I upgraded all my machines to 5.10.8 but then started getting a preponderance of Client Errors See Below

If they are marked with exit code -221, then you need not be concerned. This is a new feature implemented with 5.10.7, where a unit is aborted before it begins, in case a quorum has already been met.

This thread has more information on this below......

---

[quote]I upgraded all my machines to 5.10.8 but then started getting a preponderance of Client Errors See Below

If they are marked with exit code -221, then you need not be concerned. This is a new feature implemented with 5.10.7, where a unit is aborted before it begins, in case a quorum has already been met.


I just checked it to see what was going on with the "Aborted by project" error on my WU's. Thank you all for the information. For those of us who still have dial up this is a very bad deal. I see that i will not be crunching Seti in the future.

I had been planning to be away from an internet connection for 10 days. So downloaded 10 days worth of WU's ( it took almost all night ) and had most of them turn into "aborted by project" WU's the next day when i updated one last time.

I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

For those of us who still have dial up this is a very bad deal.
. . .
I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

Suggest you return to 5.8.16 which doesn't have the auto-abort facility. If you look at the Boinc download page, the 5.10.8 version is not now being recommended due to other problems.

---

I upgraded all my machines to 5.10.8 but then started getting a preponderance of Client Errors See Below

If they are marked with exit code -221, then you need not be concerned. This is a new feature implemented with 5.10.7, where a unit is aborted before it begins, in case a quorum has already been met.

I just checked it to see what was going on with the "Aborted by project" error on my WU's. Thank you all for the information. For those of us who still have dial up this is a very bad deal. I see that i will not be crunching Seti in the future.

I had been planning to be away from an internet connection for 10 days. So downloaded 10 days worth of WU's ( it took almost all night ) and had most of them turn into "aborted by project" WU's the next day when i updated one last time.

I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

In other words, if your time crunching is wasted, that's fine, but if your time downloading is wasted, that's a disaster?

This is a good feature. It means that your work is more relevant -- that credit means real credit, and not just crunching for crunching sake.

It does make it harder to carry a big queue, but that's been true for a long time anyway. 5.10.8 does let you request extra days and keep a short connect interval.

---

For those of us who still have dial up this is a very bad deal.
. . .
I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

Suggest you return to 5.8.16 which doesn't have the auto-abort facility. If you look at the Boinc download page, the 5.10.8 version is not now being recommended due to other problems.

Or just have a smaller cache.

---

Grant
Darwin NT

Evelyn et al

With MultiBeam coming there should be enough work that aborted by project is not an issue. Currently there is active testing to see if the workunit is sent to only two people and they return the result can work. Seti is trying to place proper value on the work everyone has completed.

I recall not to far in the past people were complaining about the workunits are sent to everyone and time was being wasted.

So somewhere in all that Seti acknowldeged (even though short of WU's) that this makes sense and working to get there.

So yes there are growing pains getting ready for MultiBeam.

Pappa

[quote]I upgraded all my machines to 5.10.8 but then started getting a preponderance of Client Errors See Below

If they are marked with exit code -221, then you need not be concerned. This is a new feature implemented with 5.10.7, where a unit is aborted before it begins, in case a quorum has already been met.


I just checked it to see what was going on with the "Aborted by project" error on my WU's. Thank you all for the information. For those of us who still have dial up this is a very bad deal. I see that i will not be crunching Seti in the future.

I had been planning to be away from an internet connection for 10 days. So downloaded 10 days worth of WU's ( it took almost all night ) and had most of them turn into "aborted by project" WU's the next day when i updated one last time.

I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

---

Please consider a Donation to the Seti Project.

I had been planning to be away from an internet connection for 10 days. So downloaded 10 days worth of WU's ( it took almost all night ) and had most of them turn into "aborted by project" WU's the next day when i updated one last time.

I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.
[/quote]
In other words, if your time crunching is wasted, that's fine, but if your time downloading is wasted, that's a disaster?

This is a good feature. It means that your work is more relevant -- that credit means real credit, and not just crunching for crunching sake.

It does make it harder to carry a big queue, but that's been true for a long time anyway. 5.10.8 does let you request extra days and keep a short connect interval.[/quote]

It means that for 6 out of 10 days my computers would have sit there running with nothing to process. I opted not to waste the power. That is the thing that is most relevant.

I had been planning to be away from an internet connection for 10 days. So downloaded 10 days worth of WU's ( it took almost all night ) and had most of them turn into "aborted by project" WU's the next day when i updated one last time.

I have slow - slow dial up and do not plan to have my phone line tied up for hours downloading WU's that i will never receive credit for.

In other words, if your time crunching is wasted, that's fine, but if your time downloading is wasted, that's a disaster?

This is a good feature. It means that your work is more relevant -- that credit means real credit, and not just crunching for crunching sake.

It does make it harder to carry a big queue, but that's been true for a long time anyway. 5.10.8 does let you request extra days and keep a short connect interval.

It means that for 6 out of 10 days my computers would have sit there running with nothing to process. I opted not to waste the power. That is the thing that is most relevant.

You reconnected in less than 10 days: if you download 10 days of work, and crunched for 10 days without reconnecting, you would have received credit for ten days work.

The "connect every 'x' days" setting is more than just an instruction to BOINC, it is also a statement that the scheduler needs to expect to hear from you once in the next ten days.

... and I guess I don't understand: if you are taking the computer(s) with you, there isn't a problem, and if you leave them behind, it's probably best to save energy and not worry about hardware (fans, power supplies, etc.) failing while no one is around.

---

I do not mind when they are aborted by project now they are saying client error aborted with the same errors. It makes my little contribution look wrong with all these client error aborted

---

I do not mind when they are aborted by project now they are saying client error aborted with the same errors. It makes my little contribution look wrong with all these client error aborted

Don't sweat it. Everyone else is in the same boat.

Where is this feature documented? Another project wants to implement this too, but can't find anything on it.

---

Dublin, California
Team: SETI.USA

Where is this feature documented? Another project wants to implement this too, but can't find anything on it.

Should be on trac...........Changeset 12322

---

I've changed the text displayed for these results on the web site to make it clear that it shouldn't be considered an error.

Let me know what you think.

Eric

---

@SETIEric@qoto.org (Mastodon)

Nice :D, Thanks Eric! ['A rose by any name...'... but it still smells sweeter, Go Figure :D]

---

"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.

Instead of

Client error Aborted

it now shows up as ....

Redundant result Cancelled by server

which is certainly self-explanatory.

Thanks, Eric.
(BTW...the change hasn't propagated to the beta results...yet)

---

An excellent change, and I agree with bounty.hunter that the terms seem self-explanatory to those of us who have been following the 'Aborted by project' saga.

However, one user has already started a thread asking for a fuller explanation: perhaps a line needs to be added under the explain link at the head of each column.

I've changed the text displayed for these results on the web site to make it clear that it shouldn't be considered an error.

Let me know what you think.

Eric

Sounds good to me for 221's. I assume 197's will still use the old terminology?

Call me a CW, I don't care to test it! :-D

Alinator

<edit> @ Richard: Yes, that's the logical next step! ;-)

An excellent change, and I agree with bounty.hunter that the terms seem self-explanatory to those of us who have been following the 'Aborted by project' saga.

However, one user has already started a thread asking for a fuller explanation: perhaps a line needs to be added under the explain link at the head of each column.

explain link ?
Where can I find that one?

---

mic.

An excellent change, and I agree with bounty.hunter that the terms seem self-explanatory to those of us who have been following the 'Aborted by project' saga.

However, one user has already started a thread asking for a fuller explanation: perhaps a line needs to be added under the explain link at the head of each column.

explain link ?
Where can I find that one?

At the top of the column where the error messages is in the results display.

---

Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.

The links at the top of the columns on the summary pages go to explanations.

Alinator