I've been keeping an eye on the result page in the server status area, and it seams that the results ready to send are climbing up dramaticly again.
It's up to 500k already.
Please tell me to just go to bed and get some sleep and that it's probably nothing...
Is berkeley ready for the big merger?

---

Those are results are ready to send to us...Users
when they climb it is a good thing
When they get to 0 , means we have no work to do

---

But the "Workunits waiting for assimilation" grows up. May be this can get a problem

greets

---

mcbeth

YES, although I don't know when or how. LOL Could be 10 seconds, 10 days, or 10 months, but there will be an outage

---

But the "Workunits waiting for assimilation" grows up. May be this can get a problem

greets

The "Waiting for assimilation" queue has been growing because they have had the assimilators turned off since they started the database merge. I see that the assimilators are now back on again, so I can only assume that the merge is complete and they are now assimilating all the pending W/U's into the master science database. As far as the "Ready to send" queue climbing, this is a good thing for now. They just recently put 2 new splitters online to create new work for the influx of new users from the closing of Seti Classic. Once the demand for new work levels off, they will disable one or more of the splitters to keep the "Ready to send" queue at the proper level based upon the demand for new work. I hope this helps to ease your mind a bit. Happy crunching.

Regards, Daniel.

---

oh well if there is one i think everyone has other projects o any ways thats why boinc was made so you will allways have work.

---

The "Waiting for assimilation" queue has been growing because they have had the assimilators turned off since they started the database merge. I see that the assimilators are now back on again, so I can only assume that the merge is complete and they are now assimilating all the pending W/U's into the master science database. {snip}.

Regards, Daniel.

I suspect that it's more a case that they have to assimilate the classic DB in "bite sized chucks". Merge a bit, assimilate it, merge some more, and so on...

Last I heard there were something like 50million results in the classic DB waiting for assimilation.
If they tried to do them all in one go it would, i assume, tie the assimilators in knots, as we have seen dramatic slowdowns when some queues grow into low 7 figures, let alone middle 8's.
IMHO.

{EDit spelling}

---

Foamy is "Lord and Master".
(Oh, + some Classic WUs too.)

The "Waiting for assimilation" queue has been growing because they have had the assimilators turned off since they started the database merge. I see that the assimilators are now back on again, so I can only assume that the merge is complete and they are now assimilating all the pending W/U's into the master science database. {snip}.

Regards, Daniel.

I suspect that it's more a case that they have to assimilate the classic DB in "bite sized chucks". Merge a bit, assimilate it, merge some more, and so on...

Last I heard there were something like 50million results in the classic DB waiting for assimilation.
If they tried to do them all in one go it would, i assume, tie the assimilators in knots, as we have seen dramatic slowdowns when some queues grow into low 7 figures, let alone middle 8's.
IMHO.

{EDit spelling}

I thought that it was 50 million WUs waiting for validation.

The assimilators are off because they would interfere with the merge of the classic science DB with the BOINC S@H DB. The assimilators take tha cannonical result and store it in the science DB. The cannonical result and WU are then available for deletion.

---

BOINC WIKI

The "Waiting for assimilation" queue has been growing because they have had the assimilators turned off since they started the database merge. I see that the assimilators are now back on again, so I can only assume that the merge is complete and they are now assimilating all the pending W/U's into the master science database. {snip}.

Regards, Daniel.

I suspect that it's more a case that they have to assimilate the classic DB in "bite sized chucks". Merge a bit, assimilate it, merge some more, and so on...

Last I heard there were something like 50million results in the classic DB waiting for assimilation.
If they tried to do them all in one go it would, i assume, tie the assimilators in knots, as we have seen dramatic slowdowns when some queues grow into low 7 figures, let alone middle 8's.
IMHO.

{EDit spelling}

I thought that it was 50 million WUs waiting for validation.

The assimilators are off because they would interfere with the merge of the classic science DB with the BOINC S@H DB. The assimilators take tha cannonical result and store it in the science DB. The cannonical result and WU are then available for deletion.

I noticed that the "Waiting for Assimilation" number has been growing steadily since yesterday...even when all four assimilators were running.

---

You will be assimilated...bunghole!

I noticed that the "Waiting for Assimilation" number has been growing steadily since yesterday...even when all four assimilators were running.

It *could* be that the science database merge that is reportedly running in the background is slowing the assimilator's access to that database, thereby causing the backlog... This would be my first guess. I guess the database merge currently rates a bit higher on the priority pole than keeping the S@H/BOINC assimilator queue drained at the moment. If you look at the nice graphics on the teamseti website, the assimilator queue started dropping like a rock, before stopping at around 600k and climbing slowly. If the assimilator queue gets TOO high, I would guess that they can always 'pause' the database merge and let the assimilators have unfettered access for a few hours to drain it. Rinse... Repeat... until the merge is complete.

So, no... I don't think it means that another painful outage is 'on the way'...

---

https://youtu.be/iY57ErBkFFE

#Texit

Don't blame me, I voted for Johnson(L) in 2016.

Truth is dangerous... especially when it challenges those in power.

I noticed that the "Waiting for Assimilation" number has been growing steadily since yesterday...even when all four assimilators were running.

It *could* be that the science database merge that is reportedly running in the background is slowing the assimilator's access to that database, thereby causing the backlog... This would be my first guess. I guess the database merge currently rates a bit higher on the priority pole than keeping the S@H/BOINC assimilator queue drained at the moment. If you look at the nice graphics on the teamseti website, the assimilator queue started dropping like a rock, before stopping at around 600k and climbing slowly. If the assimilator queue gets TOO high, I would guess that they can always 'pause' the database merge and let the assimilators have unfettered access for a few hours to drain it. Rinse... Repeat... until the merge is complete.

So, no... I don't think it means that another painful outage is 'on the way'...

Yeah that's my thought too. They made no mention of the database merge being completed...so I figure the assimilation queue will rise quite a bit before everything's done. Fortunately, the other more "critical" services are chugging along smoothly.

---

You will be assimilated...bunghole!

I noticed that the "Waiting for Assimilation" number has been growing steadily since yesterday...even when all four assimilators were running.

It *could* be that the science database merge that is reportedly running in the background is slowing the assimilator's access to that database, thereby causing the backlog... This would be my first guess. I guess the database merge currently rates a bit higher on the priority pole than keeping the S@H/BOINC assimilator queue drained at the moment. If you look at the nice graphics on the teamseti website, the assimilator queue started dropping like a rock, before stopping at around 600k and climbing slowly. If the assimilator queue gets TOO high, I would guess that they can always 'pause' the database merge and let the assimilators have unfettered access for a few hours to drain it. Rinse... Repeat... until the merge is complete.

So, no... I don't think it means that another painful outage is 'on the way'...

Though you might note that the splitters are off right now...

MJ

Yeah that's my thought too. They made no mention of the database merge being completed...so I figure the assimilation queue will rise quite a bit before everything's done. Fortunately, the other more "critical" services are chugging along smoothly.

---

Is there a outage on the way again?

Next Wednesday for sure.

---

Grant
Darwin NT

Is there a outage on the way again?

Next Wednesday for sure.

LOL!

There is a great tech news message in the Tech News area. The servers are barely keeping up. I just took my fastest machine off line for Seti. I will probably leave it that way until after the DB merge is complete. No need to kill Seti, especially when there is more then enough work getting completed right now. Of course the machine is happily crunching away on other projects.

I would suggest other people do the same. I am running BOINC on 6 machines. So taking the fastest one off line means that there is still a lot of work happening an hopefully the servers won't see any problems.

Thinks look bright, the merge will done soon, the new units will be available soon, and hopefully we will see new optimized code out soon too. I hope everyone is having a happy holiday.

---

UL/DL is disabled ATM.

Regards Hans

Edit: And up again :o)

---