but WU got reported, then 4 times where I disabled the net card, 3 times where I did it to fast and the WU wasn't reported, fourth I got it and got 20 resends :)

Noticed that a new army of Ghosts (612 of them) turned up yesterday. Hopefully this was just an temporary aberration brought about by the extreme server load.

On the 4th of October, prior to the last outage I had:

  1,221 WU 'In Progress'
    449 WU on board
------------------------
    772 Ghost Work Units (63%)

Looks like most of them timed out or were otherwise cleared during the outage because when BOINC came back up yesterday (9th of October) I only had:

    594 WU 'In Progress'
    480 WU on board
------------------------
    114 Ghost Work Units (19%)

However, this morning (10th of October) I now have:

  1,201 WU 'In Progress'
    475 WU on board
------------------------
    726 Ghost Work Units (60%)

---

Visit http://www.mediafire.com/?yyy64lh5953d7 to download GhostDet v1.05

Well, I'am about to destroy my army using Claggy's suggestion, and here are the results so far...
This mornig:

HostID = 4188485

  1,122 WU 'In Progress'
    482 WU on board
------------------------
    640 Ghost Work Units (57%)

Now:

HostID = 4188485

  1,073 WU 'In Progress'
    693 WU on board
------------------------
    380 Ghost Work Units (35%)

And the battle is far from over :-) 20wu/21 times succseeded, 19 more to go... PUFF, puff, pant, puff...

---

Who the hell is General Failure and why is he reading my harddisk?¿

Uhhh, how did you get the server to resend?

Set NNT, do a report (with at least one Wu to report), wait a few seconds, then before Boinc gets the reply, disconnect from Wireless/Disable Wireless Card/Disconnect from Net/maybe even setting Boinc to Network Activity Suspended might work,
use another computer to see if host has reported on host page, wait until Boinc has timed it's connection out, (if you have set Network Activity to suspend, you'll have have wait at least 5 minutes)
then depending if Boinc has managed to report, unset NNT, and ask for work, if you're managed to report that Wu twice, the Server will resend Ghost work instead of fresh work,
if the host page didn't update, wait for Boinc's connection to time out, then try again, waiting a few more seconds this time, the number of seconds you have to wait depends on your connection, how many tasks you're reporting, and how sluggish the Database is at that moment, try 5 seconds first, then work up from that,

Claggy

Might be able to semi-automate this process with a batch file, boincmd & Microsoft's devcon to control the adaptor. Only getting maxxed out server connection issues myself at the moment, so can't test this here right now, but might try something when my machines can get though to double report reliably.

Jason

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

Might be able to semi-automate this process with a batch file, boincmd & Microsoft's devcon to control the adaptor. Only getting maxxed out server connection issues myself at the moment, so can't test this here right now, but might try something when my machines can get though to double report reliably.

Jason

It'll be appreciated! You don't know how much :-)

---

Who the hell is General Failure and why is he reading my harddisk?¿

Might be able to semi-automate this process with a batch file, boincmd & Microsoft's devcon to control the adaptor. Only getting maxxed out server connection issues myself at the moment, so can't test this here right now, but might try something when my machines can get though to double report reliably.

Jason

It'll be appreciated! You don't know how much :-)

No guarantees, but will give it a go when I can ( If someone doesn't beat me to it ;) )

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

[Later:]
Here's what I have so far, a batch file (TrigerResends.cmd) put in my boinc application directory. No idea if it will work or not as is, but at least talks to boinc it seems. Am going to visit friends, and for now at least my main machine just gets 'couldn't connect to server' and 'HTTP error' when trying to reach the server.

Netsh should work on Windows 7 for disabling/enabling any network connection, but maybe not XP. XP may need to use devcon instead (downloaded from microsoft). The calibrated delay is set by the ping -n parameter in between '----'s

@echo off
echo Triggering Resends
echo - Step 1: Setting No New Tasks
boinccmd --project setiathome.berkeley.edu nomorework
ping -n 4 127.0.0.1 > NUL
echo - Step 2: requesting update
boinccmd --project setiathome.berkeley.edu update

REM -------------------------------------------------------------------------------------
echo    - Step 3: calibrated delay
ping -n 5 127.0.0.1 > NUL
REM -------------------------------------------------------------------------------------

echo - Step 4: Disabling network adaptor ...
netsh interface set interface name="Wireless Network Connection" admin=DISABLED
ping -n 4 127.0.0.1 > NUL
echo - Step 5: UN-Setting No New Tasks
boinccmd --project setiathome.berkeley.edu allowmorework
echo - Step 6: enable adaptor...
netsh interface set interface name="Wireless Network Connection" admin=ENABLED
ping -n 10 127.0.0.1 > NUL
echo - Step 7: Trigger update
boinccmd --project setiathome.berkeley.edu update

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.